The Archaeology of Colonialism in Medieval Ireland: Shifting Patterns of Domination and Acculturation Russell Ó Ríagáin St Catharine’s College This Dissertation is Submitted for the Degree of Master of Philosophy Department of Archaeology University of Cambridge 31 August 2010 Preface This dissertation is the product of my own work and includes nothing which is the outcome of work done in collaboration, except where specified in the text. Length 25,087 words. Reflexion I should acknowledge the fact that I am Irish, and was born and grew up in one of my case study regions. True objectivity is beyond any scholar, but it should be borne in mind that I received a primary school education heavily influenced by Irish nationalism, being constantly taught to resent the Normans, Vikings and English for what they had done to Ireland, and to glorify national heroes such as Brian Ború. This project in many ways is a reaction to this. Also, the fact that I am from a working class housing estate in an Irish provincial town, and that I grew up in 1980s (i.e. pre- Celtic Tiger) Ireland may have also affected thinking both on the period under examination in this dissertation, and on Ireland in general, having the same love/hate feelings towards my home country as so many others born in my time. The same may be true of my political past, having been a member of both the Socialist Worker’s Party and Sinn Féin at separate points of my youth. 2 ABSTRACT This project examines Scandinavian and Anglo-Norman colonialism in two Irish case- study regions, the south-east and the mid-west, by placing them on a continuum of social development. It analyses their spatial organisation and their impact on the landscape in terms of a model of colonialism based on three sub-phases: expansion, consolidation and domination. Campaign fortresses and other bridgeheads in the landscape such as the longphort and the ringwork largely belong to this phase. Mottes and Scandinavian urban settlement belong largely to the consolidation phase. The Anglo-Norman domination phase was characterised by a hierarchical configuration of monuments, including those forms already mentioned, along with masonry castles, nucleated and dispersed rural settlement and continental religious houses. They differed in a number of respects. Scandinavian colonialism was much more geographically limited, largely confined to a series of estuarine settlements which became towns over time, with possible accompanying hinterland settlement. It cannot be said to have had a domination sub-phase, rather it experienced a phase of incorporation, where the settlements came under the control of elements of the Gaelic elite. It has therefore been categorised here as non-imperial opportunistic colonialism. In contrast to the elite replacement colonialism found in Anglo-Norman Ulster and Norman England, Anglo-Norman colonialism in the case-studies was totalising, characterised by plantation colonialism, which involved the inward movement of several orders of society, and the incorporation or displacement of native groups. It involved the total reorganisation of the landscape, with the introduction of several new monument forms in a hierarchical spatial organisation. However, this was largely unsuccessful in the mid-west, and can only be said to have been successful in the south-east, and even then only until the fourteenth century, at which time the colony receded substantially. Both groups continued to be regarded as foreign elements long after the apogee of each of colonial period. While colonial acculturation was limited, over time their culture came to differ both that of their home regions and Gaelic society, which saw them become a “third nation” living in a “third space” (cf. Bhabha 1994). This was due to a combination of creolisation and hybridisation. There seems to have been extensive Gaelic acculturation in the areas of greatest contact, such as the towns in each period. 3 Contents Preface, Statement of Length, Reflexive Note 2 Abstract 3 Acknowledgements 5 Dedication 6 List of Tables 7 List of Illustrations 7 Chapter 1. Introduction and Research Context 11 1.1 Aims 11 1.2 Epistemological Setting 11 1.3 Previous Research 12 1.4 Historical Setting 22 1.5 Thesis Outline 25 Chapter 2. Methods 26 2.1 Identifying Colonial Monuments in Medieval Ireland 26 2.2 Case-Study Regions 37 2.3 Methodology 40 Chapter 3. Scandinavian Colonialism in Ireland 43 3.1 Introduction 44 3.2 Scandinavian Colonial Expansion in the Mid-West 45 3.3 Scandinavian Colonial Expansion in the South-East 50 3.4 Souterrains: A Gaelic Response to Scandinavian Colonialism? 54 3.5 Scandinavian Colonial Consolidation in the Mid-West 55 3.6 Scandinavian Colonial Consolidation in the South-East 58 4 3.7 Incorporation in the Mid-West 61 3.8 Incorporation in the South-East 63 3.9 From Scandinavian to Hiberno-Scandinavian: Creolisation, 65 Acculturation and Hybridisation Chapter 4. Anglo-Norman Colonialism in Ireland 69 4.1 Prelude: Invasion 69 4.2 Expansion 69 4.3 Anglo-Norman Colonial Expansion in the South-East 71 4.4 Anglo-Norman Colonial Expansion in the Mid-West 74 4.5 Consolidation 77 4.6 Consolidation in the South-East 78 4.7 Consolidation in the Mid-West 82 4.8 Domination 84 4.9 Colonial Domination in the South-East 88 4.10 Colonial Domination in the Mid-West 106 4.11 Decline Chapter 5. Scandinavian and Anglo-Norman Colonialism Considered 118 5.1 Expansion 119 5.2 Consolidation 121 5.3 Incorporation 122 5.4 Domination 123 5.6 Colonialism and Imperialism in Medieval Ireland and Beyond 125 Chapter 6. Epilogue: Towards a Theory of Colonialism 128 Bibliography 132 5 Acknowledgements This project would have been even more difficult to see through to completion if it were not for the help and advice of a number of people. Not least among these is my supervisor, Dr James Barrett, whose continual and tireless help and advice, even in the face of my near intractable stubbornness, throughout the project has been greatly appreciated. Aisling Murphy’s indefatigable efforts editing and proofing have also been greatly appreciated. I would also like to acknowledge the help and advice of the following: Michael Moore and Paul Walsh, Archaeological Survey of Ireland, Ed Bourke and Rachel Barrett of the National Monuments Archive Unit, Dublin, Con Manning, National Monuments Conservation Unit, Dr Kieran O’Conor, Dr Carleton Jones and Dr Elizabeth FitzPatrick, Department of Archaeology, National University of Ireland, Galway, Dr Niall Brady and Anthony Corns, The Discovery Programme, Dr Elizabeth Boyle and Dr Denis Casey, Department of Anglo-Saxon, Norse and Celtic, University of Cambridge, Prof. Stephen Mennell and Dr Seán L’Estrange, School of Sociology University College Dublin, Niamh Arthur, School of Archaeology, University College Dublin, Dr Simon Stoddart, Andrew Woods and not least David Redhouse, Department of Archaeology, University of Cambridge. I would also like to thank my friends and colleagues at the Medieval Folkmoot, and also St Catharine’s College Cambridge for enabling me to travel to a number of important conferences throughout the year. Any mistakes and/or scholarly flights of fancy are entirely my own. 6 For my mother, Catherine O’Regan, my earliest and greatest teacher, and whose lifelong courage in the face of adversity has been an inspiration to all who have known her 7 List of Tables Chapter 2 Table 1.1, the sub-processes of colonialism and associated settlement forms. Table 2.2, characteristic monuments associated with each group. Chapter 4 Table 4.1, hierarchy of Anglo-Norman settlement monuments in the domination phase, and the processes associated with them. Table 4.2, hierarchy of Anglo-Norman settlement forms in the domination phase. Table 4.3, classified Religious Houses in Co. Clare. Table 4.4, religious Houses in south-east case study. Chapter 5 Table 5.1, centralised and non centralised colonialism ideal types. Table 5.2, sub-processes of the colonial process. Table 5.3, Cultural Processes Associated with Colonialism. List of Illustrations Chapter 1 Figure 1.1, map showing elements of the political geography of pre-Anglo-Norman Ireland. Chapter 2 Figure 2.1, internal area of Lissanard Cashel, Ennis, Ireland, (author’s photograph). Figure 2.2, view of the River Suir from the western end of the longphort at Woodstown, Co. Waterford, (author’s photograph). Figure 2.3, motte at Hen Domen, Montgomery, Wales (author’s photograph). Figure 2.4, Anglo-Norman masonry castle at Trim, Co. Meath, Ireland (author’s photograph). 8 Figure 2.5, Dominican Priory at Athenry, Co. Galway, Ireland (author’s photograph). Figure 2.6, location of case study regions and modern administrative boundaries. Chapter 3 Figure 3.1, rivers and lakes mentioned in the text. Figure 3.2, Scandinavian/Hiberno-Scandinavian and Gaelic secular settlement in the Mid-West. Figure 3.3, known Scandinavian and Gaelic secular settlement in the mid-west, with undated enclosures added. Figure 3.4, Scandinavian settlement with maximum possible contemporary ecclesiastic settlement. Figure 3.5, known Scandinavian and Gaelic secular settlement in the south-east. Figure 3.6, known Scandinavian and Gaelic secular settlement in the south-east, with undated enclosures added. Figure 3.7, known Scandinavian and early ecclesiastic settlement in the south-east. Figure 3.8, Limerick with regional round towers. Figure 3.9, Scandinavian settlement and ecclesiastic round towers in the south-east . Chapter 4 Figure 4.1, early Anglo-Norman settlement activity and Gaelic secular settlement in the south-east. Figure 4.2, Anglo-Norman ringworks and campaign fortification with pre-twelfth century ecclesiastic sites and known Hiberno-Scandinavian settlement. Figure 4.3, Anglo-Norman ringworks and the pre-existing secular settlement pattern. Figure 4.4, ringworks and the known possibly contemporary Hiberno-Scandinavian and early ecclesiastic settlement. Figure 4.5, mottes, ringworks and pre-existing secular settlement in the south-east. Figure 4.6, mottes, ringworks and the pre-existing known ecclesiastical settlement pattern in the south-east. Figure 4.7, ringworks, mottes and pre-existing Gaelic secular settlement in the mid- west. Figure 4.8, ringworks, mottes and pre-Anglo-Norman ecclesiastical settlement in the mid-west. Figure 4.9, Anglo-Norman ringworks, mottes and masonry castles in the south-east. 9 Figure 4.10, Anglo-Norman castles and possible urban centres in the south-east case study region. Figure 4.11, spatial relationship between known early ecclesiastic sites and Anglo- Norman towns and boroughs. Figure 4.12, known high medieval parish churches and Anglo-Norman castles in the south east. Figure 4.13, rural nucleated settlement and known Anglo-Norman encastellation in the south-east. Figure 4.14, deserted medieval settlements and total ecclesiastic landscape. Figure 4.15, religious houses in the south-east. Figure 4.16, religious houses and early ecclesiastic settlement. Figure 4.17, religious house and the known Gaelic settlement pattern. Figure 4.18, religious houses and nucleated settlement in the south-east. Figure 4.19, religious houses and Anglo-Norman castles in the south-east. Figure 4.20, moated sites, deserted medieval settlements and religious houses. Figure 4.21, Anglo-Norman moated sites, towns and castles in the south-east. Figure 4.22, moated sites with known Gaelic secular settlement in the south-east. Figure 4.23, moated sites and minimum early ecclesiastic sites in the south-east. Figure 4.24, moated sites with early ecclesiastic sites in the south-east. Figure 4.25, Anglo-Norman castles, urban and rural nucleated settlement in the mid- west. Figure 4.26, religious Houses in Co. Clare. Figure 4.27, moated sites and Anglo-Norman castles in the mid-west. Figure 4.28, moated sites and Gaelic secular settlement in the mid-west. Figure 4.29, moated sites and early ecclesiastic sites in the mid-west. 10 Chapter 1. Introduction and Research Context 1.1 Aims This project seeks to identify the processes at work in Scandinavian and Anglo- Norman colonialism in Ireland, and their interaction with the landscape, by examining the impact of each phase of activity on the settlement pattern in two representative case-study regions. The successes, failures, similarities and differences of Scandinavian and Anglo-Norman settlement and society in Ireland are examined and compared in this project in terms of three sub-phases of the overall process, namely expansion, consolidation and domination, within an overall developmental diachronic framework. 1.2 Epistemological Setting This project is both comparative and diachronic. Elias (2000, 403) notes that ‘our habits of thinking incline us to look for “beginnings”’. He is quite critical of the tendency for historiography to identify zero-points in the trajectory of human development (2000, 403). With this in mind, rather than taking two chronologically and geographically separate colonial episodes, treating them synchronically and then comparing them, this project takes two colonial episodes occurring within the same space over time. The two colonial episodes are not treated as historical zero-points but rather as periods of accelerated change in a dynamic social continuum. This approach makes it possible to analyse the factors influencing social change and stasis in Ireland from c.700 to the late fourteenth century outside of the traditional historiographical paradigm concerned with unilinear trajectories, ultimate causes and 11 the agencies of “great” individuals. Colonialism in each phase took different forms. By taking a developmental long durée approach, analysing them both together over time, it is hoped that similar processes may be seen at work, even if the forms taken by colonialism are different in each period. The continuity of settlement evidence as a presence in the archaeological record provides a sound base from which to move outwards into further explanatory avenues. This study examines the effects of colonial activity on settlement form and patterning in the medieval Irish landscape both spatially and temporally. It assesses whether the appearance, disappearance and changing inter-relationships of certain forms and patterns were due to colonial stimuli, internal native political developments or environmental constraints. By doing so, changes in the access to or control over ideological, military, economic and political resources can be mapped out over time, illustrating shifting configurations in the exercise of social power (cf. Mann 1986; Elias 2000; Tilly 1992). This project also examines the extent to which the colonial activity can be seen as part of a planned process, and the extent to which it was organic, consisting of a series of contingent events and developments. The analysis of changes in settlement form and patterning in the landscape over time can be used to provide insights into all of these issues. 1.3 Previous Research The archaeologist, like the author of any text, is a product of his/her times and every text, to one extent or another, is infused with the author’s ideology (cf. Jenkins, 1991, 1-32; Ó Ríagáin 2008). Therefore, the choice of subject matter, its presentation and interpretation all provide points of entry into the modes of thought and the prevalent 12 discourses of an author’s age. Trigger (1984) points to the existence of a number of traditions within archaeology, namely nationalist, colonial and imperialist archaeologies. These are ideal types however, and many archaeological traditions contain elements of more than one (1984, 358, 368). He notes that ‘most archaeological traditions are probably nationalistic in orientation’ (Trigger 1984, 358). He notes that nationalist archaeology ‘is probably strongest amongst peoples who feel politically threatened, insecure or deprived of their collective rights’ (Trigger 1984, 358). He also draws attention to the connections of this form of archaeology to history (Trigger 1984, 358), and that it ‘tends to draw attention to the more recent past…to the political and cultural achievements of ancient civilisations or other forms of complex societies’ (Trigger 1984, 360). While Trigger points to Czech and Danish archaeology as being examples of the former especially (1984, 358), Ireland can be taken as another example, although there have been elements of the other two forms also at work there. Colonialist archaeology sought to justify the poor treatment of subjugated/supplanted peoples by emphasising their primitiveness and lack of accomplishments (Trigger 1984, 360). It emphasised and glorified the colonial past (Trigger 1984, 360). To an extent this also can be said to have been true for Ireland, both in archaeology and historiography. Certainly Orpen must be regarded in this light, and while his four volume Ireland Under the Normans (1911-20) remains of great importance, it remains ‘the story of the extirpation of what he calls Irish tribalism and the imposition of the “Pax Normannica”’ (Duffy 1997, 4-5). As recently as 1968, Othway-Ruthven provides a narrative of post-invasion [i.e. post- 1169] Ireland ‘unashamedly anglocentric in tone’ (Duffy 1997, 5). 13  These works are part of a long tradition of colonialist historiography stretching back to the initial years of the Anglo-Norman settlement, to literature which, like ‘all conquest literature seeks to explain to the conquerors “why we are here”’ (Bartlett 1993, 98). Giraldus Cambrensis provides a narrative of the invasion, Expugnatio Hibernica (Scott and Martin 1978) and an equally biased description of late twelfth century Irish society Topographia Hibernia (O’Meara 1982). The former must be regarded as a partial account of how his relatives had ‘stormed Ireland’ (Bartlett 1993, 98; Davies 1990, 32; cf. O’Conor 2003, 29-31), often downplaying or denigrating other leading participants. The latter is at once a description of the Irish land and society, and a justification for their subjugation. The other main textual source for the period, The Song of Dermot and the Earl,1 an anonymous rhymed chronicle in Old French probably dating to the early thirteenth century, provides a highly personalised narrative of events, while also listing in detail the initial distribution of land to the first generation of settlers (Bartlett 1993, 98). This personalised narrative has dominated much of the historiographical literature on the invasion. Historical archaeology, theoretically informed by both post-colonial and long term social dynamic approaches such as those of Elias (2000), Mann (1986), Tilly (1992) and Anderson (1974), can provide a useful alternative narrative by addressing the processes at work behind these events and individuals in these sources. Ireland is a post-colonial national state, and among the oldest at that, therefore it provides a useful study in the nationalism of a nascent state. Nationalism seems to be a necessary phase of the post-colonial condition (cf. Young 2003, 59-66; Anderson 2006, 113ff). Young notes that ‘nationalism is Janus-faced: before independence good; after independence bad’ (2003, 62). Therefore, it is hardly surprising that post- 1 Also known as The Deeds of the Normans in Ireland or La geste des Engleis en yrlande (Mullaly 2002). I will refer to it throughout as Song, followed by the line numbers 14 independence Irish nationalist archaeology did its best to ignore the archaeology of the period after the arrival of the Anglo-Normans, seen as being hugely detrimental to Irish society, and the beginning of the “800 years” so lamented by Irish Republicans. Although O’Conor (1998, 11) has already drawn attention to the following words from R.A.S. Macalister (holder of the chair of Celtic Archaeology at University College Dublin in the early years of the twentieth century) regarding the study of Anglo-Norman archaeology they are worth quoting here: In speaking of the antiquities of the period, it will be unnecessary to make more than passing allusions to those remains which are English in all but geographical situation. Such subjects are cross-legged effigies, pavement tiles, Plantagenet coins, arms and armour are a branch of English archaeology and even their extension to Ireland is much more a matter of English than Irish interest (Macalister 1928, 356). The Viking period has occupied more of a grey area, with the myth of 1014, of the supposed expulsion of the Vikings by Brian Boruma at the Battle of Clontarf, being seen as another high point. Largely though, the emphasis was on perceived high points prior to this, in halcyon days of great cultural and spiritual achievement for the Irish people (O’Conor 1998, 10-11; Barry 1987, 1). This emphasis on the so-called ‘Island of Saints and Scholars’ from the nineteenth century onwards has been a major part in the formation of Irish national identity and solidarity,2 as any individual having passed through the Irish primary education system will be aware. It looms large in the Irish conscience collective, providing the images used on several generations of coinage, paper money, stamps and other national symbols (cf. O’Conor 1998, 11). The following quote from Macalister (attention to which has already been drawn by O’Conor 1998, 11) is worth noting in this respect: 2 However, in an interesting twist, the arrival of a British missionary, Patrick, in Ireland, was also appropriated by the Anglo-Irish ascendency (cf. Binchy 1962a). 15  In these tempestuous days of ours, the young Free State of Ireland trims her argosy, and sets forth in courage and aspiration to voyage over the uncharted seas of the future. Four thousand years ago her people guided the first faltering steps of the Folk of the North on the way to civilisation. Twelve hundred years ago they shepherded a war-broken Europe upon the way of learning and the way of life. May she prove worthy of her ancient past; may she find that once more she has a mission to a bewildered, rudderless world: and may God be her speed in its fulfilment (Macalister 1928, 357). Texts such as Cahill (1995) continue to reproduce such a view, albeit largely for an Irish-American readership seeking to reinforce their largely imagined perception of Ireland and its past. Anderson (2006) draws attention to this trend in general for expatriate communities and their roles in conservative cultural nationalism. From the 1970s onwards, the balance has increasingly begun to tip back, with academic archaeology broadening its focus from the prehistoric and early Christian periods, perhaps due to Ireland’s increasing international involvement (Barry 1987, 1). The past generation has seen the Peace Process in the Six Counties, which has led to the depolarisation and de-polemicisation of the island’s past, helping to mitigate ideological distortion and remove any social stigma from studying Ireland’s colonial past. This provides further evidence of Trigger’s (1984) assertion that archaeologists and their research agendas are influenced by their socio-political environment. This shift in focus was also facilitated by the increased amount of data available from excavation and survey on the colonial period. Following the lead of Queen’s University Belfast, lectureships in medieval [i.e. high and late medieval] archaeology were set up across the Irish academy (Barry 1987, 1-2). Perhaps also the Wood Quay 16 debacle provided a means to strengthen the study of the colonial period in Ireland. At Wood Quay one of the most important excavations in medieval European archaeology was conducted in a ludicrously short period ahead of the construction of the new civic offices in Dublin in the late 1970s to public uproar. The incident illustrates that while the line of thought as evinced by Macalister remained strong in the corridors of power until comparatively recently, a new generation of the Irish public and academy were now engaging with their medieval colonial past. Barry (1987) provides the first synthesis of all the archaeological work ever carried out on the medieval period in Ireland up until that year (O’Conor 1998, 9). A number of studies have appeared subsequently dealing with the archaeology of the Anglo-Norman colonial period, such as O’Conor (1998) and O’Keeffe (2000). The former provides a study of rural settlement in Ireland in both native and colonial areas, the latter examining the evidence for urban and rural life, economic activity, and the Church in this period. For the pre-Norman period, Edwards (1990, 2005) provides culture-historical surveys of the archaeology of medieval Ireland from c.400 to the arrival of the Anglo- Normans. Mytum (1992) conducts an important large-scale study of the pre-Viking medieval Irish society from a processualist perspective, taking a systems based approach to examine the factors influencing change and stability in the social, economic, political and religious sub-systems. Valente (2008) provides a thorough survey of the Vikings in Ireland, utilising both archaeological and historical evidence. Downham (2007) discusses the Viking kings of Britain and Ireland, largely from a historical perspective. Ó Corráin (1972) and Byrne (2005a, 2005c, 2005d) provide important outlines of Ireland in the Viking Age. Mytum (2003) provides an important discussion of the archaeology of the 17 Vikings in Ireland, warning against over-estimating the level of integration between the two societies. Graham-Campbell (1976), Kenny (1987, 2005), Dolley (1987) and Sheehan (1998, 2000, 2004) provide a discussion of the numismatic and non- numismatic evidence, largely from hoarding. It is difficult to ignore the historical narrative when investigating this period. While this project is archaeological in nature, taking the distribution of monuments in the landscape as the starting point for its analysis, use will also be made of historical and historiographical texts. A number of sites are known only from documentary evidence, usually either from the annals and early legal texts (cf. Hughes 1972; Kelly 1998) or from later documents contained in collections such as the Calendars of State Papers Relating to Ireland (cf. Asplin 1987). However, it should be noted that there is often a far higher number of particular monuments in the landscape than are documented in the textual evidence. This should be borne in mind when dealing with the Irish material. A number of relevant historical and cross-disciplinary general studies have been conducted. In addition to the studies already mentioned, Dolley (1972), Martin (1987a, 1987b, 1987c, 1987d, 1994), Lydon (1987a, 1987b), Glasscock (1987), Simms (1989) and Roche (1995) provide historical accounts of the Anglo-Norman period in Ireland. Lydon (1972, 1987b, 1987c, 1987d), Watt (1987a, 1987c), and Down (1987) focus on the history of the later medieval period. Nicholls (1987, 2003), and Watt (1987b) examine the evidence for the development of Gaelic society in the high and late medieval periods. Ó Corráin (1989) and Ó Cróinín (1995) provide historical surveys of the pre-Anglo-Norman period. MacNiocaill (1972) discusses Ireland in the years c.400-800AD, with Ó Corráin (2005) providing an outline of Irish 18 society c.800. Hughes (1966, 2005a, 2005b), Bethel (1981), Ó Corráin (1981), Sharpe (1984), Etchingham (1999, 2010), and Byrne (2005b), assess the historical evidence for the Irish Church prior to the twelfth century reforms. Watt (1972) provides an outline of the historical development of the medieval Irish church beginning with these reforms. As can be seen from the above evidence, the majority of scholarship on medieval Ireland rarely treats the period in its entirety. A great number of studies on the period place the twelfth century as their starting or ending point, despite the many continuities. Duffy (1997, 2ff) warns against the use of 1169, the year of the commencement of Anglo-Norman colonialism, as a starting or finishing point for analysis, placing instead the Anglo-Norman arrival at the centre of his analysis. This project goes further still, treating both the arrival of Viking and Anglo-Norman colonists as parts of a longer term social dynamic as opposed as zero-points or ruptures. Post-Colonial Theory Post-colonial theory has shown the value of analysing the particular rather than the general, and of recognising that multiple narratives exist told by multiple voices. It has provided an important corrective to the World Systems approach which, although being an important macro-level approach can tend to over-generalise and focus too heavily on core-periphery relations (Gosden 2004; Naum 2010). It has helped us to attune our faculties to messages from the subaltern zone, to narratives of the dominated and the dispossessed (Young 2003). Might we also find such messages in the past by analysing the changes in settlement form and patterning due to colonial activity in medieval Ireland, or elsewhere in the archaeological record? 19  Said (1978; 1993) provides important alternative perspectives on the cultural aspects of colonialism, such as the process of ‘othering’ and the effects on both coloniser and colonised. Bhabha (2004) introduces the concept ‘Third Space’, a space of hybridity, effectively a contact zone between cultures where identities and cultural forms and inter-cultural relationships can be renegotiated and reshaped, often resulting in something different than the source cultures and sometimes even reshaping them. Naum (2010) applies this concept archaeologically to frontier areas in the Baltic and early colonial North America at different points in time to great effect. These concepts may also be applied to Ireland. Combining archaeological and post-colonial approaches is mutually beneficial, as the latter have often lacked long term perspective, usually confining themselves to the early modern and modern periods. Post-colonial approaches have been further limited by focusing largely on European/non-European relations. Archaeology has much to contribute to the study of colonialism. As Gosden (2004) has shown, it can supply the long term perspective often lacking in post-colonial theory, providing a basis for comparisons across time and space. To understand medieval colonialism is to better understand the shaping of Europe and later European colonialism (cf. Bartlett 1993). Naum (2010, 3) notes that the focus of scholars is naturally coloured by the sources and methods they are trained to work with. Studies through the prism of written sources can be biased towards the “official” version of events, and society (ibid). Jenkins correctly asserts that ‘history is never for itself; it is always for someone’ (1991, 21). This is true in a triple sense, it is for those narrating about, narrated about and narrated to, with each bearing complex links to the other. Certain groups of those narrated about may have the ability to control those narrating. Reflexively used, archaeology can provide alternative narratives, based on material 20 culture and archaeological monuments as opposed to potentially ideologically charged texts. Late twentieth century thought proclaimed, amongst other things, the death of the meta-narrative, perhaps justifiably, at least in some respects. As Collins notes, ‘master narratives may provide powerful characterisations of national identity and the course of history, but they minimise or just plain ignore cultural difference and specificity’ (1995, 8). No two human biographies are the same, and the same holds true for national state biographies. The path each national state, past or present, has taken to statehood has been different, there are no totally self contained political units, as Mann (1986, 1) agrees. Care should also be taken to avoid lapses into Hegelian teleology; there has not been a unilinear path of social development inevitably leading up to the present day (cf. Elias 2000). Mann (1986), Diamond (1997, 2005) and Christian (2004) have shown the benefit of a cross disciplinary comparative approach to the study of long term social dynamics. Studies such as these have moved away from the tendency of many socio- political studies of state formation and social change to derive their data largely from historical sources without paying enough attention to archaeological evidence, which limited their long term perspective. Many recent archaeological studies of state formation have been concerned with pre-historic or ancient societies, without turning the insights gained on our own not too distant past, perhaps to avoid charges of social evolutionism, teleology and over-generalisation in archaeology’s own period of anti- positivism. However, it is still possible to gain an understanding of the Big Picture, not from the meta-narratives of old, but from aggregate-narratives, combining a number of micro-narratives and region specific studies (cf. Collins 1995, 8). It is also hoped 21 that the methodology used in this project can be utilised to study other colonial episodes and examples of state formation across space and time, and that the understanding gained from this project can be used for comparative purposes in conjunction with other relevant studies. By integrating medieval Ireland into the traditions of post-colonial particularism and big-history comparative generalities, it is hoped that it will be possible to move beyond the polemicism that has blighted the study of colonial medieval Ireland. Much insight is to be gained by analysing the period through the lens of post-colonial concepts such as contact zone/third space interactions and processes of hybridity, mimicry and acculturation. Moreover, by placing Ireland within the larger European dynamic, which this project will briefly do, it is possible to gain a good deal of perspective on the processes at work. 1.4 Historical Setting The provision of a detailed historical setting is beyond the remit of this project, historical material will be introduced in the analysis where required, but some general trends regarding the development of Irish and European society will be discussed at point. Relevant full narratives are available in the various relevant texts mentioned in section 1.3. The years in question were characterised by a number of processes. The first is a centrifugal pattern of social development across Europe, from societies where social power was more heterarchically dispersed amongst a contending elite, towards a society where power has become concentrated in fewer hands. This was the case for Ireland c.700-1170, but it was also the case for the home societies of those colonising Ireland in this period. This will be discussed further in later chapters. 22  Scandinavian colonialism in Ireland was preceded by several decades of raiding, largely at Irish monastic foundations. Settlement most likely began in the 830s, most likely to provide over-wintering and a reasonably secure base of operations for raiders. The raiders most likely came from the west coast of Norway. There may have been a failed attempt to gain control over territory in Ireland at this point. Evidence for attempts by those further up the hierarchy in Scandinavia to gain control over Scandinavian activity in Ireland is to be found in the historical record in the late 840s and 850s. Some settlements seem to have become permanent in the latter half of the ninth century, and in the tenth century several towns seem to have been in operation, such as Dublin, Limerick, Waterford, Cork and Wexford. These towns became caught up in the process of Irish state formation, and by the eleventh century they were largely under some form of control by Gaelic Irish elites. Anglo-Norman colonialism began with an intervention by a group of out-of- favour nobles from the Welsh Marches under Richard FitzGilbert de Clare, or Strongbow, an on behalf of Diarmaid mac Murchadha, the recently deposed Uí Chennselaig King of Laigin. Diarmaid had probably drawn inspiration from the successful settlement of Norman groups in Scotland under David I, which had resulted in a major boost for that king’s power (Bartlett 1993; Duncan 1975). The military superiority of a group at the heart of military developments in Europe and beyond over the previous two centuries meant that a large swathe of territory quickly came under the control of this king and his allies. This led Henry II, king of England to arrive in Ireland with a large force, in order to bring this group under his control. This meant that colonialism in Ireland was centrally controlled from then on. Anglo- Norman colonists gained control of a large part of Ireland, holding some of it for a considerable length of time. However, from the late thirteenth century that control 23 began to decline due largely to demographic, financial and environmental factors, and to military and political developments elsewhere. The remit of this project ends in the late fourteenth century, as it is at this point the Irish social developmental trajectory enters another phase, with the contraction of the area under control from London to the greater Dublin area, and the southern part of Wexford, and the rise to prominence of Anglo-Irish families of original Anglo-Norman extraction, prior to the Tudor reconquest in the sixteenth century. 24 Figure 1.1, map showing elements of the political geography of pre-Anglo-Norman Ireland. Province- kingdoms are capitalised, and the names of lesser kingdoms and peoples mentioned later in the text are in lower case. 1.5 Thesis Outline Chapter 2 discusses the plausibility of the identification of certain monuments as colonial, and the related epistemological issues. It also introduces the case-study regions, and discusses in full the methodology and categories of evidence used in the 25 project. Chapter 3 deals with Scandinavian colonial settlement, while also addressing pre-colonial Gaelic settlement, and Gaelic settlement contemporary to colonial activity in each of the case-study regions. It also deals briefly with the pre-colonial raiding phase. It assesses the effects of colonial activity on both coloniser and colonised along with the success and failures of the Scandinavian colonial endeavour in terms of the proposed expansion, consolidation and domination model. Chapter 4 does the same for the Anglo-Norman period, beginning with a continuation of the description of Gaelic society, before discussing the re-ordering of the landscape in the phases of Anglo-Norman expansion, consolidation and domination, while also discussing the related parallel processes of acculturation, creolisation and hybridisation, and the retraction of the colony. Chapter 5 comparatively discusses the key features of the colonial phases, while also discussing cross-cultural parallels, and puts forward a theory of colonialism. The final section of the chapter puts forward a theory of colonialism based on the work conducted in this thesis, in the light of other existing theories, and its suitability for use in the analysis of other colonial processes at other points in space and time discussed. Chapter 2. Methods 2.1 Identifying Colonial Monuments in Medieval Ireland This project identifies certain archaeological monuments with certain groups in medieval Ireland. This has not been done without first considering the caveats 26 provided by recent scholarship on ethnicity and identity. Ethnicity is an extremely relevant topic in modern society. The widespread ethnic conflict of recent decades has led to an increasing prominence of scholarship related to ethnicity across the human and social sciences (cf. Edwards 1985, 115; Jones 1997, 40ff). Moreover, recent advances in the study of population genetics have brought the genetic aspect of ethnicity back into discussion (Cavalli-Sforza 2000; Jobling 2001; Capelli et al 2003; McEvoy et al 2006; Moore et al 2006; Hill, Jobling and Bradley 2000). While ethnicity and genetics are linked, ethnicity is largely socially constructed; ethnic identity involves a person’s identification with a broader group which has been set apart from other groups on the basis of a perceived cultural difference and/or common descent (Jones 1997, 13). This is supported by the findings of studies such as McEvoy et al (2006),3 which have shown that an entire ethnic identity and associated material cultural package can be adopted by a “people” not usually associated with it. This is indicative of the highly problematic nature of the identification of certain types of material culture with supposedly discrete cultural units and ethnic groups. The reading of unsuitable present day conceptions of ethnicity and identity back into the past can be epistemologically unsound (cf. Jones 1997). Nelson (2003) draws attention to existence of the holding of multiple identities by various groups and individuals in the medieval period and to the fluid nature even of ethnic identity. This is highly relevant, as we shall see for Irish Viking Age cities. Ethnicity can be an extremely important anchorage for identity, but not the only anchorage. There are numerous examples of multiple overlapping identities to be found throughout the record of human society. Religious affiliation, consumption choice, geographical 3 While population genetics techniques, such as the examination of Y-chromosomal traits in certain populations, are still in their infancy, genetic evidence could become as important for archaeology as radiocarbondating. 27 location and participation in mass events can all provide such anchorages in contemporary society (cf. Durkheim 1893; 1912; Bourdieu 1984; Maffesoli 1996). Material culture can be an important mediator in identity formation, providing a means of shaping both habitus and identity (cf. Bourdieu 1977; DeMarrais 2004; Gosden 2004. The culture history archaeological paradigm was/is concerned with attempting to categorise artefacts and settlement forms into discrete culturally and biologically homogenous units. However, it fails to take cultural processes into account, it fails to address overlapping identities, and fails to take the epicurean nature of humanity into account. Archaeological thought has sought to move on from the description of past archaeological cultures and onto themes such as environmental adaptation, state formation, and the phenomenological aspects of monument and material culture amongst others. It has largely sought to move beyond the identification of peoples, and to even ignore it. However, the history of knowledge has thought us that paradigm shifts can tend to move too completely from those paradigms preceding it (cf. Kuhn 1970; Popper 2002). Certain groups did build specific monuments and used certain types of portable material culture. Therefore, it is both possible and plausible to identify certain groups with certain forms of monument, at least over short spaces of time. In a colonial or migratory episode, new monument forms may appear in an area outside of the society within which they originated. They may be taken, at least initially as indicating activity by a new non-native group in a new area. However, changes may occur in their use and meaning after an initial period of conservatism, and they may be adopted by elements of native society. 28  This is not an apology for culture history. The existence of a historical record for the period in question means that it is possible to explain the appearance of new monument forms in the landscape in terms of their colonial (or non-colonial) nature. Certain monuments can be identified as part of a colonial package, all with origins outside Ireland. Other monuments might be developed in a colonial setting as a response to conditions there, or to changing goals there. Cultural homogeneity does not necessarily preclude genetic biological heterogeneity. Despite the existence of colonial society, the ethnic make-up of that society might be mixed, as will be discussed regarding Hiberno-Scandinavian towns. The monuments in this project belong to a number of groupings: Gaelic, (Hiberno-)Scandinavian and Anglo-Norman. Colonial monuments also can be seen as basically belonging to one of three sub-processes of colonialism: expansion, consolidation and domination. While these sub-processes will be considered in full later in the thesis, along with related processes such as acculturation, hybridisation and creolisation it is worth discussing them at this point. Sub Process Expansion Consolidation Domination4 Characteristic Campaign Centres of Hierarchical Settlements Fortresses exploitation configuration Viking Period Longphort Nucleated Town, Example settlement, Town rural settlement Anglo-Norman Ringwork Motte Masonry Castle Period Example Table 2.1, the sub-processes of colonialism and associated settlement forms The term Gaelic will be used to signify native Irish in this project. While the terms are interchangeable, the later use of the terms Anglo-Irish (acculturated, hybridised and/or creolised Anglo-Normans) and Gaelic-Irish when addressing very 4 As will be discussed later, the domination phase for Scandinavian activity in Ireland was more one of incorporation and the settlement form listed in the table still is representative of this phase 29 late medieval and early modern Irish society means that it would be useful to use related terminology here. Norse and Viking are largely interchangeable, although neither is entirely accurate, as there were many sub-groupings within Scandinavian society (cf. Nelson 2003; Roesdahl 1998). In Ireland, they referred to themselves as Ostmen.5 The term Scandinavian will be used in this project. The terms English and Norman have both been used to describe the post-1169 invaders, however the term Anglo-Norman is more correct, as it refers to the level of cultural change undergone by the Norman elite in their interaction with the English land and people, just as the term Norman does for the Danes in Normandy. However, Anglo-Norman as a term is not perfect, itself being short hand for a disparate group with ties to the Welsh Marches, England, Anjou, Normandy and Flanders.6 Any attempt to estimate medieval populations is more than likely doomed to failure, and most attempts are little more than guesswork (cf. Bartlett 1993). Trying to ascertain the ethnic make-up of that population is even more difficult. Without total excavation, the populations of the Hiberno-Scandinavian towns must remain in the realm of conjecture. The figures in the annals are prone to literary convention, exaggeration and hyperbole, as is Giraldus Cambrensis. The high medieval population was probably somewhere between 675,000 and 1,500,000 (Russell 1966, 502). Glasscock notes that the flow of settlers from the east may have been more a steady trickle than a deluge, although in some areas manorial records illustrate that English and Welsh settlers, most likely attracted by the possibility of acquiring increased social status, formed an important element of the south-eastern population 5 Literally men of the East 6 Due to the large number of the initial colonists coming from south-west Wales while also being part of a cross-English channel elite, they could therefore be called Cambro-Norman, or even Cambro- Plantaganet or Cambro-Angevin due to the complexities of royal succession in the period, so even Anglo-Cambro-Norman-Angevin would also suffice. However, for clarity’s sake, Anglo-Norman will suffice 30 by c.1300 (1987, 212; cf. Othway-Ruthven 1951). However, it would seem that the overall colonial demographic impact island-wide was relatively low (Glasscock 1987, 213). This is extremely important when addressing the decline of the colony in Ireland. The ringfort, ráth for those of earth, cashel for those of stone, crannóg (pl. crannóige), and souterrain have all been identified with pre-colonial and colonial period Gaelic society (cf. Edwards 1990, 2005; Mytum 1992; O’Conor 1998, Barry 1987 in general, cf. Proudfoot 1961; Lynn 1975a; 1975b; McNeill 1975; Barrett and Graham 1975; McCormick 1995, 2008; Stout 1997; Kerr 2007; Comber 2008 and FitzPatrick 2009 on ringforts, O’Sullivan 1998; O’Sullivan 2000, 8; Edwards 1990; 2005; O’Conor 1998, 79ff; Barry 1987, 18-20 on crannóige, on souterrains cf. Clinton 2001; Stout 1997 and McCormick 2008). The dating of ringforts is disputed, and both the sample of absolute dates, ~0.1% of the total number and the geographical biases in excavation illustrate that generalisations for the entire island cannot be made. As it stands, the majority seem to date to between c.600-900. However, the evidence put forward in papers such as FitzPatrick (2009) has shown that they continued in use in some areas far beyond this. Ringforts, along with the souterrain, will be taken as a proxy for Gaelic settlement throughout the project. 31 Figure 2.1, internal area of Lissanard Cashel, Ennis, Ireland, (author’s photograph) Each ringfort has been positively identified by the Archaeological Survey of Ireland (ASI), and classified as a ráth, cashel or unclassified ringfort (where a ringfort which cannot be definitively assigned to either class). There remains a margin of error for any interpretation, namely those sites classified as “enclosure” by the ASI, the nature of whose remains defy classification and dating, and may be of several different shapes. A number may have been ringfort sites, or were possibly constructed by colonists, or as part of more recent farming activities, and so they are a highly problematic dataset, but one that needs to be given at least some consideration. It is possible to identify Gaelic ecclesiastic activity in the landscape. According to Ó Carragáin (2009), there were no new ecclesiastic foundations in Ireland after c.800AD. This means that ecclesiastic enclosures and round towers can be taken as indicating the distribution of early medieval monastic sites in Ireland, as 32 the latter, although later, are all located on the former. This also holds for some later cathedral sites, and some churches, as some, but not all of these originated in this pre- 800 period. It would seem as though both the Scandinavian longphort (Kelly and O’Donovan 1998; Kelly and Maas 1999; Maas 2008; cf. Ó Floinn 1998; Gibbons 2005; Gibbons and Gibbons 2009) and town (Wallace 2005; 1985a; Clarke 1998; Hurley 1998; 2010; Hurley, Scully and McCutcheon 1997) fall into the category of adaptive responses by a colonial group to local conditions. While there might possibly have been towns in Ireland prior to the Viking period (cf. Doherty 1980; 1982; 1985; cf. Valente 1998; Swift 1998), it would seem as though they were adopted into the Viking settlement pattern right across their zone of influence at roughly the same time. The appearance of cathedrals and churches in Hiberno- Scandinavian towns provides strong indication of the potentially hybrid nature of these settlements. At this point in time, the extent of Scandinavian/Hiberno- Scandinavian hinterland and rural settlement remains unknown, but important work has been carried out by Bradley (1988, 2009) determining the possible extents of such hinterlands, and by Kelly (2010) on rural settlement. The distribution of so-called Viking Age hoards has not been mapped, as they are more of an indication of interaction between coloniser and colonised rather than of colonial settlement. Sheehan (2000, 54) notes that ‘with the exception of two recently discovered late tenth-century coin-hoards from Dublin, there are no silver hoards on record from unequivocally Scandinavian contexts in Ireland at all’. Kenny (1987, 514) notes that ‘the great majority of hoards are likely to have been deposited by Irish rather than by Viking hoarders’. Kenny also agrees with Graham-Campbell’s assertion that 'there is no clear-cut evidence in these [hoard] distributions to support 33 any suggestion of significant inland Scandinavian settlement during the period from the mid-9th to the mid-11th century' (Kenny 1987, 515; Graham-Campbell 1976, 46). Figure 2.2, view of the River Suir from the western end of the longphort at Woodstown, Co. Waterford, (author’s photograph). The ringwork, motte and masonry castle all have their points of origin in the Rhine-Loire region of Europe as part of the process of post-Carolingian feudalisation (cf. de Meulemeester & O’Conor 2007; Bloch 1961; Elias 2000). They were adopted into the hybridised settlement pattern of Scandinavian colonists in northern Francia in the tenth and eleventh centuries, and formed central elements of the Norman colonial package in England (Allen Brown 1970; 1985; Cathcart King 1988; Platt 1978; Pounds 1990). Orderic Vitalis thought that it was by their castles that the Normans established themselves in England (Platt 1978, 3). They then became part of the colonial package for the Anglo-Norman expansion into Ireland after 1169 (cf. McNeill 1997; Sweetman 1999). 34 Figure 2.3, motte at Hen Domen, Montgomery, Wales (author’s photograph). Figure 2.4, Anglo-Norman masonry castle at Trim, Co. Meath, Ireland (author’s photograph). 35  Over 5000 moated sites are known in England, dating from the twelfth to the eighteenth centuries (Clarke 1984, 49). They are found from Ireland in the west to what is now Poland in the east, although they vary in date by region (Wilson 1985, 56-7). In England and the Low Countries they can be found at the centre of nucleated settlements or in dispersed settings from the twelfth century onwards (Wilson 1985, 57). In Ireland moated sites largely date to the period after the initial Anglo-Norman incursion in Ireland, possibly representing a secondary movement of settlers (cf. Glasscock 1970; Barry 1977, 1987, 2000; O’Conor 1998; O’Keeffe 2000). Rural nucleated settlement also formed part of the Anglo-Norman colonial package, having not been present in Ireland prior to the invasion. These are represented in the landscape today by the monuments classed as deserted medieval settlements, although some modern villages may occupy the sites of medieval villages (Buchanan 1970; Glasscock 1970; Barry 1977, 1987, 2000; O’Conor 1998; O’Keeffe 2000). O’Conor notes that a large number of villages may be undiscovered, especially those abandoned earliest (1998, 46ff). The substantial remains uncovered at sites such as Mullaghmast, Co. Kildare (Stephenson 2009) would seem to support this. Dispersed or semi-dispersed undefended houses have also been uncovered by excavation in what would seem to be areas within the Anglo-Norman maximum zone of influence (O’Conor 1998; Barry 1987, 2000), but the true distribution of this form may be as difficult to assess as its Neolithic predecessor (Ó Ríagáin 2010). Urbanism in the Anglo-Norman period is addressed by Graham (1985, 1993, 2000), Wallace (1985b), Bradley (1985, 1995), MacNiocaill (1985), Barry (1987), and O’Keeffe (2000). 36 Figure 2.5, Dominican Priory at Athenry, Co. Galway, Ireland (author’s photograph). While the appearance of parishes followed the division of the island into a number of territorial dioceses in the twelfth century reform of the Irish church, the process was in its infancy on the arrival of the Anglo-Normans (Duffy 1997, 73; Watt 1972, 209). It was in the years after the invasion that the widespread appearance of parishes occurred, usually in tandem with the division of colonial land into manors (Duffy 1997, 73). Therefore, medieval parishes and parish churches can be taken as indicators of Anglo-Norman settlement activity, at least in areas held by Anglo- Norman colonists. Continental religious orders, such as the Franciscans, Augustinians, Dominicans, Benedictines, Cistercians, Carmelites, Knights Hospitallers, Fratres Cruciferi, and the Order of Tiron, are associated with the period after the twelfth century reform of the Church in both Ireland and Europe. While some of their arrivals, such as that of the Cistercians, date to before the arrival of the Anglo-Normans, they can be considered as playing a part in the colonial process, as they represent growing outside influence in Ireland. The location and patronage (using textual evidence) of these new houses are extremely important for this study, as 37 both are strong indicators of the relationship between an ideology shaping institution like the Church with both native and colonial secular power. Secular Ecclesiastical Gaelic Settlement Ringfort, crannóg, Church, ecclesiastic site; souterrain, deserted cathedral, round tower; settlement (later) religious house (later) Scandinavian Settlement Longphort, town Church, cathedral Anglo-Norman Settlement Ringwork, motte, masonry Church, cathedral, castle, moated site, religious house deserted settlement, dispersed house, town Table 2.2, Characteristic monuments associated with each group 2.2 Case-Study Regions The mid-west and south-east have been chosen as representative case studies as the colonial process in both areas was similar in the Viking period and differed slightly in the Anglo-Norman period. Also, as Wallace states, ‘among Hiberno-Norse towns, Dublin has received the lion's share of scholarly attention’ (1992, 36). The south-east was the point of entry in the Anglo-Norman period, for both political and geographical reasons, and remained one of the core areas of the colony even after its retraction. The process of colonisation in the mid-west seems to have begun a number of years afterwards due to its geographical location and political figuration. Anglo- Norman colonial activity in both areas saw quite different results, with the south-east traditionally being regarded as having been much more anglicised and the mid-west having seen little more than a century of Anglo-Norman activity. The areas contain a number of historical polities. Co. Clare largely corresponds to the kingdom of Tuad 38 Mumu or Thomond.7 The urban centre at Limerick had a strong relationship with this polity, and at various points in the historical record seems to have been part of it, and is therefore included. The southern shore of the Shannon Estuary has also been included in order to gain a more regional perspective, although the focus will largely be on Clare. The south-eastern case-study area corresponds largely to the Uí Chennselaig portion of the kingdom of Laigin (what is now the south-eastern part of Leinster), the kingdom of the Osraige, roughly corresponding with the modern county of Kilkenny, which at various points was part of Munster and Leinster, and the eastern portion of modern Co. Waterford, which saw both Scandinavian and Anglo-Norman colonial settlement. 7 Literally “North Munster”, it can also be found in sources as Tuadh Mumhan, 39 Figure 2.6, location of case-study regions and modern administrative boundaries 40 2.3 Methodology The project is necessarily desk based, examining the existing literature on the relevant sites and mapping the settlement pattern over time using the data available from the Archaeological Survey of Ireland (ASI). The settlement forms associated with colonialism have already been identified and surveyed by the ASI and their classification has largely been retained, although some adjustments have been necessary after consultation with the relevant scholarly literature. One such adjustment has been to divide the monument class ‘church’ by time period, which has only been possible here mainly for Wexford and Waterford, using the information and aerial photographs at http://www.archaeology.ie. Elsewhere, the entire set of pre- 1700 churches has been used. The monument class ‘longphort’ has as yet not been utilised by the ASI, but is used in this project. Occasionally, sites have also been reclassified in the light of recent scholarship. A scoping analysis was carried out in the early months in order to set the extent of the case-study areas, in order to ensure the feasibility of the project. This was done by forming a spreadsheet containing each of the sites to be used, using the information available from the (ASI) via the website http://www.archaeology.ie and the relevant published inventories. The spreadsheet contained fields noting the SMR number, townland, county, classification and National Grid coordinates of each site. In addition to this, fields noting the parish, barony, spatially associated monuments and potential classificatory doubt were included where relevant. This was followed by a thorough examination of the existing literature on the relevant monument forms. A period of consultation of the relevant primary textual material was also necessary, in order to gain an emic view of the monuments featured 41 in this project, and to ensure that a fuller picture of the settlement pattern was obtained. Once the dataset had been finalised, the relevant shapefile data was downloaded from http://www.archaeology.ie and distribution maps formulated using the Geographical Information Systems (GIS) program Archmap 9.2. The relevant geographical data was obtained via: http://www.gdem.aster.ersdac.or.jp/ for the elevation geodata, http://www.diva-gis.org/datadown for the river/lake data, and the Dept. of Geography, National University of Ireland Galway for the Ireland shapefile. Once this phase had been completed it was possible to form distribution maps of the relevant monuments in a number of time-slices in order to examine their distribution pattern and through this the process of colonialism in both the Scandinavian and Anglo-Norman phases. These time-slices are intended to be representative of the expansion, consolidation and domination phases of colonial activity. The relationship between the physical environment and the colonial process has been assessed by superimposing the relevant layers on the same map. Mountains and forest may have provided natural limitations to colonial activity. River systems played an important role in both colonisation and interaction between colonised and non-colonised areas. Areas of obvious colonial activity are interpreted as those with moderate-to- high densities of colonial monuments based, for present purposes, on qualitative examination of the distribution maps. These monuments may be related synchronically or diachronically. That is they may have a contemporary spatial relationship, or they may be related to each other across time, as different phases of related activity. Examples of the former would be the distribution of mottes in a colonial area, or the distribution of mottes and Anglo-Norman medieval parish churches. Examples of the latter are found in the location of later masonry castles on 42 the site of earlier ringworks or on the former sites of Gaelic pre-colonial high-status monuments. Almost every distribution naturally has its outliers, and the distributions analysed here are no exception. These can be in the form of single, or very small numbers of, colonial monuments appearing in the midst of areas characterised by Gaelic settlement. These may be due to the adoption by natives of colonial forms via social processes such as acculturation, or they may be indicative of failed attempts at expansion due to the agency of single individuals or families. They may also be part of a larger pattern which remains undetected. These outliers will be discussed in the analysis on a case by case basis where necessary. There may also be areas of mixed distribution that cannot be chronologically separated. This may mean the coexistence in some areas of colonial and native settlement forms, which is highly likely to have been the case in many areas and when dealing with these care has been taken to heed the warning of Jones (1997) and not read modern conceptions of identity back into the archaeo-historical record. Due to the time and spatial constraints involved in a master’s thesis, statistical tests of significance for spatial analysis will not be utilised in the project. This decision has not been taken lightly, as techniques such as Monte Carlo simulation (Conolly and Lake 2006, 161-2) can be useful. By doing so, the data may be utilised in future research for such statistical tests, possibly for comparative purposes with another, non-Irish, dataset at PhD level. All Irish language placenames contained in the annalistic references have been translated into their modern English names. Where I have not been able to make the translation myself, use has been made of Hogan’s Onomasticon Goedelicum 43 (http://publish.ucc.ie/doi/locus/A#navtop), and every placename has been checked against this document. Chapter 3. Scandinavian Colonialism in Ireland Figure 3.1, river and lakes mentioned in the text, important conduits for Scandinavian activity. 44 3.1 Introduction While there may have been some rural settlement, the main impact of Scandinavian colonialism on the settlement pattern has been urban, with several of Ireland’s main cities being Norse foundations, including the three largest in the 26 Counties: Dublin, Cork and Limerick, in addition to Waterford and the towns of Wexford, Wicklow and Arklow (cf. Wallace 1985a; ). Most of these settlements were located in what were originally politically peripheral locations, at least from a Gaelic perspective (cf. Ó Floinn 1998). From a Scandinavian perspective, such locations, with their access to water, were perfect for the expansion and consolidation colonial phases, providing bases of operations for activity further upstream. However, it is difficult to identify a phase of Scandinavian colonial domination in Ireland. Scandinavian settlements became caught up in the centrifugal state formation process of the tenth to twelfth centuries in Ireland and incorporated into Irish socio-political life. That is not to say the colonists became fully acculturated, rather they underwent a mixture of hybridisation and creolisation in what is best termed the Hiberno-Scandinavian settlement phase. The chapter will first turn to the expansion phase for each of the case-studies, before discussing the consolidation and incorporation phases, before finishing with a discussion of the processes of acculturation, creolisation and hybridisation, in addition to further exploring problems of identity and ethnicity. A combination of archaeological and textual evidence is necessary to gain a better understanding of Scandinavian settlement in Ireland, and recourse will be made to textual evidence in order to set the archaeological evidence in context. A great deal of textual information has survived via annals such as Chronicon Scotorum (CS), Annála Uladh (AU), the Annals of the Four Masters (AFM), which provide a useful aid to the archaeologist. 45 3.2 Scandinavian Colonial Expansion in the Mid-West The Shannon, Ireland’s longest river, capable of providing access far inland, seems to have been a major hub of Scandinavian activity from the 830s onwards. A battle was in northern Kerry, over “the heathens”8 in 834 (CS834, AU834.8). Mungret and other churches in west Munster were burned the following year (CS835). AU836.10 notes the Déis Tuaisceirt’9 defeated by the raiders, in a year which Ó Cróinín (1995, 237) dubs ‘a crescendo of Viking activity, particularly in the west’. The raids on Iniscealtra, close to the Déis Tuaisceirt heartland (CS837) and Clonmacnoise (CS837, CS842) provide evidence for Scandinavian raiding further up the Shannon system. The foreigners of Loch Ree, the second largest lake on the Shannon, are mentioned in AFM842.13, AFM842.14 and CS844. In CS845, Scandinavians under Tuirgéis built a dún on Loch Ree, the second largest lake on the Shannon, from which they raided Clonmacnoise. It is in this context that the longphort (pl. longphuirt), lit. ‘ship-fortress’,10 at Athlunkard, derived from Áth an Longphort [the Ford of the Longphort] in Fairyhill townland, Co. Clare must be examined. The term dún, which usually refers to a high status fortress, might also be used interchangeably in the contemporary sources. It is also later used to describe the enclosing wall of a town (cf. Maas 2008; AI972.1). Longphort distribution is vital to the understanding of the expansionary phase of Viking colonialism. The form that these fortresses took has been a matter of some debate, with a D-shaped structure, similar to that at Repton in England (Roesdahl 1998, 236-8), being put forward by some scholars (Kelly & Maas 1999). The vast 8 The term Viking is of course never used, instead the usual terms are variants of gentibh, best translated as heathens or gentiles, and gall, foreigner 9 This group would later become known as the Dál gCais (cf. Ó Cróinín 1995, 237) 10 Later usage of the term in contemporary sources are less precise, where it becomes interchangeable with the word for campaign fortress, dúnadh, and later again as one of a number of generic words for fortress (Maas 2008; O’Conor 1998). 46 majority of evidence for longphuirt is textual. Ó Floinn (1998, 164), citing the paucity of archaeological evidence, points out that there may not have been some form of pre-existing longphort-type defended settlement model in Ireland or elsewhere, and instead points to a possible opportunistic use of pre-existing monuments. Iron objects dating from the final centuries of the first millennium AD were found on the site, including a plough coulter, a spearhead, a spear-butt and a small ring or hoop (Kelly and O’Donovan 1998, 14). Nearby, on the shore of St Thomas' Island, an iron axe was found which still contained part of a wooden handle, and a possible Scandinavian weight was found at Corbally on the riverbank opposite (Kelly and O’Donovan 1998, 14). A detailed survey of the site revealed the presence of a D- shaped site, enclosed by a curved rampart, located where a stream runs into the River Shannon and is open to the stream and river (Kelly and O’Donovan 1998, 14). The site lies just below the tidal range of the river near a fording point and modern bridge (Kelly and O’Donovan 1998, 14). Beyond the rampart there is an intractable marsh which affords natural protection. Within the enclosure, which is 75m long and 30m wide, there is an oval raised area measuring 20m by 12.5m, protected formerly by a ditch and counterscarp bank (Kelly and O’Donovan 1998, 14). The interpretation of the site as being a longphort is disputed by Gibbons (2005) and Gibbons and Gibbons (2008), where the interpretation of the artifactual evidence as diagnostic is questioned, as are the site’s defensibility and the perception of the D-shape as being diagnostic of a Viking encampment in Ireland and beyond. While there would be some degree of justification in such a minimalist interpretation, especially if the site’s interpretation was based on just one category of evidence, the 47 convergence of several possibly diagnostic features in one site indicates that the site was a longphort. It is highly probable that the activity in the lower Shannon region, the activity in and around Loch Ree and the site at Athlunkard are related. Kelly and O’Donovan (1998, 16) may be correct in their assertion that Athlunkard may have been intended to secure the approach to Loch Ree, or it may be that activity on Loch Ree was secondary to Athlunkard. Regardless, a deliberate decision is apparent in the placing of a base near a ford close to the tidal range of the Shannon and on a major inland lake close to the confluence of the Shannon and several navigable tributaries. The temporal relationship of Athlunkard to Limerick is difficult to ascertain. There are a number of possibilities. The sites may have existed contemporaneously, or sequentially. It may be that Athlunkard was the original Scandinavian settlement in the region (Hodkinson 2002; cf. Kelly and O’Donovan 1998, 16; Gibbons 2005). CS845 refers to the ‘ships of Limerick’, which seems to indicate some form of tie to the area, so perhaps there was a longphort already established in the area at this point, either at the site of the later town at King's Island, or at Athlunkard. Athlunkard may have preceded the settlement at Limerick, or it may have been an outlying fortification protecting the ford and related to the defence of the main settlement at Limerick (Kelly and O’Donovan 1998, 16). If they were contemporary, then a significant element of planning went into Scandinavian expansion in the region. Ó Floinn (1998, 162) notes the location of many of the Viking bases at the borders between kingdoms. It is evident from figures 3.2 and 3.3 that Scandinavian settlement in the mid-west took place in liminal zones in Irish political geography. The longphort at Athlunkard, located on the River Shannon, lies on the frontier between the Dál gCais and the Eóganacht. Moreover, it occupies an area relatively 48 empty of Gaelic secular settlement when mapped along with the known diagnostic monuments, as can be seen from figure 3.2. Figure 3.2, Scandinavian/Hiberno-Scandinavian and Gaelic secular settlement in the Mid-West, note the low settlement density around the colonial sites. Figure 3.3, known Scandinavian and Gaelic secular settlement in the mid-west, with undated enclosures added. 49  As can be seen from figure 3.4, Scandinavian settlement in the mid-west seems to have had more of an association with early ecclesiastical settlement than it had with secular settlement in figure 3.2. Athlunkard and the ecclesiastic site at Kilquane are in close association, and were probably contemporary. This is echoed again at Wexford, Cork and Dublin, but not at Waterford or Woodstown. From the evidence of the maps, it might be tentatively stated that Scandinavian settlement in the mid-west was located in an immediate area containing low secular settlement, but could have been associated with ecclesiastic settlement Figure 3.4, Scandinavian settlement with maximum possible contemporary ecclesiastic settlement. 50 3.3 Scandinavian Colonial Expansion in the South-East Although the eastern half of Ireland saw much more raiding activity from the 790s to the early 830s and beyond, the south-east case-study saw much less raiding activity than further north, even after the intensification of the 830s (Valente 2008, 167-8). 825 saw a Scandinavian defeat of the Osraige (CS825, AU825.12) and the plundering of Inis Daimle11 (CS825). CS828 saw a battle-rout (cathraoinedh) inflicted by the Uí Cheinnselaig and the community of Taghmon on the “heathens”. This was probably a failed raid as opposed to an attempt at colonial activity. In 835, both Clonmore and Ferns were raided (CS835, AU835.5). The former must have sat at the head of an efficient economic system, as it suffered devastation (vastatio) on Christmas Eve the following year, with many being carried off, possibly indicating slave raiding (CS836). The Rivers Suir, Nore and Barrow, known as the Three Sisters, enter the sea together, and collectively provide access deep inland. The Slaney, which reaches the sea at Wexford, also provides access inland. It is hardly surprising that the known Scandinavian settlement in the region would be associated with these two inland waterway systems. It is probable that the longphort at Woodstown represents the earliest known Scandinavian settlement in the region. The site was unknown and undocumented prior to the site’s discovery and partial excavation ahead of a road scheme from 2003 to 2007 (O’Brien and Russell 2005; Russell 2003). The excavations point towards the site being occupied in the ninth and/or tenth centuries, by a Scandinavian or Hiberno-Scandinavian community, with several phases of activity evident (O’Brien and Russell 2005). While its ditch and D-shaped enclosure suggest military functions, the site seems to have been a centre of trade and 11 Probably Kilmokea on Great Island, but could be Little Island (cf. Charles-Edwards 2006, 150). 51 production, as demonstrated by the quantity of lead weights and hack silver, in addition to evidence for iron, copper alloy, silver, glass and perhaps lead-working, woodworking, ship repair and textile production (O’Brien and Russell 2005). Woodstown is also in an area of low Gaelic settlement on the periphery of two Gaelic polities, i.e. the Déisi Mumhan and Osraige, and relatively close to the provincial boundary between Mumu and Laigin. While it lies several kilometres upstream on the Suir from the confluence of the three rivers, it would still have provided access to each. As can be seen from figures 3.5 and 3.6 the proposal that Scandinavian settlement took place in areas of low Gaelic settlement density holds. There may have been a temporal overlap between Woodstown and the settlement at Waterford, but an answer to this question will have to await further excavation at both sites. Scholars remain unsure as to whether or not there was a longphort at Waterford prior to the activity so far uncovered by excavation, and it would most likely have been in the area associated by Hurley with the initial phase of development (Hurley 2010, 155, fig. 15.1). 52 Figure 3.5, known Scandinavian and Gaelic secular settlement in the south-east Figure 3.6, known Scandinavian and Gaelic secular settlement in the south-east, with undated enclosures added. Even within this maximum possible extent, the argument that Scandinavian settlement avoided areas of high secular settlement holds. 53  Figure 3.7 illustrates the relatively low ecclesiastical settlement density around the Scandinavian settlements at Woodstown and Waterford. However, Wexford, which has not been discussed up until this point, seems to be in an area of relative density, with the site of the possible early ecclesiastic foundation of St Iberius’ (Moore 1996, 163) directly in association with the later town. Perhaps this may have been the site of the initial longphort, if Ó Floinn’s (1998, 164) suggestion that some longphuirt may be actually reused native settlement forms is valid, although there were probably more defensible locations in the immediate area. The placing of a longphort at the mouth of a major Irish river, the Slaney, would certainly conform to the pattern of longphort placement identified elsewhere in the two case-studies under examination. Figure 3.7, known Scandinavian and early ecclesiastic settlement in the south-east, the term ecclesiastic site refers to sites definitely datable to the early medieval, with a bias towards Co. Wexford and the eastern part of Co. Waterford apparent, where I have been better able to map church sites by period, cf. chapter 2. 54  It is significant that rivers, which served as boundaries for the Gaelic Irish, served as conduits for Scandinavian raiders and colonists. This can be taken as signifying different perceptions of space in both groups. For the Gaelic Irish, these boundaries served to limit action, whereas for Scandinavians they facilitated action. The location of Scandinavian settlement on the frontiers of Gaelic polities may have eventually been encouraged by the elites of these polities, provided they could come to some form of peaceable arrangement. They could have provided an important buffer zone between rival polities. The transition from fortified encampment to urban centre in the three Norse/Hiberno Norse towns under consideration must have taken place with at least some elite Gaelic collusion. Limerick is located c.80km from the open sea, and while the Shannon Estuary is several kilometres wide in places, it would have been difficult for regular shipping to have taken place without threat of attack unless there was some form of local alliance. It is here that some of the historical evidence may be of use, as the rise of the Dál gCais under Cennétig and his sons may have been facilitated by the rise of Limerick (cf. Ó Cróinín 1995, 237). The same may also be possibly said for the Osraige, whose rise to prominence corresponded with Scandinavian settlement in the south-east (Downham 2004). 3.4 Souterrains: A Gaelic Response to Scandinavian Colonialism? McCormick claims that the construction of souterrains is a reaction to the commercialisation of slavery (2008, 221), an intensification which probably coincided with the colonial consolidation phase (cf. Holm 1986). Souterrains are essentially underground chambers, dating to c.750-c.1250 (Clinton, 2001, 95), although few absolute dates are currently available.12 Clinton asserts that the range of impedimental devices indicates the defensive nature of a sizeable number of souterrains in Ireland, 12 This should be rectified with the publication of many of the recent developer led excavations. 55 and thus their construction primarily as refuges (2001, 60), although they have been death traps in the event of a sustained attack (Edwards 1990, 30). However, a number of souterrains also have ramp or stepped entrances, which would seem to imply ease of access and thus a storage function (Clinton 2001, 60), possibly for the both pastoral and arable agricultural surplus (Stout 1997). Any direct connections to Scandinavian activity are tenuous. Their distributions are uneven, and their only discernable relationship to the pattern of Scandinavian settlement is that it occurs in areas associated with low local souterrain density, as can be seen from figures 3.2 and 3.5. They probably were related more to the commercialisation of agriculture than of slavery, although they may have occasionally functioned as shelters from such raids. 3.5 Scandinavian Colonial Consolidation in the Mid-West The growth of Scandinavian colonial towns in Ireland should be regarded as part of the sub-process of consolidation rather than domination, representative of a strengthened grip on what was already held. While hinterlands seem to have developed around them, they were not part of a colonial central place hierarchy of domination like their later Anglo-Norman counterparts, and the third phase of Scandinavian colonialism in Ireland was rather one of incorporation, as will later be discussed. Surprisingly little is know about the genesis and early years of the Hiberno- Scandinavian towns in Ireland. Each may have been preceded by a longphort, as was most likely the case in Dublin. Wallace (2005, 818) notes that ‘although well documented historically, viking-age Limerick has not been as archaeologically productive as either Dublin or Waterford’. The Scandinavian settlement at Limerick 56 was situated underneath the area of the city known as King’s Island13 (Wallace 2005, 818). The location of the settlement on a river island parallels the siting of Cork (cf. Wallace 1992, 37-8; Wallace 2005; Hurley 2010). At King’s Island, Wiggins (1990) uncovered evidence for three probably Scandinavian houses and a 10.1m stretch of a clay bank revetted by a limestone wall,14 dating to the twelfth century based on artifactual evidence, and which seems to have been bonded to the later east curtain wall of Limerick Castle. This may well have been part of the wall referred to by Giraldus Cambrensis (Expugnatio 2.7.30), and although it may belong to the later incorporation phase, it may have been related to an earlier feature. While acknowledging the existence of several ninth century references to Limerick, Wallace asserts that Limerick appears to have been founded in 92215 (2005, 818). It seems to have been a base of operations for extensive raiding on the Shannon system in this period (AI922.2, AU922.3, CS922, AU922.3, AFM920.19). However, this seems quite late in the chronology of expansion, and earlier references exist, such as that of 845 mentioned earlier. It is mentioned again in 866 as being the base of operations of a Norse Jarl Tomrar (FA337),16 who died later the same year on the Isle of Man (FA 331, FA 340),17 an indication of the mobility of these individuals. In 887, the “foreigners” of Limerick, are mentioned as again being slaughtered by the Connachta (CS887, AFM884.16). 13 Also known as King John’s Island or Englishtown, both references to the Anglo-Norman phase of activity. 14 Surviving to a maximum height of 1.7m, with a 2.8m deep ditch. The houses have parallels possible at Dublin and York. 15 By Tamar mac Ailche, who established a longphort there 16 A very strong, very rough, merciless man of the Norwegians (duine aindreannda, agarbh, aindgidh eisidhe do Lochlannachaib). 17 At Port Mannan, he is reported as having died of insanity there due to a miracle of Clonfert’s long deceased vengeful patron St Brendan, in retribution for having raided the site FA 331, FA 340). His followers seem to have been besieged and defeated after this by the Ciarraige, with ‘great quantities of gold and silver and beautiful women’ left behind (FA341), perhaps an indication of slave raiding and trading in the region. Whether this took place at Limerick or somewhere else in difficult to surmise. 57  The tenth century settlers at Limerick seem to have been engaged in continuous military activity, either with Gaelic groups or other groups of Scandinavians, possibly in an attempt to establish hegemony among the other towns. The conflicts with Scandinavians from Dublin (AU924.3) and Waterford (AI927.2) provide evidence of this, and in the latter case, Waterford was fought by an alliance between the Mumu and Limerick. It is interesting to note that from this point forward, there are no more references to the Limerick Scandinavians raiding in northern Munster, the area where the Dál gCais were growing in power. This surely infers some form of connection between the two groups. This is further supported by the fact that what raids and battles they do take part in are against rivals to the Dál gCais, such as the Connachta (CS929, AFM927.13; CS930; CS934; CS940) and the Osraige (CS930, AI930.1), although the latter seems to have been in collusion with the Dublin Scandinavians. While the evidence is strongest for that of Dublin (cf. Bradley 1988; 2009), Limerick also must have had some form of hinterland, which may correspond to the twelfth century diocese of Limerick, or to its medieval deanery, or to the Cantred of the Ostmen of Limerick of the Anglo-Norman period (Bradley 1988, 62). The most likely area for that earlier hinterland is probably the area relatively empty of Gaelic settlement visible in figure 3.2. The genesis of this hinterland and their entering into some form of alliance with the Dál gCais must be connected. While such an alliance would have been mutually beneficial, especially militarily, it would also have provided the Gaelic elite with a major economic power source, due to the commercial nature of Scandinavian settlement. It would also have meant that Limerick’s residents could agriculturally exploit the town’s environs without threat of attack. The later integration of Limerick into the territory of the Dál gCais would have meant that the 58 territory in its entirety became its hinterland, and any earlier hinterland may have been obscured. 3.6 Scandinavian Colonial Consolidation in the South-East Wallace asserts that the settlement at Waterford, or Port Láirge in Irish, founded in either 853 or 914, favouring the latter (2005, 818-9). The nature of the archaeological evidence as it stands means that very little is known about early Waterford, with the vast majority of evidence being from the early eleventh century onwards (Wallace 2005, 818; Hurley 2010). Wallace’s interpretation seems largely to be based on FA458 and CS914, which note that a great Norse fleet landed at Port Láirge in 914. That this was a foundation event is difficult to surmise. The arrival of the ships might indicate an intention to urbanise an existing settlement, or the beginning of a relocation from Woodstown to Waterford. The site may have been intended to rival Woodstown. Regarding Waterford, Hurley (2010, 154) states that ‘the assumption is that the initial longphort or dún formed the nucleus of the settlements and expanded over the years’, and he sees it ‘as likely as not’ that any earlier phases were in a different location. He also draws attention to the possibility that further research at Woodstown might provide a better perspective on both sites. Like Dublin, (Hiberno-)Scandinavian Waterford is located on a ridge at the confluence of a major river estuary, the Suir, and a minor tributary, St John’s River (Wallace 1992, 37-8; Hurley 2010, 155-7). The marshy ground around the tributary would have provided added protection, as at Woodstown and Athlunkard. Four phases of development of the Hiberno-Scandinavian town have been identified, moving progressively westward between the river and its tributary (Hurley 2010, 155, fig. 15.1). 59  The defences at Waterford seem to parallel those of Dublin (Wallace 1992; 2005, 822; Hurley 2010), and to a lesser extent Limerick (Wallace 2005, 822). A section of bank c.35m long, c.4m wide, and possibly originally 3m high, with a 2- 2.5m deep ditch, which was 8.5m wide at the top and 2.5m wide at the base, with oak beams, dated to 1070-90, possibly forming some form of superstructure in parts of the bank (Wallace 2005, 822; Hurley et al 1997). This bank was demolished in the second quarter of the twelfth century, and a wall built on its back-filled ditch (Wallace 2005, 822). Also similar to Dublin were a series of vertical joints indicating that the wall was built in sections, possibly be different work gangs (Wallace 2005, 822; Hurley et al 1997). This can be taken as evidence for central planning by an authority able to organise work gangs, and although this evidence is largely from the incorporation phase, such centrality may have been a feature of the consolidation phase. The similarity between the defences of the Scandinavian towns can be taken as indicative of a shared concept of defensive technology, and thus a shared culture. The possible hinterland for Waterford most likely corresponds to the area of low settlement density around it evident from figure 3.4. This area seems to correspond to the later Anglo-Norman period Cantred of the Ostmen, or Offath (Bradley 1988, 62-65). Wexford, Loch Garman in Irish, was also built at the confluence of an estuary and tributary, the Slaney and Abbey River respectively (Wallace 1992, 37-8; 2005, 819). The excavations at Bride Street yielded a building sequence from c.1000 to 1300, with a realignment occurring in the mid to late eleventh century, possibly due to an expansion beyond the initial core area (Bourke 1995, 33-6; Wallace 2005, 818, 827). Therefore, Wexford dates to at least the tenth century, and may have been 60 based on an initial longphort, as noted earlier. Few documentary references to the settlement are extant, and the earliest mentions the foreigners of Wexford as having lost a battle, along with those of Waterford (AFM888.6). As can be seen from figure 3.5, the settlement at Wexford is located in an area of low local Gaelic secular settlement. Figure 3.7 seems to indicate a higher degree of ecclesiastical settlement in the area in comparison to Limerick and Waterford. It is more like Cork and Dublin in this respect. At least three major forests are known as having been in the area in the Anglo-Norman period, and it is highly likely that they were even more extensive earlier (cf. Colfer 1987, 68, Roche 1995, 35). One of these, the Forest of Taghmon, surely accounts for the large gap in the settlement pattern apparent in southern Wexford in figures 3.5 and 3.7 (M. Moore, ASI, pers. comm. 2010). The extensive estuarine lowlands known as the Wexford Slobs to the north of the harbour also contributed to the site’s isolation over land. It would be tempting to postulate that the site was settled with the consent of the local Uí Chennselaig dynasty, who may well have viewed the location as being otherwise of little economic use. Wexford is sited in low lying land, of the type ráth and cashel builders avoided (Stout 1997). The benefits of having an outlet for trade operating within their zone of influence may have been apparent to the dynasty. However, in 935, the “foreigners” of Wexford are mentioned as having killed Cinaedh mac Coirbre, king of Uí Chennselaig (CS935, AFM933.8). It is more likely then, at least from this evidence, that the site began life as a longphort built to provide a base of operations for the Slaney basin. That said, some form of local consent may have been necessary as the settlement grew, which holds for all of the Scandinavian settlements which developed into Hiberno-Scandinavian towns. It would seem as though the modern barony of Forth served as the hinterland for Hiberno-Scandinavian 61 Wexford, and Bradley (1988, 62) draws attention to a 1283 inventory noting that there were 40 poor Ostmen where there once had been 100 rich ones. This area is also an area characterised by low Gaelic secular settlement, as can be seen from figure 3.5. 3.7 Incorporation in the Mid-West Limerick was brought under the control of the Dál gCais, their immediate Gaelic neighbours in the 960s and 970s. It was burned by them in 967 (AI967.2, AU967.5), and again in 972, with its foreign inhabitants banished the same year (AI972.1). In 977, Brian Boruma, king of Dál gCais and Mumu effectively ended the Scandinavian royal dynasty of Limerick by killing Imhar and his two sons (AI977.2, CS977, AFM975.8). This can be taken as signifying the incorporation of Limerick into the Mumu polity, and it seems to have become the seat of the Dál gCais the following century, although its inhabitants continue to be referred to as foreigners up until the Anglo- Norman period. It means that Limerick now becomes as much a part of the land orientated local central place hierarchy, being the seat of the regional elite, as well as continuing as a nodal point in the network of Scandinavian settlements connected to each other by water. The appearance of so many round towers in the region from the tenth to twelfth centuries must surely be linked to this. The wealth accrued by controlling Limerick would have filtered through the patronage network of the regional elite, with legitimating institutions such as ecclesiastic sites benefiting greatly, which would have contributed to the ecclesiastic building boom of the period (cf. Ó Ríagáin 2010b). While ostensibly being nodal points in the network of ideological power relations, they were also major economic centres. Their long association with grain production (cf. Edwards 1990; 2005; Stout 1997) may have 62 also seen them involved in trade with Limerick, further boosting their position financially. This could have further contributed to the boom in round tower and stone church construction, as well as possibly funding the crafting of portable objects such as manuscripts and metalwork. As can be seen from the map, there are nine round towers within the direct Dál gCais sphere of influence, with a further three lying in the contact zone with the Connachta to the north. This is quite a large proportion of the overall island-wide distribution, which stands at approximately 96,18 and it is interesting to note that a similar phenomenon occurs in the greater Dublin area. This may also be the period in which some Irish ecclesiastic sites took on some of the characteristics of towns, due to the opportunities provided by the Scandinavian inspired contraction of space and the continued concentration of power in the hands of patrons.19 18 This figure for the 32 Counties combines the figures from each of the two relevant SMRs, which stand at 83 and 12 respectively, plus the second tower at Devinish, Co. Fermanagh which is not listed separately in the SMR for the Six Counties. 19 This could be a matter for some debate, however, as only one major ecclesiastic site has been excavated to date, and the possible urban nature of ecclesiastic sites has been hotly debated by Irish medievalists (cf. Ó Corráin 1972; Doherty 1980, 1982, 1985; Swift 1998; Valente 1998, Etchingham 2010). 63 Figure 3.8, Limerick with regional round towers 3.8 Incorporation in the South-East The Scandinavians of Waterford seem to have followed a similar path to those in Limerick in the tenth century, there are references to alliances with kings of Mumu, (AFM937.14, AFM937.15, AFM967.12). Their raiding activity seems to have been conducted to the north, in the territory of the Southern Uí Néill and the northern Laigin, both rivals of their immediate neighbours. AU983.2 has a reference to a Gilla Pátraic son of Ímar of Waterford, an indication of the adoption of Gaelic names by the Scandinavian elite, probably due to intermarriage. AI984.2 has an alliance between the Mumu and Waterford, against the Osraige, Uí Chennselaig and Dublin. A similar alliance came together to attack Connacht (AFM987.4 and CS988), and in AI1002.4 they are part of Brian Burma’s contingent against the Ulaid. From the late 1030s, a decade which saw three kings of Waterford killed, control over the city seems to have oscillated between the Scandinavian elite, kings of Laigin, and occasionally Mumu 64 (AFM1031.20; AFM1035.3; AU1035.5; AU1037.4; AFM1037.9; AFM1037.15; AFM1134.12; AFM1137.12; AFM1140.10). In 1088, the “foreigners” of Wexford, along with those of Dublin and Waterford, had a ‘great slaughter’ inflicted on them by the Uí Echach of Mumu on the day they intended to plunder Cork (AU1088.4). This is significant in that it shows three Hiberno-Scandinavian towns being defeated by a Gaelic group while acting together possibly against a fourth. It may also indicate that Cork had by this point gone the way of Limerick and been incorporated into its local Gaelic polity. Wexford was later plundered by Toirdealbach Ua Conchobuir as part of a raid into the territory of the Laigin, into which it seems to have become incorporated by this point (AU1128.6). Evidence of the town being walled is provided by Giraldus Cambrensis (Expugnatio 1.3.32, 1.3.35). Whether or not it was capable of having the 2000 fighting men mentioned by Gerald (Expugnatio 1.3.27-8) is difficult to surmise, as the Expugnatio is as much a propaganda piece as a historical narrative. A number of round towers are located in the south-east case-study region, cf. figure 3.9, and a number of them are located close to the Barrow and Nore rivers, possibly indicating a similar connection to that put forward for the mid-west. 65 Figure 3.9, Scandinavian settlement and ecclesiastic round towers in the south-east 3.9 From Scandinavian to Hiberno-Scandinavian: Creolisation, Acculturation and Hybridisation According to Gosden (2004), colonial activity can change both coloniser and colonised. This can certainly be said to hold true for Scandinavians in Ireland. Material culture seems to have developed in a unique trajectory in the Hiberno- Scandinavian towns, as evinced by the unique style of ornamental items such as armrings (Sheehan 1998, 2000, 2005). Wallace (2005) draws attention to the differences between Hiberno-Scandinavian towns and those in Scandinavia. The use of horizontal log walls in contemporary Baltic and Scandinavian houses illustrates the divergent development of towns there and in Ireland, with differences also apparent in defences and yard layout in the excavations at Oslo, Trondheim and Tonsberg (Wallace 2005, 831). 66  Seven house types have so far been identified in Hiberno-Scandinavian towns (cf. Wallace 1992; 2005; Hurley 2010). Their distribution at present may represent the extent of excavation in each of the towns, rather than their actual distribution. Of these seven, only types 1 and 4 seem to have parallels outside Ireland. Type 1 has been the most common type found so far, and has been uncovered at Dublin, Waterford and Wexford, but not Limerick (Wallace 2005, 828). The only known parallels for the type 1 house in Scandinavia are from Kaupang, which was abandoned c.900 (Valente 2008, 63). The type also seems to have been in use in non- Scandinavian northern Europe. This can be taken as being an example of conservative creolisation, where colonists adhere to a cultural trait of their homeland long after it has gone out of use there. Wallace notes a number of parallels with York. Anglo-Scandinavian York was located at the confluence of the Ouse and the Foss, the evidence for a defensive embankment uncovered at Hungate is similar in scale to that used at Hiberno-Scandinavian towns, and York also had sunken featured structures (Wallace 2005, 830-1). It should also be noted that Dublin and York were ruled by the same dynasty until the mid-tenth century (Wallace 2005, 831; Downham 2007), a dynasty which also held Limerick and Waterford until the tenth and eleventh centuries respectively. Mytum (2003) wisely questions the level of integration between the two populations, it would seem as though there was little cultural syncretism beyond the towns, and even there the level of hybridity seems to have been low. Some acculturation did occur, probably in both directions, it was never total in the case of Scandinavian settlers. They participated in Irish society, but were not of it, the inhabitants of the towns retained a distinct identity. One example of acculturation may be the adoption of Christianity in the early eleventh century by the colonists. 67 However, the clergy of Dublin, Waterford, Cork, Wexford and Limerick looked east for metropolitan guidance, to Canterbury, as did the Hiberno-Scandinavian cathedrals at Dublin, Limerick and Waterford for a number of years (Watt 1972). This surely must be taken as a sign of apartedness in the elites of these towns. Furthermore, the inhabitants of each of the Hiberno-Scandinavian towns are termed as “foreigners” in the annals right up until the Anglo-Norman period. In fact, in some ways, Scandinavian colonialism helped foster a stronger shared sense of Gaelic identity, by providing a means for definition by contradistinction (cf. Binchy 1962b). While a number of loan-words from Old Norse were adopted into Middle Irish over the course of Scandinavian activity in Ireland, the overall number was still comparatively low compared with Manx and Gaelic of Scotland, and limited largely to urban life, military and nautical matters (cf. Ó Cróinín 1995, 269-70). This can be taken as an indicator of low Gaelic acculturation. Also, while rectilinear houses do seem to be adopted some time in the eighth or ninth century (Edwards 1990, 26-7) on Gaelic secular settlement sites, there is no evidence for Hiberno-Scandinavian house forms being adopted outside of the towns’ zones of immediate influence. We must see the inhabitants of the towns as living in some form of “third space”, (cf. Bhabha 1994), such as that occupied by the later Anglo-Irish descendents of Anglo-Norman settlers, or by creoles in Latin America. They were culturally different from both their homeland and their adopted land, especially after a number of generations, they were creolised. Furthermore, in this “third space”, identities could be negotiated and renegotiated, often through interaction with material culture (cf. Gosden 2004; DeMarrais 2004). Identity can be something that is chosen, and it is possible to have multiple identities. Those living in Hiberno-Scandinavian towns may well have adopted dual or even shifting identities. 68  This would seem to have been the case with people of Gaelic origin living in the Hiberno-Scandinavian towns. The results of the study by McEvoy et al on the link between supposed Norse derived Irish surnames and geographically specific Y- chromosomal traits imply ‘a very limited general Norse genetic legacy in Ireland dating to the Viking period’ (McEvoy et al 2006, 1292). Furthermore, they suggest that ‘the Norse genetic component of the distinctive hybrid Hiberno-Norse culture appears absent’, which they take as implying that perhaps the numbers of Norse settlers in these sites was small compared to the overall population, possibly restricted to a thin upper stratum, and that the bulk of the population of these sites ‘may simply have been culturally adapted indigenous Irish’ (McEvoy et al 2006, 1292). It also implies that the surnames may have been acquired in the bilingual environment through familiar names (McEvoy et al 2006, 1289). Studies such as these indicate the dangers of assigning biological ethnicity based on material culture. However, it should be borne in mind that they used a sample of 47 men representing 26 surnames associated with Scandinavian ancestry in Ireland. Perhaps at some point in the future a larger study will be done with a more statistically significant sized sample group. 69 Chapter 4. Anglo-Norman Colonialism in Ireland 4.1 Prelude: Invasion The exact points of entry are known from historical sources, the first was at Bannow, where a force of c.500 men landed in May 1169 under Robert FitzStephen (Expugnatio1.3 lines 4-5)20 and Maurice de Prendergast, (Song line 455), before going on to take Wexford. A second lay c.4km to the south-west at Baginbun, Co. Wexford, whose elevated position on a promontory overlooking a beach location, made it a fine staging point for further military campaigning, and has been associated with the 1170 campaign of Raymond le Gros (O’Conor 2003, 19; Expugnatio 1.13). It is possible that this was intended to be the main point of entry for the large force due to arrive under his cousin de Clare (O’Conor 2003, 18, 24ff). However, combined Gaelic and Ostman forces attacked the site and were greatly defeated (Song lines), enabling de Clare to sail directly to Waterford and take it in August 1170 (Song lines 1414-95, 1500-15; O’Conor 2003). 4.2 Expansion The ringwork must be regarded as the principal monument of the expansion phases of Anglo-Norman expansion in Ireland. The same is also true for Britain in the years after the Norman Conquest. For many years the established wisdom assigned this role to mottes, but the complexities of motte construction, as shown by Higham and Barker (2000) at Hen Domen, has rendered that monument unsuitable for the task of territorial acquisition. 20 I am refferring to the Expugnatio by book, section and line, as opposed to page numbers, and Song of Dermot And the Earl by line rather than page number, so as to render it easier for the reader to locate the references in the various editons 70  A ringwork castle consists of earthen banks with an external ditch, usually circular or sub-circular and usually palisaded (Sweetman 1999, 4; Barry 1987, 46; O’Keeffe 2000, 30). Internal buildings would largely have been timber built, although internal stone buildings may possibly have been used on occasion. Their identification in the landscape can be problematic, due to their morphological similarity to ringforts (Barry 1987, 45). In Ireland, they generally had more elaborate defences and were larger in form than the ubiquitous Irish ringfort, being 30-60m in diameter on average, (de Meulemeester and O’Conor 2007, 325), as opposed to an average diameter c30m. However, it should be noted that numerous examples of ringforts of this size exist in the archaeological record. The ASI has 109 ringworks listed on its database (www.archaeology.ie) for the 26 counties.21 A number have been excavated, such as Pollardstown, Co. Kildare (Fanning 1973-4), Clonard, Co. Meath (Sweetman 1978b), and in the case-studies, Béal Ború/Ballyvalley (O’Kelly 1962), Castletobin (Barry 1987, 48), Co. Kilkenny, and Ferrycarraig, Co. Wexford (Bennett 2004-5, 190). At the high status Anglo- Norman Castle at Trim, Co. Meath (Sweetman 1978a) uncovered evidence for a preceding ringwork. Similar evidence exists for preceding ringworks at similar sites in the case-studies, at Ferns (Sweetman 1979) Limerick, Wiggins (1994), Kilkenny, Carlow (Sweetman, 1999, 6) and Adare, Co. Limerick (Barry 1987, 49ff). The key feature of ringworks is that they could have been constructed relatively quickly, (O’Conor, 2002, 174) which would have been a very attractive feature to an expanding polity. This, combined with their defensibility, would have made them ideal campaign headquarters and residences for a newly established power. They had already been successfully used in the Norman subjugation of 21 The majority of these are found in Co. Tipperary, a bias probably due to research activities as opposed to being representative of their original distribution 71 England and Wales (cf. Platt 1978), and they would not be used in Ireland for the same purpose. Therefore, they are regarded in this project as being monuments of the expansion phase of the colonial process. 4.3 Anglo-Norman Colonial Expansion in the South-East The south-east was the Anglo-Norman point of entry into Ireland. Ringworks were built to lay a claim on territory granted to various Anglo-Norman lords involved in the initial activity in Ireland to establish manors, in order to repay their service. In order to provide the necessary incentive, these grants would have had to consist of land favourable to arable agriculture. Their siting represents the centres of the initial settlement, although some may have been replaced with mottes in the consolidation phase. As can be seen from figure 4.1 the majority of ringworks in the south-east are located in strategic locations in the landscape, with several located near to the main river routes. They are also in areas of relatively low local Gaelic secular settlement. Whether this is due to the later destruction of Gaelic settlement sites by arable agriculture (Barrett and Graham 1975; cf. Stout 1997) or the Anglo-Normans avoidance areas of dense Gaelic settlement is difficult to surmise. Another reason for this “avoidance” is that, as Stout (1997) has shown, ringfort builders preferred well drained moderately elevated locations. Such locations were favourable to pastoral agriculture. The Anglo-Norman economy was based on arable agriculture, to which large low lying areas of the south-east are suitable. This difference helps explain the largely complementary distribution. 72 Figure 4.1, early Anglo-Norman settlement activity and Gaelic secular settlement in the south-east Figure 4.2, Anglo-Norman ringworks (labelled) and campaign fortification with maximum pre-twelfth century ecclesiastic sites (including all pre-1700 churches in Kilkenny and Carlow) and known Hiberno-Scandinavian settlement 73  As can be seen from Figure 4.2, ringworks were sometimes located at high status ecclesiastical sites. Ringworks were located at Great Island, at Kilkenny and close to the cathedral site of Ferns. While Kilkenny had not yet acquired cathedral status, it was already an important monastic site, as evinced by its round tower, and it was also possibly the location of some form of high status secular settlement associated with the Osraige (Bradley 1990, 64-6; AFM1146; Song lines 1268-1391). It was located at the centre of a small fertile plain in the centre of the modern county, in a series of knolls at a fording point on the River Nore close to where it is joined by the Breagagh (Bradley 1990, 64). The strategic location, close to a nodal point in the native Irish network of power relations is significant. It follows a pattern adhered to in all zones of Norman and Anglo-Norman colonialism, and indeed in colonialism in general, (cf. Platt 1978; Colas 2007; Darwin 2007). The location of a ringwork at Ferns is also significant. It was a major centre of power in Ireland since at least the ninth century, a possible monastic town with significant economic, political and ideological resources (Ó Corráin 1972, 87). In the twelfth century it was both a diocesan centre and the caput of the Uí Chennselaig dynasty. Controlling the site would have been extremely important for the establishment of colonial hegemony. The ringwork may have been built there after the death of the colonists’ main ally, Diarmaid mac Murchadha in 1171 in order to secure the Anglo-Norman hold on the area against Irish claimants to his kingship. Another ringwork was located at Great Island, Co. Wexford, c.1.5km from Kilmokea, which probably had been the important ecclesiastic site Inis Daimle (cf. Charles-Edwards 2006, 150). It was another strategic location, lying close to where the confluence of the Nore and Barrow rivers join the Suir. The ringwork at Ferrycarraig/Newtown was on high ground overlooking the mouth of the Slaney, and 74 the ecclesiastic site at Saunderscourt lies close by to the north. The ringwork at Castletobin, Co. Kilkenny lies less than 1.5km from the ecclesiastical site at Tullamaine and c.3.5km from the ecclesiastical site at Greatwood. The ringwork at Carlow lies close to the junction of the Barrow and a tributary. The location of two ringworks in area of higher Gaelic settlement density in the north-east of the case- study possibly represents an attempt to subdue that area. 4.4 Anglo-Norman Colonial Expansion in the Mid-West Modern Co. Clare was a highly charged contact zone in the Anglo-Norman period, with the Gaelic Ua Briain dynasty and their Dál gCais ancestors having held sway in the region for several centuries. It can be identified as one of the locations where the wave of Anglo-Norman colonialism broke and began to recede in the fourteenth century. Although Limerick was taken by the “grey foreigners”22 in 1175, it does not seem to have been securely held until after 1200 (AI1175.6; AI1176.3, AI1176.6, AI1202.2). This means that Anglo-Norman colonialism began somewhat later there in the mid-west, which may be the reason for the identification of fewer ringworks there. Figures 4.3 and 4.4 illustrate the spatial relationships between these monuments of expansion and the existing secular and ecclesiastic settlement patterns. Ringworks are almost exclusively located in areas of little Gaelic secular settlement. This may be due to the fact that indigenous and colonial communities favoured environments conducive to different agricultural systems, or ringworks being built in areas of lesser threat. The exceptions to this pattern are the Clare sites at Ballyvalley and Ballycarroll, and the ringwork at Carrigafoyle Co. Kerry. Each of the three of 22 The term used is Gaill Glassa, surely a reference to the armour of the Anglo-Normans, and if so, then it illustrates the uniqueness of such armour in the eyes of Irish contemporaries, in turn indicating the military/technological advantage held by the colonists 75 these is in an area of relatively dense Gaelic settlement. The first directly overlies the site of Béal Ború, a high status Gaelic site associated with the Dál gCais dated to around the turn of the millennium (O’Kelly 1962, cf. Twohig 1978). The choice of this particular site seems to have been both militarily strategic and symbolic. It was located overlooking the first fording point below Loch Derg. It was also quite close to Killaloe and the related site of Cenn Corad, the symbolic caput Ua Briain dynasty prior to their taking residence in Limerick. Killaloe was the both an important ecclesiastical site, at the head of a diocese, and was possibly urbanised (cf. Bradley 1988, 64). The reuse of such a site can therefore be taken as a symbolic appropriation of the Ua Briain power. Figure 4.3, Anglo-Norman ringworks (labelled) and the pre-existing secular settlement pattern 76 Figure 4.4, ringworks and the potentially contemporary Hiberno-Scandinavian and ecclesiastic settlement, including the maximum undated church distribution The placement of a ringwork at Limerick, a pre-exiting nodal point in the network of power relations would have resulted in greater military, economic, political and ideological colonial power in the region. Located on an island at the head of the Shannon estuary, the site could control traffic from the sea to the river system. Like other Hiberno-Scandinavian towns, Limerick was part of a the network of trade routes in north-west Europe, and it probably supported large scale craft production and fishing, as at Dublin (cf. Wallace 2005; AI1105.8). It was also an ideological centre, including a cathedral, with previous ties to Canterbury, and a clergy to legitimise the hegemony of the new settlers. Politically, it had been the caput of the Ua Briain dynasty, and while they had fallen into relative decline, they continued to sit at the head of a kleptocratic regional taxation and political system. The northernmost possible monument of expansion in the case-study region is at Ballycarroll, Co. Clare. The interpretation of the site is problematic, however, as it 77 is difficult to discern whether the partially surviving earthworks there are those of a ringfort or a ringwork. If this site was a ringwork, it would represent the maximum extent of penetration into Gaelic territory in the region. The ringwork at Carrigafoyle is in an extremely strategic location on the southern shore of the Shannon estuary in modern Co. Kerry. It lies in a pocket of localised low density Gaelic secular settlement c.800m from the ecclesiastic site at Carrig Island. Ringworks in the mid-west case-study were largely located in low lying areas, possibly favourable to arable agriculture, as can be seen from the group of ringworks to the south of Limerick, they often were located in strategic locations, often close to water, and could on occasion be located in pre-existing centres of power, such as at Béal Ború and Limerick. Those in less strategic locations or in areas of low pre- existing settlement, such as the series south of Limerick probably represent the granting of land in areas favourable to arable agriculture. 4.5 Consolidation Mottes are the salient monument of Anglo-Norman consolidation, representative of a tightening colonial grip, at least in their primary phase of construction, as of course they went on to become centres of domination afterwards. Their morphology is such that it renders them visible in the landscape. They were large mounds of layers of earth and gravel. That they still stand with such steep slopes today in many parts of north-west Europe is a tribute to the engineering of their builders. They would have had substantial defences, usually of timber, with a tower and palisade located at the summit of the mound. In certain instances mottes may have been constructed by the filling in of a previous ringwork (Cathcart King 1988, 42), as the castle was elaborated in a period of post-expansion consolidation. This process may help 78 account for the low numbers of ringworks in Ireland compared to mottes, which stands at 423 mottes to 109 ringworks at present for the 26 counties (www.archaeology.ie). Mottes often had one or more adjacent defended enclosures, known as baileys, usually constructed in the same manner as a ringwork. These may be of various size and shape, depending on the needs and exact function of the motte within the community. It also indicates a hierarchy of protection at work, with the tower providing a castle within a castle for the lord and his family, differentiating them from those protected solely by the bailey, such as the lord’s retinue and possibly some of the settlers which may have begun to settle the land Mottes had multiple functions. From a military perspective, they were strong fortresses suitable for maintaining the positions gained in an expansion phase. Their morphology was such that they were the materialisation of the power of the new regional elite (cf. DeMarrais 2004; Elias 1983, 41ff). They would have been among the largest structures in Ireland at that time, with only ecclesiastic round towers taller than them, and no structures more massive. They also became economic nodes, functioning as manorial centres in the following phase of colonial domination. 4.6 Consolidation in the South-East As can be seen from figure 4.5, many more mottes than ringworks are found in the south-east case-study. They are to be found in areas of varying previous settlement density. They avoid some areas of dense Gaelic settlement such the north and west of Co. Wexford, possibly indicating continuity of Gaelic settlement patterns and a low colonial impact in these areas (cf. Barrett and Graham 1975). There was an avoidance of high elevations, although mottes are found in river valleys in areas of upland. This 79 would seem to point to their function as manorial centres associated with cereal production, as the corpus of historical evidence also indicates. Figure 4.5, mottes (labelled), ringworks and pre-existing secular settlement in the south-east 80  The motte at Castlesow and the ringwork at Toberfinnick 120m to the north are possibly associated, either temporally, with one replacing the other, or as part of a contemporary spatial hierarchy. This may also have been the case with the motte at Salville/Motabeg and the ringwork at Dunanore are located at either side of the Slaney. The latter site is arguably in a more strategic location, close to the confluence of the Slaney and a tributary. In Co. Kilkenny, the motte and bailey at Westcourt Demesne and the ringwork at Castletobin possibly share a similar association, lying 1.6km apart. Figure 4.6, mottes, ringworks and the pre-existing known ecclesiastical settlement pattern in the south- east, please refer back to figure 4.6 for motte names As can be seen from figure 4.6, there is at least some correspondence between the pre-existing ecclesiastical settlement pattern and motte distribution. O’Keeffe (2000, 16-17) draws attention to a similar pattern in the Lordship of Meath. This may be taken as indicating church collusion. The motte and possible early ecclesiastic site 81 at Coolbunnia, on Waterford Harbour lie c.240m apart, indicating either seizure of collusion, with the latter being most likely. A motte and bailey are located close to the important early ecclesiastical site at St Mullin’s on the River Barrow, and the decision to locate there may have been both a strategic and symbolic appropriation of social power resources. In north Co. Wexford, the possible early ecclesiastic site at Barnadown and the motte and Bailey at Loggan Lower lie c.770m apart, and the possible early ecclesiastical site at Glebe in Kilenor and the motte at Pallis Lower lie 1.6km apart. In the south of the county, the motte and ecclesiastical site at Duncormick lie less than 200m apart, and in the townland of Hooks, a motte and early ecclesiastic site lie c.50m apart. In western Wexford, a motte and possible early ecclesiastic site lie c.300m apart at Killegney. At Doonooney, a motte and ecclesiastic site are located c.200m apart. To the north of Wexford Harbour, the ecclesiastic site of Beggerin Island and the motte at Ballinamorragh are located c.1.4km apart. At Grangefertagh, in north-west Co. Kilkenny, itself an area of dense Gaelic settlement, a motte and ecclesiastic site lie c.700m apart. At Killamerry in the south of the county, a motte and ecclesiastic site are roughly 100m apart. At Lismateige, a motte and ecclesiastical site are located c.200m apart, with a small river between them, but in the same townland. In Co. Waterford, the motte at Pembrokestown and the ecclesiastic site at Lougdeheen are located less than 1km apart. From this information, it is possible to say that a spatial link exists between some Anglo-Norman mottes and ecclesiastic sites in the case study region. These latter sites would have been nodal points in the ideological power network, as well as possibly being arable agricultural economic production centres. Therefore, it would have been strategically astute for colonists to encastellate in close proximity to such 82 centres, so as to consolidate their power resources. It would seem as though some mottes replaced ringworks, either directly by being constructed over them (cf. Cathcart King 1988, 42), or figuratively. Motte and ringwork builders in the south- east avoided higher elevations, which while most likely for agricultural reasons, also meant that they made no major incursions into areas of Gaelic settlement. 4.7 Consolidation in the Mid-West Colonial settlement activity in Clare seems mainly confined to the area east of the Fergus, where the mottes at Clonmoney West and Bunratty East are located. Both are located close to the coast in low lying areas characterised by limited local Gaelic secular settlement evidence. As can be seen from figure 4.7, the number of unclassified ringforts between this area and Limerick is significant, as it may indicate site destruction due to arable agriculture, or perhaps an earlier abandonment of Gaelic secular sites in this area, perhaps due to possibly having been part of Hiberno- Scandinavian Limerick’s hinterland. South of the Shannon, the motte and bailey at Shanid Upper is located in an area of dense Gaelic settlement. The motte at Burgesbeg, Co. Tipperary is in a strategic location below the Arra Mountains, overlooking the route of the Slighe Dála, one the main medieval land routes through Ireland. As it stands there seems to be no correlation between motte distribution and pre-Anglo-Norman ecclesiastic sites in the mid-west case-study region. Commenting on Connacht, O’Keeffe notes that the distribution map of mottes is ‘almost a freeze-frame graphic of colonial diffusion up to 1200’ and notes that we know that the Anglo-Normans spread further west, but that mottes do not seem to be a feature of this (2000, 29-30). Ringworks may have been utilised instead (2000, 30). This seems to also hold for the mid-west case-study, as more ringworks than mottes 83 have been identified there, as can be seen from figure 4.7. It is likely that the reason that fewer mottes appear west of the Shannon because the labour was not available for motte construction, meaning that fewer mottes were built, and that fewer ringworks were replaced with mottes compared to elsewhere. It would also seem as though there is little or no correlation between the location of mottes and early ecclesiastic sites, cf. figure 4.8. Therefore, it can be said that consolidation was limited in the mid-west, as had expansion before it. Figure 4.7, ringworks, mottes and pre-existing Gaelic secular settlement in the mid-west 84 Figure 4.8, ringworks, mottes and pre-existing ecclesiastical settlement in the mid-west, including entire pre-1700 church dataset 4.8 Domination Anglo-Norman masonry castles must be largely, but not exclusively, assigned to the phase of domination, when the area had already been won and consolidated. They may theoretically have played a part in colonial expansion phases, such as in the Levant, but the man-hours necessary to construct this form would have made them very inefficient for the purpose. As noted earlier, a number of them seem to be sited on former ringwork castles, such as at Kilkenny, Carlow, Ferns, Adare and Limerick, indicating both continuity of settlement, and the strategic nature of settlement in the expansion phase. There are instances of stone castles being used in consolidation activity, but their expense would have made them inefficient for this purpose, and in this phase there would still have been the danger of such powerful military technology falling into native hands. 85  Both mottes and masonry castles indicate that time could be safely spent on construction, which is indicative of less time being assigned to primary military activity. However, a hierarchy of settlement is also a feature of the domination sub- phase of colonialism, and so nucleated settlements such as towns and villages, dispersed colonial settlement, Continental religious houses and the continued use of monuments constructed as part of the previous phases of expansion and consolidation must all be considered as part of the domination phase. In addition to their roles as centres for the exercise of social power, as centres for redistribution, proto-industrial and legal activity, these castles of the domination phase were also highly metonymic, representing the new political system in the minds of those viewing the castle. They were the materialisation of an entire social system. They provide prime examples of the use of conspicuous monumentality as means of social control (cf. Elias 1983; Veblen 1925). While their form may have been due to status anxiety (cf. de Botton 2004), peer polity competition (cf. Renfrew 1986) or the edifice complex associated with absolute power, their primary symbolic function was the portrayal of a message of dominance. Monument Form Sub-Process Example Royal Castle Masonry Domination Kilkenny Castle, Limerick Castle Seigniorial Castle Masonry or Domination, Enniscorthy, Bunratty, Motte Consolidation Ferns Manorial Castle Motte or Consolidation, Glascarraig, Ringwork Expansion, Domination Co. Wexford Moated Site Domination Ballygub New, Co. Kilkenny Undefended House Domination Clonmore, Co. Kilkenny Table 4.1, hierarchy of settlement monuments, and the processes associated with them. Where associated with more than one sub-process, they have been listed in the order of their level of association. 86 Settlement Position in Network Example City Regional Hub Waterford, Kilkenny Town Sub-regional Hub New Ross Borough Rural Hub Bannow, Bunratty Village Rural Hub Newtown Jerpoint Dispersed Defended Settlement Rural node Moated sites Dispersed Rural node Undefended rural house Table 4.2, hierarchy of settlement forms, each level can serve as a hub for the nodal points of the level below it. This is only an approximation, for the purpose of explanation, rather than a positivistic attempt to apply cybernetics to the Anglo-Norman settlement pattern. Anglo-Norman masonry castles in Ireland sat at the head of a local and regional central place hierarchy. Tables 4.1 and 4.2 place them in their hierarchical context, along with the other monuments associated with the domination phase. For military, economic and administrative purposes, it stands to reason that castles would adhere to some form of spatial patterning, especially in a colonial setting. Their siting was influenced by a number of factors. The first would be actual location within the local polity. The castle would have to be in a location best suited to exploit the economic resources of the area. It would also have to be in a defensible location, for military reasons. It would have to be visible in order to maximise its cognitive/symbolic potential. The siting of the centre would also have to take into account the location of other such centres, both for military and economic reasons. The extent of nucleation is difficult to surmise. McNeill, commenting on the spatial organisation of colonialism in Ulster notes that the manorial centres of the earldom of Ulster were places where tenants went to periodically as opposed to being permanent centres of population (1980, 84-5). However, it could be argued that Ulster is an exception, the form colonialism took there was somewhat different than 87 other parts of the country, being largely characterised by a replacement of the existing elite with a new elite without an influx of the lower orders of a colonial society (McNeill 1980; O’Conor 1998). Graham (1975, 224-8), in his study of Anglo- Norman settlement in the lordship of Meath identifies 98 possible former villages, each located where castles and churches juxtaposed in the landscape. However, O’Keeffe (2000, 71) notes that there is little archaeological evidence for this at the majority of these sites as yet. Colfer (1987) puts forward a model of settlement in Wexford based largely on Graham’s methodology, but it would seem that there is little evidence as it stands for those sites having been nucleated settlements (M. Moore, ASI, 2010 pers. comm.). The conferral of borough status usually indicates the incorporation of a town. However, in Ireland it seems to have also occurred with “rural boroughs”, where borough status was conferred on villages or even virtually empty space for political or economic purposes, or to incentivise settlement (cf. O’Conor 1998, 43ff). 88 4.9 Colonial Domination in the South-East Figure 4.9, Anglo-Norman ringworks, mottes and masonry castles (labelled) in the south-east Considerable evidence exists for a colonial settlement hierarchy in operation in the south-east case-study in the phase of colonial domination. It is significant that not all mottes and ringworks were replaced with masonry castles in the domination phase, as figure 4.9 illustrates. This can be taken as indicative of a settlement hierarchy, even without recourse to documentary evidence. An indication of continuity from the phases of expansion and consolidation to the phase of colonial domination can be ascertained from figure 4.9. A number of masonry castles are located on or close to the sites of pre-existing castles. The excavations at Kilkenny, Carlow, and Ferns Castles, all nodal points in the local network of power relations, have provided evidence for direct continuity from ringwork to masonry castle. From the information provided by figure 4.9, Great Island can be added to this group. All were located at important points on major 89 rivers, bar Ferns, and only Carlow did not have some form of high status secular or monastic settlement prior to the arrival of the Anglo-Normans. There is also evidence on the map for continuity from motte to masonry castle. The masonry castle at Clonmore, Co. Wexford is located c.800m from the motte at Minvaud Upper. Whether or not one replaced the other is difficult to ascertain, but they were certainly either synchronically or diachronically related. The masonry castle at Tullowphelim Co. Carlow seems to have replaced the motte there. The masonry castle at Enniscorthy, Co. Wexford is located upstream from the motte at Salville or Motabeg and the ringwork at Dunanore located at either side of the Slaney. According to Colfer (1987, 76), there may have been a motte at Enniscorthy prior to the construction of the masonry castle constructed by the Prendergasts in the 1220s. Nucleated Settlement Figure 4.10, Anglo-Norman castles and possible urban centres in the south-east case-study region 90  The Hiberno-Scandinavian towns at Waterford and Wexford both had masonry castles constructed within their confines. This was strategic from both a military and symbolic perspective, as it was militarily important to hold such sites, and it was symbolically important to be seen holding them. As can be seen from figure 4.11, a number of new urban settlements came into being in the Anglo-Norman colonial period. The majority are associated with castles, which might precede the foundation of the town, or post-date it. An exception seems to be New Ross, although it may be that a castle there has not been identified or classified. Figure 4.11, spatial relationship between known early ecclesiastic sites and Anglo-Norman towns and boroughs, with non-diagnostic pre-1700 church sites left out As can be seen from figure 4.11, there is very little association between pre- Anglo-Norman ecclesiastic settlement and Anglo-Norman towns and boroughs. Again, New Ross is an exception. The only other towns with a relationship with an 91 earlier ecclesiastic site are the aforementioned Kilkenny and Ferns, and the Hiberno- Scandinavian town at Wexford. Future research may uncover further relationships. Parishes and manors were often coterminous, in much the same manner as parishes often represent secular political boundaries in Gaelic Ireland in the high and late medieval (cf. MacCotter 2008; Watt 1972). Therefore, by mapping known high medieval parish churches, it is possible to gain an approximation of the overall density of settlement, as the higher the number of parish churches in a region, the more intensely that region is settled. Also, manors and their subdivisions were often based on the ability to pay the so-called Knight’s Fees, which were in lieu of military service to the feudal overlord (cf. Brooks 1950). These fees underpinned the entire feudal system in the greater Angevin empire to which Anglo-Norman Ireland belonged (Bloch 1961; Critchley 1978). The more intensely cultivated and productive land was, the more profit could be extracted from it, with the likely result of it becoming more densely populated and subdivided. So by mapping the distribution of high medieval parish churches it is possible to gain an approximation of the intensity of land use and the related settlement pattern. As can be seen from figure 4.12, the south and east of Wexford seems to have been quite densely settled, the south in particular. However, castles seem to be relatively low in number in the area compared with other parts of the case-study region. The area, corresponding to the baronies of Forth and Bargy seems to have remained one of the most anglicised parts of the island until relatively modern times, both in custom and language (Colfer 1987; Roche 1987; Estyn Evans 1957). A separate dialect of English, Yola, survived there until the nineteenth century. This had been an area of low pre-colonial Gaelic settlement, so perhaps less pacification was necessary. It may be then that manorial centres took a less well-defended form in 92 the region, or it may be that several ringworks remain unidentified in the quagmire of generic unclassified enclosures located in the region, as seen in figure 3.6. The construction of a large number of tower houses in the area in the later medieval period may obscure any earlier encastellation there. Figure 4.12, known high medieval parish churches and Anglo-Norman castles in the south east The low-lying nature of the land would not have been seen as ideal by pastoralists (cf. Stout 1997). Perhaps also victory of Raymond le Gros over all of the potentially hostile parties prior to the invasion proper negated the need for extensive encastellation in the area. It is also interesting to note that the proposed hinterlands of Hiberno-Scandinavian Wexford and Waterford seem to be located in dense zones of Anglo-Norman settlement, possibly an indication of the swift incorporation of Hiberno-Scandinavian settlers into the new colonial polity. It is also interesting to note the correspondence between low densities of parish churches and the area of low density of encastellation in north-east Wexford, 93 which may have continued to be held by Gaelic groups, and which had Wexford’s highest density of Gaelic settlement sites. However, this may also just indicate that this region was ill suited to arable agriculture and better suited to pastoralism (cf. Stout 1997). Figure 4.13, rural nucleated settlement and known Anglo-Norman encastellation in the south-east Rural nucleated settlement A number of deserted medieval settlements are known in Ireland, a large proportion of which are in the south-east case-study region (Buchanan 1970; Glasscock 1970; Barry 1977, 1987, 2000; O’Conor 1998; O’Keeffe 2000. As noted in chapter 2, their true distribution is difficult to assess, as many may remain undiscovered with little or no diagnostic surface features, and none have been identified as yet in Co. Wexford, despite the obvious density of settlement there. From what evidence exists, they have no direct associations with masonry castles, but a number do seem to have been associated with mottes, such as at Coolbunnia, Co. Waterford and Fiddown, Co. 94 Figure 4.14, deserted medieval settlements and total ecclesiastic landscape. This is a necessarily dense and complicated map, whose function is to show the variety of possible associations. The category “high-late medieval church” refers to non-parochial churches from this period. 95 Kilkenny. The ringwork at Rathealy, Co. Kilkenny is the only directly associated ringwork in the region. The only directly associated pre-Norman ecclesiastic sites are Killamery, Castlestown, and Grangefertagh, Co. Kilkenny and Old Leighlin, Co. Carlow. However, as can be seen from figure 4.14, a number of high medieval parish churches and undated churches are directly associated, a similar pattern to that in England at the time (cf. Clarke 1984). Religious houses Appearance of parishes was just one manifestation of the major reforms in the Church in the twelfth century, which saw the island divided into several territorial dioceses and arch dioceses, largely representative of the contemporary configurations of secular and ecclesiastic power (Watt 1972, 1-26). It is at this point that the decline in fortunes of many earlier ecclesiastic institutions accelerated, as many sites, such as Clonmacnoise, lost considerable power due to these reforms. The reforms saw the introduction of a number of Continental religious orders, beginning with the Cistercian foundation at Mellifont in 1157 (Watt 1972, 43; Duffy 1997, 55), and accelerating in the Anglo-Norman period. This must be regarded as part of the colonial process, as it drew Ireland further into the European scene (cf. Martin 1987b, 50-1). Ab initio these new houses had politically powerful patrons, a pattern which would continue into the Anglo-Norman colonial period. This patronage associated the secular elite with the contemporary forces of modernity, and also with a pan- European aristocracy. By linking themselves with such an important ideological source of power, these patrons were significantly bolstering their power resources. The Church was the principal ideology-forming institution in medieval Europe. It 96 was responsible more than any other social institution for the formation of the symbolic universe of explanation within which social action took place and which shaped experience and legitimated the social order (cf. Berger and Luckmann 1966; Berger 1967). Therefore, the Church must be regarded as the principal agent of reification in medieval Ireland, and indeed medieval Europe. The term reification, in the way used by Lukaćs (1968, 83ff), might best be regarded as “thingification”, in that human qualities come to be regarded as things (Craib 1984, 16). Social relations may become reified, in that they can become regarded as being as natural and unchangeable as the laws of nature. This can mean that an existing social hierarchy can be regarded as being as natural and unchangeable as the sun crossing the sky on a daily basis (cf. Lukaćs 1968; Craib 1984). Therefore, any element of society seeking to establish and maintain hegemony would have to do so in collusion with the Church, in order to make the existence of such a hierarchy, and the position of individuals and groups of individuals within it seem as being “common sense” in the minds of each individual (cf. Gramsci 1971). Ergo, the Church, and the new religious orders conducting pastoral activity in the community could help legitimate colonial activity, or indeed legitimate resistance to it. The Normans and the groups subsequently related to them managed to have themselves associated with religious reform, meaning the Church was in many ways favourable to their expansionary activity. There also was another key connection, that of a shared habitus (cf. Bourdieu 1977, 1984) between the new secular and ecclesiastical elite. They often came from the same families or social circles, and thus had a similar world view, with one group favourable to the other, much the same as the relationship between politicians and bankers in modern society, where a shared habitus makes each group favourable to the other (cf. Mills 1956). 97  Some religious orders seem to have been more connected with the colonists than others, such as the Dominicans, but few absolutely so, bar rare orders such as the Order of Tiron. Some orders seem to have had a preference for urban locations, others for rural locations, as can be surmised from figures 4.15 to 4.19, and tables 4.3 and 4.4. Religious houses seem to avoid areas of high early ecclesiastic settlement, cf. figure 4.15, which indicates that the old and new forms co-existed for a period of time. Although they did contribute to their social obsolescence there seems to be little physical replacement of older ecclesiastic sites in figure 4.16, and Swan (1983) notes that rural religious houses seem to be on green-field. As can be seen from figure 4.17, religious houses seem to avoid the areas associated with mid and high density Gaelic secular settlement, although this is probably connected more with the mode of agriculture practiced at rural religious houses than any potential Gaelic hostility. Evidence for the mode of agriculture has been provided by excavations such as those at Kells, Co. Kilkenny (Clyne 2007). The plant remains assemblage at Kells point towards a wheat based economy, a pattern repeated at other similar sites, and which was a departure from the predominantly barley based package practiced by earlier ecclesiastic sites (Monk 2007, 483). Roche (1987, 108) notes that religious houses seem to be absent from the area of densest Anglo-Norman settlement in southern Wexford, as can be seen from figures 4.14, 4.18, and 4.19. At present this largely defies explanation, but it may be related to the extent of infeudation there, in that there was no land left to grant. A number of religious houses are in close association to non-urban mottes, such as at Glascarraig, Co. Wexford, Aghade/Castlegrace, Co. Carlow, the unclassified house at Loughill and the motte at Castlemarket, and the Cistercians at Barrowmount and the motte at Powerstown East, Co. Kilkenny. A number of other 98 religious houses and mottes are in association in an urban setting. Table 4.3 describes the urban/rural nature of each classified religious house in the case-study, and notes its founder, which is an indication of the colonial or non-colonial nature of each site where that information is available. From this table it can be surmised that c.70% of the classified religious houses are urban, and that 10 had colonial patrons, and 5 had Gaelic benefactors, with only four houses dating to before the Anglo-Norman period. As can be seen from figure 4.20, they had little or no association with moated sites. Location Order Date Patron Urban/Rural Carlow AGHADE Augustinian nuns 1151 Diarmaid mac Rural Murchadha OLDLEIGHLIN Augustinian Friars 1163- Laurence O’Toole Cathedral site LISNEVAGH, Augustinian Friars - - Rural TOBINSTOWN possible TULLOWBEG Augustinian Friars 1314 Simon Humbard and Urban High Talun LEIGHLINBRIDGE Carmelite Friars 1272 Earliest in Ireland – G Urban &H KILLERRIG Knights Hospitallers Pre-1212 Gilbert de Borand Rural Wexford ABBEYDOWN Augustinian Canons - - Rural FERNS DEMESNE Augustinian Canons 1160 Diarmaid mac Urban Murchadha SAINTJOHNS Order of St Victor, 1230 Gerald de Prendergast Urban, near then Augustinian Enniscorthy Canons possible CLONMINES Augustinian Friars c. 1317 Kavanghs Urban? NEW ROSS Augustinian Friars 1320 - Urban TAGHMON Augustinian, Of 12th, on - Urban Arrouaise Nuns earlier nunnery DUNBRODY Cistercian Monks 1171-5 Harvey De Rural Montmorency, Builwas and Dublin connections too 99 TINTERN Cistercian Monks 1200 William Marshal Rural ROSBERCON Dominican Friars 1267 Graces or Walshes Urban ENNISCORTHY Franciscan Friars 1460 Donald Kavanagh Urban TOWNPARKS (ST. Franciscan Friars 1230 - Urban PETER'S PAR.) NEW ROSS Fratres Cruciferi c. 1195 William Marshall Urban (Franciscans) (by 1295) GLASCARRIG Monks Of The Order 1193 Cantetans or Condons, Rural NORTH Of Tiron daughterhouse of St Dogmells' Pembrokeshire Waterford WATERFORD CITY Augustinian Canons before King John? Urban 1200 WATERFORD CITY Benedictine Monks Poss 1185 King John? United w Urban Bath 1204 WATERFORD CITY Dominican Friars 1230 - Urban WATERFORD CITY Franciscan Friars 1245 - Urban Kilkenny GARDENS (ST. Augustinian - - Urban JOHN'S PAR.) Canons INISTIOGE Augustinian - - Urban Canons RATHDUFF Augustinian 1193, or Geoffrey FitzRobert, Urban (MADDEN) or Kells Canons 1204-06 motherhouse at Bodmin CALLAN NORTH Augustinian Friars Urban KNOCKTOPHERAB Carmelite Friars Urban BEY BARROWMOUNT Cistercian Monks Rural GRAIGUENAMANA Cistercian Monks Rural GH JERPOINTABBEY Cistercian Monks 1160? Nucleated GARDENS (ST. Dominican Friars 1225 William Marshall the Urban CANICE PAR.) Younger ST. MARY'S Franciscan Friars 1232-40 Urban PARISH Table 4.3, classified religious houses in south-east case-study, listing the townland, order, date if known, founder and level of rurality. Compiled from the information in Watt (1972), Moore (1996, 1999), Clyne (2007) and Brindley & Kilfeather (1993). 100 Figure 4.15, geographical distribution of religious houses in the south-east by order, although the clustering of some sites masks the distribution, for full distribution, see table 4.3. Figure 4.16, religious houses and known pre-Anglo-Norman ecclesiastic settlement in the south-east 101 Figure 4.17, religious house and the known Gaelic settlement pattern in the south-east Figure 4.18, religious houses and nucleated settlement in the south-east 102 Figure 4.19, religious houses and Anglo-Norman castles in the south-east Figure 4.20, moated sites, deserted medieval settlements and religious houses 103 Moated sites Moated sites belong to the mature part of the domination phase. The sites are usually rectilinear, characterised by wide, usually flat-bottomed, sometimes water-filled, ditches around a central area (Barry 1977; 1987). O’Keeffe calculates that it could take three men a year’s work to construct one (2000, 73), which may indicate a system of mutual aid amongst colonists. They were part of a process of cultivation of comparably peripheral land, probably dating to between 1225-1325, after the manors had already been laid out and granted (Empey 1982-3, Barry 1977). Their low density in the heavily infeudated areas of east Waterford and south Wexford supports this, cf. figure 4.21. They are possibly representative of the colonial regime’s growing monopolisation of the legitimate use of violence, negating the need for settlement to cluster around castles. However, they may also have been symptomatic of political instability in instances where the moat postdates the buildings within it (Barry 1977, 96-7; O’Keeffe 2000, 75). Intriguingly, O’Keeffe points out that existing agricultural activity had long been producing a surplus, and that perhaps this secondary phase was related to a group of entrepreneurs aiming to profit from the high prices in England and France brought about by extensive military activity, although this military activity would be quite late in the chronology of moated sites (O’Keeffe 2000, 77-80, cf. Lydon 1987b, 197ff). As can be seen from figure 4.21 there is little direct association between moated sites and pre-existing Anglo-Norman castles. In some areas, such as central Wexford, a number of them seem to loosely cluster around castle sites however, possibly part of a localised settlement hierarchy. Their densest distribution runs south-west to north east through the centre of Co. Wexford, possibly an indication of 104 the colonists attempting to expand their zone of total control. Higher elevations are avoided, which seems to indicate a predominantly arable based agricultural package practiced by the inhabitants of the sites. The fact that densest distributions are in areas of low Gaelic secular settlement further supports this, cf. figure 4.22. There does not seem to be any direct spatial association between the distributions of moated sites and early ecclesiastic sites, cf. figure 4.23, or with later rural religious houses, figure 4.20, despite favouring similar agricultural land. Figure 4.21, Anglo-Norman moated sites, towns and castles in the south-east. Due to the number of moated sites in the south-east, this and the subsequent maps will necessarily be rather dense, but not so dense as to preclude the identification of spatial relationships 105 Figure 4.22, moated sites with known Gaelic secular settlement in the south-east Figure 4.23, moated sites with maximum pre-Anglo-Norman ecclesiastic sites in the south-east 106  The south-east saw a period of extensive colonial domination between the late 1170s and the mid-fourteenth century. It was characterised by a hierarchy of colonial settlement, centred on castles and urban centres. These sites were often located on earlier sites involved with the processes of expansion and consolidation. That is not to say that the mottes and ringworks involved in the earlier phases became obsolete, rather they were incorporated into the settlement hierarchy when not replaced, and they themselves functioned as local centres of domination. The land became substantially infeudated quite quickly, and each of the castles would have served as a manorial centre. Moated sites, dispersed houses and deserted settlements are all indicators of extensive colonial activity in the area, and each would have had its place in the settlement hierarchy, just as their occupants had their place in a reified social hierarchy, legitimised by the Church. The period saw the appearance of a network of parishes, usually coextensive to manorial divisions, each with its own church, providing material evidence for the extent and density of Anglo-Norman settlement in the region. The period also saw the appearance of a number of largely urban Continental religious houses in the region, indicative of the relationship between the first two orders of medieval society, those who pray and those who fight. 4.10 Colonial Domination in the Mid-West It would seem as though Anglo-Norman influence in Co. Clare was confined east of the Fergus Estuary, and centred on a number of royal land grants. The stone castle at Bunratty would have replaced the motte, which in turn may have replaced an initial ringwork. The stone castles at Adare and Limerick replaced ringwork castles, as noted earlier. The stone castle at Quin was begun in 1280 (Hodkinson (2004, 56), perhaps in response to the loss of caislen Clair Atha Da Charad in 1270. This latter 107 castle has not been precisely located, but was most likely sited at the site of the later tower house at Clarecastle, or perhaps at the later Ua Briain caput at Clonroad (cf. Ó Dálaigh 1989, 40; Hodkinson 2004, 57; AC1270.9 AHC1405.1). Figure 4.25, Anglo-Norman castles, urban and rural nucleated settlement in the mid-west Urban and Rural Nucleated Settlement The Hiberno-Scandinavian town at Limerick continues in importance in the Anglo- Norman period. A number of speculative land grants were made in the greater Limerick area in the late twelfth century. It would seem as though the area to the immediate north of Limerick was intensely settled, as there is evidence for three nucleated rural settlements there, probably indicative of a local settlement hierarchy. As mentioned earlier, Killaloe also probably witnessed a degree of pre-Anglo-Norman urbanisation, and it and Bunratty would seem to be part of a more regional hierarchy centred on Limerick. 108  Bunratty formed the centre of a borough, planted with “plebeian English”23 around a stone castle built by de Clare (CT1277), which by 1287 had 226 burgages and a watermill at Bunratty, in an area of land, Tradaree,24 granted to various Anglo- Norman families,25 and substantially under cultivation (Sweetman 1876-86, Vol. 1, no. 2920; O’Brien 1999, 34; Murphy 1980, 19). The deserted medieval settlement at Feenish Island in the Fergus Estuary would seem to be connected to this settlement. Textual evidence also exists for a vill at “Clarin”, with a weekly market and yearly fair (Sweetman 1876-86, Vol.2, no. 155). Ó Dálaigh asserts that this may have been at Clarecastle or the modern townland of Claureen in Ennis (1989, 40). It may also have been the deserted settlement at Clonroad. The settlement at Quin may have been Anglo-Norman, based around the castle. However, the castle was most likely in Anglo-Norman hands for 30-40 years, and a Franciscan friary replaced it after the colony fell into decline, and the settlement may actually have been associated with this phase of activity, as at Ennis. Parishes The distribution of medieval parishes in Co. Clare may not be of the same use as that of Co. Wexford. Many of the parishes were in Gaelic areas, and corresponded to Gaelic land divisions. Furthermore, I was unable to separate the pre-1700 churches dataset for Clare or the parts of Limerick and Kerry included in the project. Religious Houses Only the religious houses in Co. Clare have been considered, Clare forms the centre of the case study. Cistercian Corcomroe Abbey (McInerney 1979, 4; McMahon 1997, 24), Augustinian Clare and Killone Abbey (McMahon 1993, 22) were patronised by 23 O’Grady’s translation, the term used could also translate as “common grey foreigners”. 24 There are many variant spellings in the sources, such as Tradry and Tradery. 25 To Arnold Keting in 1199 (Murphy 1999, 34), followed by de Muscegros, an Anglo-Norman knight in 1248, who built fortifications at Tradry and Ocormack (Sweetman 1876-86, Vol. 1, no. 2920), in turn followed by the de Clare family in 1276. 109 the O Brian family. Augustinian Kilshanny is deep in Gaelic territory in the west of the county. The deserted settlement at Corcomroe must therefore be related to the abbey rather than to any colonial activity. The abbey Inchicronan lies close to the possible ringwork and hall house at Ballycarroll, and the abbey on Canon Island is located across the Fergus Estuary from the main zone of Anglo-Norman influence. The unclassified houses at Cratloe and Illaunmore may well be connected to Anglo- Norman activity. Location Order Date Patron Corcomroe Cistercian 1190s Ua Brian Clare Augustinian 1189 Ua Brian Killone Augustinian nuns Ua Brian Kilshanny Augustinian 1194? Clare Abbey Ennis Franciscan 1240 or 1280s Ua Brian Quin Franciscan Ua Brian Table 4.4, Classified Religious Houses in Co. Clare, complied from Gwynn and Hadcock (1970), Watt (1972), McMahon (1993, 1997). Figure 4.26, Religious Houses and nucleated settlement in Clare, with Anglo-Norman encastellation 110 Moated Sites As can be seen from figure 4.27, a low number of moated sites were built in Co. Clare, with the highest concentration in the mid-west being along the south of the Shannon Estuary, in an area associated with high density Gaelic secular settlement (figure 3.3). This concentration can be seen as indicating a higher Anglo-Norman presence in the area in the late thirteenth and early fourteenth centuries (cf. Empey 1982-3; Barry 1977). However, most of the activity in Clare dates to the same period, which may indicate that moated sites were monuments of expansion and consolidation there. In contrast with the south-east, all of the moated sites in the mid-west seem to be in areas of high Gaelic secular settlement, possibly due to a forced appropriation of land from an incumbent Gaelic population. Also of interest is the fact that there seems to be no association between moated sites and pre-existing Anglo-Norman settlement. Figure 4.27, moated sites and Anglo-Norman castles in the mid-west 111  The low numbers of moated sites in proximity to Limerick may indicate that land grants in this area may have been made prior to the spread of moated sites into the area. It also seems as though it corresponds to the possible hinterland for the town in the Hiberno-Scandinavian period, possibly indicating a continuity of Hiberno- Scandinavian settlement (cf. Bradley 1988). The three moated sites in west Clare are interesting. The southernmost, at Colmanstown on the Shannon Estuary would seem to be part of the pattern on the other shore. It may be an indicator of organic colonial settlement, where a family or group moved beyond the frontier of colonial expansion, similar to Israeli settlers in Palestine in recent years. It may also be an indicator of acculturation, in this case the native adoption of a colonial settlement form. The moated sites at Castlepark and Finnor More are probably example of the latter process, a phenomenon also noted at Cloonfree, Co. Roscommon by O’Conor (1998, 87). It is also of interest that these two moated sites are located close to ecclesiastic sites. The remaining moated site in Co. Clare is located in the area most associated with Anglo-Norman activity. 112 Figure 4.28, moated sites and Gaelic secular settlement in the mid-west Figure 4.29, moated sites and minimum early ecclesiastic sites in the mid-west It can be seen from the evidence provided for the mid-west that Anglo- Norman activity was much more limited there than in the south-east. Only in the area south of the Shannon and in south-east Clare can it be said to have developed beyond 113 the phases of expansion and consolidation. Even the appearance of moated sites may be representative of expansion rather than domination. For Co. Clare, there seems to have been no colonial patronage of Continental religious houses. 4.11 Decline By the fifteenth century, direct control by the London government had been reduced to two areas, the so-called Pale, which was basically Dublin and its extended hinterland, and the southernmost baronies of Wexford, Forth and Bargy (Duffy 1997; Roche 1987, 109-110). However, this does not entirely indicate a complete loss of power by colonists in the rest of Ireland, rather that the London government lost control over them. The retraction of the colony was due to the convergence of a number of causal factors, environmental, social and political. It provides an important case-study for studying the declines of colonies in other locations in space and time. Firstly, Ireland must be regarded as an example of a ‘half-conquered land’, where ‘beyond the sometimes precariously held colonial towns and fiefs lay native populations that were not subjected’ (Bartlett 1993, 301). Other examples of this would be the German Lithuanian settlements and the Crusades (Bartlett 1993, 301). The Anglo-Norman colonial endeavour in Ireland effectively ran out of people and could expand no further, and could not fully consolidate what it held, especially in contact zone areas outside of the colonial heartland of the east and south-east. As can be seen from the cartographic evidence provided, even in the densely colonised south-east there were areas where Gaelic settlement possibly continued, probably due to its suitability the pastoral agriculture practiced by such groups. This certainly provides evidence for a half-conquered land. It would seem as though the 114 previous Gaelic elite began to regroup from the late thirteenth century, at the beginning of what historians term the Gaelic Revival (cf. Lydon 1972; Lydon 1987b; Watt 1987a; Watt 1987b; Duffy 1997, 134ff). Many of the principal Gaelic septs had never been displaced or acculturated by the arrival of the Anglo-Normans, even in areas close to the main zone of activity (cf. Empey 1990, 89, Bartlett 1993, 301). Empey (1990, 89) notes that the position of the Anglo-Norman lords in many areas was already precarious before the close of the thirteenth century. He draws attention to the various impositions of levies and protection payments being made to the Gaelic Irish in the later thirteenth century in Kilkenny. Roche draws attention to similar payments, known as the “black rent” exacted by Gaelic groups in modern Co. Wexford from areas such as Forth and Bargy (1987, 109-110). Another major causal factor was the loss of control by the London government over colonial magnates in Ireland (cf. Watt1987a; 1987b). By the thirteenth century, many of the large land grants in Ireland were held by absentee lords, including the king, which further destabilised colonial authority in Ireland. Many of these grants were taken up by groups such as the Butlers, further boosting their authority. Elias (2000) outlines a process he terms feudalization, where the granting of territory by a ruler for services rendered aiding the maintenance or expansion of that ruler’s territory leads to a devolution of part of that ruler’s power, a loss of authority and an eventual threat to their position. This can be seen to have occurred in a number of historical cases, such as England under Stephen or John, or in this case in Anglo- Norman Ireland. The rise of the Butlers provides an excellent example of this process, and one which partly occurs within the confines of the south-east case-study. 115  A great deal of intermarriage took place between these magnatial families (Watt 1987b, 353), and with Gaelic elite families such as the Ua Briain (Watt 1987b, 353-4; Lydon 1972). However, such marriage ties did not impede their competing for position with each in a similar fashion to pre-Anglo-Norman Gaelic society. Intermarriage brings the analysis to two important processes at work in colonial Ireland, acculturation and creolisation. Creolisation is the maintenance, and often exaggeration, of cultural traits of the colonial point of origin in a colonial setting, with later cultural developments in the homeland failing to take place in the more conservative colonial setting. A prime example of this would be the development of Creole society in Latin America. Therefore, as a means of maintaining an identity threatened by being a minority in a colonial setting, creolisation led elements of Anglo-Norman society to resemble the society of twelfth century England long into the fourteenth century and beyond. Turning to acculturation, which is the process of transference of elements of one society to another over the course of their contact in geographical and cultural space, there is evidence for it occurring in both directions. The early generations of Anglo-Norman settlers had intermarried with Gaelic elements of society, a common feature in Norman colonial activity since the years of the settlement of Normandy itself (Lydon 1972, 57). This would have greatly contributed to the adoption of elements of the Gaelic language and customs, as mothers are key transmitters of cultural traits, being much more involved in early habitus formation than fathers (cf. Elias 2000; Bourdieu 1977). The interaction between children crucial for understanding processes such as this (cf. Pinker 1994), and further investigation of this aspect of cultural syncretism in Ireland could provide fruitful results. 116  More than any other document of the period, the Statutes of Kilkenny (1366) represent the processes of creolisation and acculturation ongoing in Ireland at the time. They are an attempt to set out the legitimate cultural traits of colonists, attempting to counter-act the “degeneracy” feared by the London government. Both processes are usually taken as being signs of colonial degeneracy in the homeland, as can be seen from the various descriptions of Spanish Creoles back in Castile in the early modern period, where Los Peninsulares regarded their Creole cousins in South America as totally degenerate in body and mind and no longer true Spaniards (cf. Darwin 2007; Colas 2007). Similar views were held in London regarding the medieval colonists in Ireland, and Lydon (1972, 57) draws attention to the use of the term degeneres from 1297 in a series of parliamentary enactments related to the English in Ireland who had adopted certain Irish habits. However, the case for acculturation should not be overstated, as even as late as the sixteenth and seventeenth centuries the descendants of the Anglo-Norman colonists were still regarded as foreigners by the majority of the Gaelic Irish population. Although by that point they had repeatedly married into many Gaelic families, especially on the periphery of their influence, the very fact that they began to refer to themselves as Old English in the seventeenth century shows that in order to retain their identity they defined themselves in contradistinction to the other groups on the island. Interestingly, James I referred to them as ‘half-subjects of mine’ in 1614 (Lydon 1988, 172) and they were regarded as degenerate due to their environment by many in seventeenth century England. Acculturation also took place in the other direction, as may be seen in the eventual use of the castle form, both in the reuse of Anglo-Norman castles themselves, the possible construction castles based on the Anglo-Norman model in 117 the thirteenth century, and the adoption of the tower house form (Barry 1987; O’Conor 1998; O’Keeffe 2000), along with settlement in Anglo-Norman towns, and possibly the adoption of the language in places. The so-called Gaelic resurgence from the thirteenth century later may have arrested some of these trends of acculturation. The decline is materially manifested in the abandonment of various castles and settlements, such as at the borough at Bunratty and a number of deserted medieval settlements. However, as already stated in the case of the latter many remained in use until the early modern period, and it is these that are the most visible in the landscape (cf. O’Conor 1998). The decline of Bunratty is reasonably well documented in the Calendar of State Papers, and in Caithréim Thoirdhealbhaigh. Environmental factors may have played a significant role in the decline of marginal contact zone boroughs such as Bunratty, which would have been affected most by the decline in weather conditions, which led to worsening harvests and the Great European Famine. They would also have been affected by the failure to continue the expansion of the colony, the devolution of power in colonial Ireland and the attendant resurgence of the Gaelic elite. 118 Chapter 5. Scandinavian and Anglo-Norman Colonialism Considered Scandinavian colonialism in Ireland seems to have been more commercially orientated than land orientated. The nature of Scandinavian colonial activity in Ireland was not lost on Hore (1857, 433), who states that ‘Dublin, Limerick, and Waterford were to them as Calcutta, Madras, and Bombay to our Indian merchants in the days of Clive’. This was probably due to the failure to win any tracts of land in the ninth century, although there may have been rural settlement in areas of the western seaboard devoid of Gaelic settlement, and in similar conditions to Scandinavian settlement in Scotland (cf. Kelly 2010; Barrett 2003; 2004). It seems to have been a largely urban phenomenon in its mature phase, and those urban centres may have had hinterlands. Anglo-Norman colonialism was totalising, involving the reordering of geographic and social space in the areas into which it spread. It was largely based on land, although existing urban centres were incorporated into the colonial system, and new towns were encouraged. Anglo-Norman colonialism followed a model of expansion-consolidation-domination in each area of Ireland it occurred. That said, it still had more regional variation than Scandinavian activity, which seems to have developed in a similar fashion in both case studies, and in areas such as Dublin and Cork, although this may change with the work being done on possible rural settlement in Connacht. Anglo-Norman colonialism in Ulster differed greatly from the south- east. The former area saw elite-replacement colonialism, where a martial colonial elite displaced the former elite at the head of the same exploitative system, with some alterations. This had had parallels in the Norman Conquest of England and in aspects of the Ottonian expansion east of the Elbe. The south-east was characterised by plantation colonialism, which saw the movement of several orders of society and the 119 displacement or incorporation of those previously there. Anglo-Norman colonialism in the mid-west represents an ultimately unsuccessful attempt at the latter form. Across Europe, Scandinavian colonialism overall was much more varied, although in all instances it showed a remarkable adaptive capability, ranging from the rapid acculturation of Danish settlers in northern Francia, to the rural, townless settlement of Scotland, from the culturally conservative settlement of the terra nullius of Iceland and the failed settlement of North America, to the hybrid nature of the Rus’. 5.1 Expansion From the evidence outlined in chapters 3 and 4, both phases of activity saw the construction of campaign fortresses. Some of the early campaign fortifications may have only been intended as temporary bridgeheads in the landscape, such as the Anglo-Norman fortifications at Baginbun, Co. Wexford and perhaps some longphuirt with single annalistic references. However, the majority of monuments of expansion seem to have been established with a view to relatively long term activity. Evidence for this is provided by the durability of Anglo-Norman ringworks, which helped to firmly establish a power base in whatever regions they moved into. The longphort at Woodstown, Co. Waterford has provided evidence for its use over several decades, as outlined above, and its size indicates an intention of maintaining a presence in the region. Further excavation at longphort sites may prove this to be the case for several others, including Athlunkard. The distribution maps have shown that the known longphuirt in the two case- study regions were located in areas of low Gaelic secular settlement, possibly due to the suitability of lowland areas for pastoralism. There seems to have been a greater 120 ecclesiastical presence in these areas, Wexford and Athlunkard both seem to have been located in close proximity to early ecclesiastic sites. Their location in political borderlands is significant, as it would seem as though they sought to establish themselves at the interstices of existing power networks, possibly due to their numerical inferiority. Their location indicates a different conception of geographical space, as for Scandinavians water was a conduit, rather than a terminus of authority. Ringworks also largely avoided areas associated with a high density of Gaelic settlement, with some exceptions, probably due to agricultural requirements. The obtaining of land conducive to arable agriculture would have been a major incentive for colonists. The grants made in the initial years of the Anglo-Norman colony were on such land. These areas would have been doubly favourable to colonial activity due to their possibly lower Gaelic settlement densities, thus rendering the displacement or subjugation of the local population much less difficult. This would in effect leave the land ‘half-conquered’, which in turn would have major ramifications from the late thirteenth century (cf. Bartlett 1993). Some ringworks were located at strategically important locations, such as at Béal Ború and Kilkenny, where sites formerly associated with Gaelic elite settlement were either appropriated and reused, or had new fortifications built in close proximity. The same also seems to have occurred at Hiberno-Scandinavian settlements. These actions could have been for symbolic reasons, a statement of the colonists’ intent on replacing the old order, and/or for military reasons, such as the gaining of control over major land routes, rivers and their interstices. Ringworks may also have been placed at such strategic locations with no pre-existing elite settlement, such as at Carlow. 121 5.2 Consolidation The consolidation sub-phase of Scandinavian colonial activity is evinced by urban settlement. As noted in chapter 3, little is known about the early years of Scandinavian urbanism in Ireland, but it is likely that the towns in both case studies began as longphuirt, possibly passing through an as yet properly unidentified intermediate phase. These towns may or may not have acquired hinterlands, as they must have obtained food from some source. However, their size in the consolidation phases is difficult to ascertain, and it may be that they obtained the necessary provisions by trade, possibly with arable surplus producing ecclesiastic sites. As Christian (2005, 245) observes, ‘like stars, cities and states reorganise and energise the smaller objects within their gravitational field’. The Hiberno-Scandinavian towns in Ireland certainly distorted Irish society and the local settlement pattern around them, but as will be discussed in the next phase of analysis, they failed to reorder that pattern and attain a dominant position within it, being instead incorporated into it. The motte is the metonymic monument for the consolidation sub-phase of Anglo-Norman colonialism in Ireland. The labour necessary for their construction would seem to preclude their suitability for expansionary activity. They indicate a strengthening grip on the landscape, and a major statement of intent. The ability on the part of their builders to manipulate the resources necessary for their construction would not have been lost on contemporaries. Their construction could be used to structure the experience of contemporaries, materialising the new social order by encoding it in the landscape, in turn (re)shaping the habitus of contemporaries via their interaction with this altered landscape. This holds for all elite Anglo-Norman colonial monuments, but especially for mottes and masonry castles. In cases where a motte replaced a ringwork a message of continuity and permanence would have been 122 portrayed to those lower orders of society, native and colonial under the sway of its occupant. The replacement of a native nodal point in the network of power relations with a motte would have portrayed a message of the supersedence of the old elite with a new and more powerful one. Their construction on new sites would have indicated the arrival of totalising elite, with the ability to literally reorder the landscape. 5.3 Incorporation While Anglo-Norman colonialism had a very distinct domination phase, the same cannot be said for Scandinavian colonialism in Ireland. It might even be said that it is at this point that the expansion-consolidation-domination model breaks down. This is significant, as it can be taken as an indication that not all examples of colonialism reached a phase of domination. Instead, Scandinavian colonialism in Ireland went through a phase of incorporation, as colonial settlements came to be dominated by Gaelic elite groups. Hiberno-Scandinavian urban sites developed an interesting spatial duality. They were part of a central place hierarchy in Gaelic socio-political and economic geography, but they also were nodal points in a trans-European urban network. A similar duality of being both connected to and disconnected from local geographic space is to be seen at other points in time in the Hanseatic, Phoenician and central and eastern Greek cities (cf. Tilly 1992; Gosden 2004). It is also found in the contemporary globalised system of cities, are at once part of a local/national central place hierarchy and nodal points in an international space contracting urban network. They also occupied an interesting point in social space. They were Hiberno- Scandinavian, but cannot be said to have been culturally Gaelic or Scandinavian. They were part of Gaelic society, but not Gaelic, they were part of the trans-European 123 Scandinavian diaspora, but differed in many ways from their home culture. They occupied what Bhabhi (1994) terms as “third space”. They arrived at this juncture due to a number of cultural processes associated with colonialism. The first of these is creolisation, where cultural traits from the point of origin at the time of departure for the colonists are conserved and even exaggerated. Perhaps this might be behind the continued use of a house type in Ireland only found at Kaupang, a site which had fallen out of use in the decades around 900AD (Valente 2008, 63). There was some diffusion of material culture style between the two societies, but the coloniser was also changed via Christianisation. The later Hiberno- Scandinavian towns were major ideological centres, with the large cathedrals located in Dublin, Limerick, Cork and Waterford. It is worth noting that the clergy of these settlements looked east to Canterbury for ecclesiastic guidance, either due to cultural separateness from Irish society or due to the nature of the contemporary Irish church (cf. Watt 1972). Hiberno-Scandinavian society was a hybrid society. The recent genetic study by McEvoy et al (2006) has illustrated that the actual numbers of settlers of Scandinavian origin in Ireland may have been much lower than suspected, possibly being confined to the elites of the relevant settlement. This would mean that many of the people being referred to as “foreigners” as late as the eleventh and twelfth century may have been acculturated Gaelic Irish. 5.4 Domination As can be seen from the distribution maps, Anglo-Norman colonial activity had a greater impact on the settlement pattern than Scandinavian activity, introducing several new forms of monument and a new settlement hierarchy. While it was very 124 much driven by the desire for land, it also involved several aspects of social activity, including commerce. Anglo-Norman society did not distort the existing settlement pattern with the insertion of new nodal points in the landscape of power like Scandinavian colonialism. Instead it sought to entirely re-order social and physical space. An exception to this might be the continued use of native land subdivisions (MacCotter 2008). They seem to have seen themselves as conquering both a people and a landscape, as can be surmised from a reading of The History and Topography of Ireland. Anglo-Norman colonialism in Ireland reached its apogee from the late twelfth to the early fourteenth centuries. However, the level of colonial domination varied between regions, as can be seen from the case studies. The south-east saw much greater activity, as the distribution of Anglo-Norman parish churches, deserted settlements, urban settlements, continental religious houses and moated sites for each case-study, discussed in chapter 4, illustrates. The establishment of a hierarchy of settlement was an extremely important feature of colonial domination. It echoed the rigid contemporary Anglo-Norman hierarchical social pyramid. In the period of colonial domination, those castles at the upper end of the social hierarchy tended to be large masonry castles, with mottes and even ringworks continuing as castles of the middle ranks of the colonial nobility. All served as centres for the accumulation and redistribution of agricultural surplus, high status portable material culture both locally derived and from long distance trade, craft centres and as administrative centres. They could provide the focus for nucleated or semi-nucleated clusteration. This would account for the low number of masonry castles in comparison to mottes and ringworks, as can be seen from figures 4.9 and 125 4.25. High status castles seem to have been related to the distribution of urban centres, both often located at key strategic points in the landscape. 5.6 Colonialism and Imperialism in Medieval Ireland and Beyond Much more evidence exists for Anglo-Norman colonialism in Ireland, and on first inspection it seems the more useful of the two periods for developing a theory of colonialism. However, the study of Scandinavian colonialism also provides important lessons, especially when compared with the subsequent Anglo-Norman period. Said defines colonialism as being ‘the implanting of settlements on distant territory’, and being ‘almost always a consequence of imperialism, which he sees as meaning ‘the practice, the theory, and the attitudes of a dominating metropolitan centre ruling a distant territory’ (1993, 8). While this definition applies to the Anglo- Norman period, it would seem as though Scandinavian colonialism in Ireland is part of the minority of non-imperial colonialisms alluded to by Said. Both phases began as sheer opportunism. Arguably Scandinavian colonialism continued as such, and it seems to have defied attempts at centralisation until the incorporation of the colonies into the Irish political scene. The arrival of various groups attempting to impose authority over Scandinavian settlements in Ireland in the mid-tenth century seem to have only be partially successful, with authority quickly becoming devolved, Dublin, Waterford and Limerick quite quickly had their own kings. No metropolitan centre can be identified for Scandinavian colonial activity in Ireland. Therefore, the Scandinavian section of the project is important, as it provides an example of non-imperial colonialism. Competition between competing counter-poised heterarchical elites western Scandinavia may have inspired raiding. It could have provided a means to 126 concentrate resources and maintain or expand social power at the expense of neighbours in a pre-emptive action (cf. Elias 2000; Tilly 1992; Roesdahl 1998; Barrett 2008). Many groups in western Scandinavia may have suffered a diminished social position due to the effects of elite overpopulation and the attendant contraction of elite social space due to the growth of centralised authority, and this may explain the genesis of more permanent settlement in Ireland and elsewhere. The growing centralisation of the Danish crown, even in its early stages in the ninth century may have caused further diminution of authority of other members of the western Scandinavian elite, which may further have inspired a desire to relocate for elements of this elite. Central authority was imposed within little more than a year in Ireland, with Henry II having learned from the experience of his ancestors and cousins elsewhere in space and time, acting decisively to prevent feudalization in Ireland, and de Clare setting himself up as de facto king of Ireland. Anglo-Norman activity in Ireland can therefore be seen as part of an imperial figuration, a tenuous Angevin Empire, consisting of England, parts of France, Wales and Ireland, with a metropolitan centre at London. Anglo-Norman colonialism in Ireland had itself started as sheer opportunism, like that of their cousins in southern Italy, started as pure opportunism on the part of a social group with few other opportunities, such as the non-inheriting sons, a group capable of causing extreme social instability, as they were largely cut off from inheritance, and they all could not be accommodated in the Church (cf. Elias 2000). Therefore, they became the main agents in warfare and polity expansion. Elite overpopulation such as this is a phenomenon common to many hierarchical societies. It most likely accounts for many of the changes occurring in medieval Ireland prior to 127 the arrival of Scandinavian and Anglo-Norman colonists. Hierarchical societies are pyramidically structured, and elite overpopulation at the higher levels can cause a downward movement through the social orders for some elements of society. Unlike Anglo-Norman colonialism, Scandinavian colonialism failed to make any large territorial gains in Ireland. Whether raiding or trading, their activity seems to largely have been economic in Ireland. Anglo-Norman colonists sought a more varied power base, strategically gaining control over ideological, military, economic and political resources. This meant that they had a much greater impact both in the short and long term. However, for a number of reasons discussed at the end of chapter 4, they failed to impose total hegemony, with Ireland remaining a “half- conquered” land. This eventually led to the decline of the centralised colony from the late thirteenth century, if not the decline of some of the colonial magnatial families in Ireland. Both colonial phases saw differing degrees of incorporation into native society, although they both became “third nations” living in a “third space” (cf. Bhabha 1994), different from both their home culture and the native culture of Ireland. Both continue to be referred to as “foreigners” or other such terms long after the fading of their colonial spurts. Both can be said to have become creolised societies, but there seems to have been a low level of acculturation on the part of the colonists. There may have been a much higher incidence of acculturation of Gaelic elements of society, especially in contact zones and third spaces such as the towns of both periods, and possibly in areas of dense colonial settlement such as southern Wexford. The development of Hiberno-Scandinavian culture after the arrival of the Anglo-Normans is difficult to surmise, and would reward further study. They seem to have been incorporated into either Gaelic or Anglo-Norman society, much the same as 128 Anglo-Norman groups came to be absorbed into Gaelic and English society after the re-colonisation of the island in the sixteenth and seventeenth centuries. To conclude, Anglo-Norman and Scandinavian activity in Ireland must be regarded as being emblematic of different forms of colonialism. Moreover, Anglo- Norman activity in Ireland itself provides examples of two different forms of colonialism, the elite replacement colonialism as seen in Ulster, and plantation colonialism, as seen in the south-east. Scandinavian colonialism must be regarded as opportunistic colonialism. The two phases differed greatly in their approaches and impacts, and provide a worthwhile comparative exercise. Scandinavian colonialism was commercially orientated, and had a low landscape impact and lacked centrality, and was therefore not connected to an imperial project. Anglo-Norman colonialism was imperialistic, totalising and had a major impact on the landscape, and indeed on the subsequent historical development of Ireland. Chapter 6. Epilogue: Towards a Theory of Colonialism As noted throughout the study, colonialism can be seen as a process, which is made up of three main sub-processes: expansion, consolidation and domination. Acculturation, a related parallel process, is an extremely important indicator of colonial success, often highly visible in the settlement record, with natives adopting colonial forms. It can positively or negatively reinforce these sub-processes, and is thus somewhat of a wild card. There may have been a phase of pre-colonial contact, which can either be short or long term, with varying intensity. 129 Type Opportunistic Elite Replacement Plantation Social Bottom Up Top Down Near total Characteristic Main Motivation Economic Economic, political Political Centrality Non-imperial Both Imperial Irish Examples Vikings in Ireland Anglo-Norman Anglo-Norman Ulster Wexford European Hanseatic Cities, Norman England, Levant, Wales Examples Greek Cities Italy and Sicily Approximation to Shared Cultural Shared Cultural Terra Nullius Gosden Typology Milieu, Middle Milieu, Terra Ground Nullius Table 5.1, centralised and non centralised colonialism ideal types Sub Process Expansion Consolidation Domination Dominant source of Militaristic Economic Political, social power Ideological Characteristic Campaign Centres of Hierarchical Monuments Fortresses exploitation configuration Assoc. with forms Opportunistic, Opportunistic, Elite Replacement, from Table 1. Elite Replacement, Elite Plantation Plantation Replacement, Plantation Table 5.2, sub-processes of the colonial process. The terms in bold represent the key term associated with each dub-phase Forms of Colonialism Colonialism can take many forms. The first of these is small scale movement, and is termed here as opportunistic colonialism. It concerns the relatively small scale movement of people, with low overall impact on the landscape. It can lead to changes in the society into which it occurs, but those changes are not enforced changes imposed by colonists. This form can often be a precursor to one of the other two 130 forms, or it can remain as it is. Examples are to be found in the Scandinavian settlement of Ireland, the Hanseatic cites of the Baltic, and the Greek cities of the central and western Mediterranean. The second of these is ‘elite replacement colonialism’, where an incoming group of colonists displace the native elite. The region’s basic socio-economic structure often remains the same, although new elements may be introduced so that the colonist can profit further from the situation. In a more extreme case, society may be reordered along the lines of the colonising society, although with only a small number of colonisers at the apex of the new society, such as in the case of the Roman Empire. An example of the former variant would be Anglo-Norman activity in Ulster (cf. McNeill 1980). It is an extremely common phenomenon in the historical record, and can be seen at work in British India. The third of these is ‘plantation colonialism’ where a movement of several orders of society into a region occurs. It is usually associated with imperial projects, as it usually results in an extremely firm grip on the subjugated area, at least for a time. It involves the displacement and replacement of colonised natives in the settlement pattern and society, or their incorporation into these. However, it can also involve the movement into previously unpopulated space, such as in the case of the settlement of Iceland. In an interesting piece of casuistry, those carrying out plantation colonialism may portray an area as devoid of settlement, as being terra nullius, due to its non-cultivation by natives, such as in the case of the European expansion into North America. Ireland provides examples of plantation colonialism at two different points in time, in the high medieval, as discussed throughout this 131 project, and in the sixteenth and seventeenth centuries. Plantation colonialism has also been seen in the European settlement of North America, the Chinese expansion into Central Asia and in an unusual example, the post-war colonisation of Palestine. Processes associated with colonialism A number of socio-cultural processes are associated with colonialism, namely acculturation, hybridisation, and creolisation. Gosden (2004) notes that colonialism can change both the coloniser and colonised. This certainly is the case in a number of historical instances, the most salient of which is the Industrial Revolution in Europe Acculturation Hybridisation Creolisation Characteristic One dominant culture, Two cultures form a Colonial culture does either native or colonial new culture not keep pace with culture exerts strong socio-cultural in influence source region Irish Example Irish speaking Anglo- Hiberno-Norse towns Colonial South Normans, Gaelic in Ireland Wexford encastellation Global Example Roman Britain Phoenicians in the Spanish in Latin Maghreb America Table 5.3, Cultural Processes Associated with Colonialism 132 BIBLIOGRAPHY All websites listed have been checked, and found to be available as of 30 August 2010. Primary Caithréim Thoirdhealbhaigh or the Triumphs of Turlough (CT) http://www.ucc.ie/celt/published/G100062/index.html Dermot and the Earl http://www.ucc.ie/celt/published/F250001-001/index.html Chronicon Scotorum (CS) http://www.ucc.ie/celt/published/T100016/index.html Fragmentary Annals of Ireland (FA) http://www.ucc.ie/celt/published/T100017/index.html http://www.ucc.ie/celt/published/T100017/index.html (trans.) http://www.ucc.ie/celt/published/T100017/index.html The Annals of Connacht or Annála Connacht (AC) http://www.ucc.ie/celt/published/G100011/index.html (Trans.)) http://www.ucc.ie/celt/published/T100011/index.html The Annals of Inisfallen (AI) http://www.ucc.ie/celt/published/G100004/index.html (trans.) http://www.ucc.ie/celt/published/T100004/index.html The Annals of Loch Cé (LC) http://www.ucc.ie/celt/published/G100010A/index.html http://www.ucc.ie/celt/published/G100010B/index.html (trans.) http://www.ucc.ie/celt/published/T100010A/index.html (trans.) http://www.ucc.ie/celt/published/T100010B/index.html 133 The Annals of the Four Masters (AFM) http://www.ucc.ie/celt/published/g100005b/index.html (Trans.) Http://www.ucc.ie/celt/published/t100005b/index.html The Annals of Ulster or Annála Uladh (AU) http://www.ucc.ie/celt/published/g100001a/index.html http://www.ucc.ie/celt/published/g100001b/index.html (Trans.) http://www.ucc.ie/celt/published/t100001a/index.html (Trans.) http://www.ucc.ie/celt/published/t100001b/index.html Adomnán of Iona, 1995. The Life of St. Columba, Sharpe, R. (trans.). Harmondsworth: Penguin. Bede, 1994. The Ecclesiastical History of the English People, Colgrave, B. (trans.). Oxford: Oxford World’s Classics. Binchy, D. (ed.) 1941. Críth Gablach. Dublin: Stationary Office. Brooks, E. St John. 1950. Knights’ Fees in Counties Wexford, Carlow and Kilkenny. Dublin: Government of Ireland Stationary Office. Giraldus Cambrensis, 1978. Expugnatio, Scott, A.B & F.X. Martin (trans.). Dublin: Royal Irish Academy. Giraldus Cambrensis, 1982. The History and Topography of Ireland, O’Meara, J. (trans). Harmondsworth: Penguin. Clyn, Friar John, Annalium Hiberniae Chronicon ad annum MCCCXLIX (JC) (Latin) http://www.ucc.ie/celt/published/L100011/index.html (trans.) http://www.ucc.ie/celt/published/T250001-001/index.html Mills J. & M.C. Griffiths (eds) 1905-14. Calendar of Justiciary Rolls, Ireland (3 vols). Dublin 134 Sweetman, H.S. (ed.) 1876-86. Calendar of Documents Relating to Ireland, 1171- 1307 (5 vols). London Secondary Allen Brown, R. 1970. English Castles, Third Edition. London: Batsford Allen Brown, R. 1985. Castles. Aylesbury: Shire. Anderson, B. 2006. Imagined Communities, Revised Edition. London: Verso. Anderson, P. 1974. Passages from Antiquity to Feudalism. London: Verso. Barrett, G.F. and Graham, B.J. 1975. Some considerations concerning the dating and distribution of ring-forts in Ireland. Ulster Journal of Archaeology 38: 33–45. Barrett, J.H. 2003. Culture contact in Viking Age Scotland. In Barrett, J. (ed.) Contact, Continuity, and Collapse: The Norse Colonization of the North Atlantic, 73- 112. Turnhout: Brepols Publishers. Barrett, J.H. 2004. Beyond war or peace: the study of culture contact in Viking-age Scotland. In Hines, Lane & Redknap (eds.) Land, Sea and Home: Proceedings of a conference on Viking-period settlement, 207-18. Leeds: Maney, Society for Medieval Archaeology Monograph 20. Barrett, J.H. 2008. What caused the Viking Age? Antiquity, 82: 671-685. Barry, T.B. 1977. Medieval Moated Sites of South-East Ireland. Oxford: BAR Barry, T.B. 1987. The Archaeology of Medieval Ireland. London: Routledge. Barry, T.B. 2000. Rural settlement in medieval Ireland. In T.B. Barry (ed.), A History of Settlement in Ireland, 110-23. London. 135 Bartlett, R. 1992. The Making of Europe. Harmondsworth: Penguin. Bendix, R. 1978. Kings or People: Power and the Mandate to Rule. Berkeley and Los Angeles: University of California Press. Bennett, I. 2004-5. Archaeological excavations in County Wexford: a review of the last 35 years. Journal of the Wexford History Society, 20: 184-196. Berger, P. & Luckmann, T. 1966. The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Harmondsworth: Penguin. Berger, P. 1967. The Sacred Canopy/The Social Reality of Religion. Harmondsworth: Penguin. Bethel, DLT 1981. The Originality of the Early Irish Church. Journal of the Royal Society of Antiquaries of Ireland 111: 36-49. Bhabha, H.K. 1994. The Location of Culture. London: Routledge. Binchy, D.A. 1962a. Patrick and his biographers, ancient and modern. Studia Hibernica 2: 7-173. Binchy. D.A. 1962b. The passing of the old order. In Ó Cuív (ed.) The Impact of the Scandinavian Invasions on the Celtic-speaking Peoples c.800-1100AD, Proceedings from the International Congress of Celtic Studies, 119-32. Dublin: Dublin Institute for Advanced Studies. Bloch, M. 1961. Feudal Society. London: Routledge Bourdieu, P. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. Bourdieu, P. 1984. Distinction: A Social Critique of the Judgement of Taste. London: Routledge. 136 Bourke, E. 1995. Life in the Sunny South-East: Housing and Domestic Economy in Viking and Medieval Wexford. Archaeology Ireland, 9.3: 33-6. Bradley, J. 1985. Planned Anglo-Norman towns in Ireland. In Clarke & Simms (eds) The Comparative History of Urban Origins in Non-Roman Europe: Ireland, Wales, Denmark, Germany, Poland and Russia from the Ninth to the Thirteenth Century, 411-468. BAR International Series 255. Bradley, J. 1988. The interpretation of Scandinavian settlement in Ireland. In Bradley, J. (ed.) Settlement & Society in Medieval Ireland, 49-78. Kilkenny: Boethius. Bradley, J. 1990. The early development of the medieval town of Kilkenny. In Nolan & Whelan (eds.) Kilkenny: History and Society, Interdisciplinary Essays on the History of an Irish County, 63-73. Dublin: Geography Publications. Bradley, J. 1995. Walled Towns in Ireland. Dublin: Country House. Bradley, J. 2009. Some reflections on the problem of Scandinavian settlement in the hinterland of Dublin during the ninth century. In Bradley, Fletcher & Simms (eds) Dublin in the Medieval World, Studies in Honour of Howard B. Clarke, 39-62. Dublin: Four Courts Press. Brady, N. 2006. Mills in medieval Ireland: looking beyond design. In Steve Walton (ed.) Wind and Water in the Middle Ages. Fluid technologies from Antiquity to the Renaissance, 39-68. Tempe: Arizona Center for Medieval and Renaissance Studies. Braudel, F. 1993. A History of Civilisations, Harmondsworth: Penguin. Brindley, A. & Kilfeather, A. 1993. Archaeological Inventory of County Carlow. Dublin: Government of Ireland Stationary Office. Buchanan, R.H. 1970. Rural Settlement in Ireland. In N. Stephens & R.E. Glasscock (eds), Irish Geographical Studies in honour of E. Estyn Evans, 146-161. Belfast: Queen’s University Press. 137 Byrne, F.J. 1987. The trembling sod: Ireland in 1169. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534. 1-42. Oxford: Oxford University Press. Byrne, F.J. 2005a. The Viking Age. In Ó Cróinín, D. (ed.), A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 609-634. Oxford: University Press. Byrne, F.J. 2005b. Church and politics, c.750-c.1100. In Ó Cróinín, D. (ed.), A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 656-679. Oxford: University Press. Byrne, F.J. 2005c. Ireland before the Battle of Clontarf . In Ó Cróinín, D. (ed.), A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 852-861. Oxford: University Press. Byrne, F.J. 2005d. Ireland and her neighbours, c.1014-c.1072. In Ó Cróinín, D. (ed.), A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 862-898. Oxford: University Press. Cahill, T. 1995. How the Irish Saved Civilization : the untold story of Ireland's Heroic Role from the Fall of Rome to the Rise of Medieval Europe. London: Hodder & Stoughton. Campbell, E. 2007. Continental and Mediterranean Imports to Atlantic Britain and Ireland, AD 400-800. York: Council for British Archaeology Capelli et al 2003. A Y chromosome census of the British Isles. Current Biology 13: 979-984. Cathcart King, D.J. 1988. The Castle in England and Wales. Beckenham: Croom Helm. Cavalli-Sforza, L.L. 2000.Genes, Peoples and Languages. London: Allen Lane. 138 Christian, D. 2004. Maps of Time: An Introduction to Big History. Berkeley: University of California Press. Critchley, J.S. 1978. Feudalism. London: George Allen & Unwin. Clarke, Helen (1984). The Archaeology of Medieval England. London: British Museum. Clarke, Howard 1998. Proto-towns and towns in Ireland and Britain in the ninth and tenth centuries. In Clarke, Ní Mhaonaigh & Ó Floinn (eds) Ireland and Scandinavia in the Early Viking Age, 331-380. Dublin: Four Courts Press. Clinton, M. 2001. The Souterrains of Ireland. Bray: Wordwell. Clyne, M. 2007. Kells Priory, Co. Kilkenny: Archaeological Excavations by T. Fanning & M. Clyne. Dublin: Government of Ireland. Craib, I. 1984. Modern Social Theory, from Parsons to Habermas. Brighton: Wheatsheaf. Colas, A. 2007. Empire. Cambridge: Polity. Colfer, B. 1987. Anglo-Norman settlement in County Wexford. In Whelan, K. (ed.) Wexford: History and Society, 65-101. Dublin: Geography Publications. Collins, J. 1995. Architectures of Excess: Cultural Life in the Information Age. London: Routledge. Comber, M. 2008. The Economy of the Ringfort and Contemporary Settlement in Early Medieval Ireland. BAR Int. Ser 1773. Conolly, J & Lake, M. 2006. Geographical Information Systems in Archaeology. Cambridge: Cambridge University Press. 139 Critchley, J. Feudalism. London: George Allen & Unwin. De Botton, A. 2004. Status Anxiety. Harmondsworth: Penguin. DeMarrais, E. 2004. The Materialization of Culture. In DeMarrais, Gosden & Renfrew (eds) Rethinking Materiality: The Engagement of Mind With the Material World. Cambridge: McDonald Institute For Archaeological Research. De Meulemeester, J. & O’Conor, K. 2007. Fortifications. In Graham-Campbell & Valor (eds.) The Archaeology of Medieval Europe, Volume 1: Eighth to Twelfth Centuries AD, 316-41. Aarhus: Aarhus University Press. Darnton, R. 1984. The Great Cat Massacre and Other Episodes in French Cultural History. New York: Basic Books. Darwin, J. 2007. After Tamerlane: The Rise and Fall of Global Empires, 1400-2000. Harmondsworth: Penguin. Diamond, J. 1997. Guns Germs and Steel: a Short History of Everybody for the Last 13,000 Years. London: Vintage Books. Diamond, J. 2005. Collapes: How Some Societies Choose to Fail or Survive. Harmondsworth: Penguin. Doherty, C. 1980. Exchange and trade in Early Medieval Ireland. Journal of the Royal Society of Antiquaries of Ireland 110: 67–89. Doherty, C. 1982. Some aspects of hagiography as a source for Irish economic history. Peritia 1: 300-328. Doherty, C. 1985. The monastic town in early medieval Ireland. In Clarke & Simms (eds) The Comparative History of Urban Origins in Non-Roman Europe: Ireland, Wales, Denmark, Germany, Poland and Russia from the Ninth to the Thirteenth Century, 45-76. BAR International Series 255. 140 Dolley, M. 1987. Coinage, to 1534: the sign of the times. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 816-826. Oxford: Oxford University Press. Dolley, M. 1972. Anglo-Norman Ireland. Dublin: Gill & Macmillan. Down, K. 1987. Colonial society and economy. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 439-491. Oxford: Oxford University Press. Downham, C. 2004. The career of Cearbhall of Osraighe. Ossory, Laois and Leinster, 1: 1-18. Downham, C. 2007. Viking Kings of Britain and Ireland: The Dynasty of Ívarr to AD 1014. Edinburgh: Dunedin. Duncan, A.M.D. 1975. Scotland: The Making of the Kingdom. Edinburgh: Oliver & Boyd. Duffy, S. 1997. Ireland in the Middle Ages. Dublin: Gill & Macmillan. Edwards, J. 1985. Language, Society and Identity. Oxford: Blackwell. Edwards, N. 1990. The Archaeology of Early Medieval Ireland. London: Batsford. Edwards, N. 2005. The archaeology of early medieval Ireland, c.400-1169: settlement and economy. In Ó Cróinín, D. (ed.), A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 235-300. Oxford: Oxford University Press. Elias, N. 1983. The Court Society. Oxford: Blackwell. Elias, N. 2000[1939], The Civilising Process, Revised Edition. Oxford: Blackwell. 141 Empey A. 1982-3. Medieval Knocktopher: a study in manorial settlement. Old Kilkenny Review 2: 239-42, 441-52. Empey, A. 1990. County Kilkenny in the Anglo-Norman period. In Nolan & Whelan (eds.) Kilkenny: History and Society, Interdisciplinary Essays on the History of an Irish County, 75-95. Dublin: Geography Publications. Estyn Evans, E. 1957. Irish Folk Ways. London: Routledge. Etchingham, C. 1999. Church Organisation in Ireland, AD 650-1000. Dublin: Laigin Press. Etchingham, C. 2010. The Irish Monastic Town: Is this a Valid Concept, Kathleen Hughes Memorial Lectures 8. Cambridge: Hughes Hall & Department of Anglo- Saxon, Norse and Celtic, University of Cambridge. Fanning, T. 1973-4. Excavations of a ringfort at Pollardstown, Co. Kildare. Journal of the Kildare Archaeological Society, 15: 251-261. Fanning, T. 1981. Excavation of an Early Christian cemetery and settlement at Reask, County Kerry. Proceedings of the Royal Irish Academy 81C: 67-172. FitzPatrick, E. 2009. Native Enclosed Settlement and the Problem of the Irish ‘Ring- fort’. Medieval Archaeology 53: 271-307. Foucault, M. 1967. Madness and Civilisation. London: Routledge. Foucault, M. 1970. The Order of Things. London: Routledge. Foucault, M. 1977 Discipline and Punish: The Birth of the Prison. Harmondsworth: Penguin. 142 Gibbons, M. 1990. The archaeology of early settlement in county Kilkenny, in W. Nolan & K. Whelan (eds.) Kilkenny: History and Society, 1–32. Dublin: Geography Publications. Gibbons, M. 2005. Athlunkard (Áth-an-Longphort): A reassessment of the proposed Viking fortress in Fairyhill Td, County Clare. The Other Clare 29: 22–5. Gibbons, M. & Gibbons, M. 2009. The search for the ninth-century longphort: early Viking-age Norse fortifications and the origins of urbanisation in Ireland. In S. Duffy (ed.) Medieval Dublin VIII: Proceedings of the Friends of Medieval Dublin Symposium 2006, 9–20. Dublin: Four Courts Press. Glasscock, R.E. 1970. Moated Sites, and Deserted Villages: Two Neglected Aspects of Anglo-Norman Settlement in Ireland. In N. Stephens & R.E. Glasscock (eds), Irish Geographical Studies in honour of E. Estyn Evans, 162-177. Belfast: Queen’s University Press. Glasscock, R.E. 1987. Land and people, c.1300. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 205-239. Oxford: Oxford University Press. Gosden, C. 2004. Archaeology and Colonialism: Cultural Contact from 5000 BC to the Present. Cambridge: Cambridge University Press. Graham, B.J. 1985. Anglo-Norman colonization and the size and spread of the colonial town in medieval Ireland. In Clarke & Simms (eds) The Comparative History of Urban Origins in Non-Roman Europe: Ireland, Wales, Denmark, Germany, Poland and Russia from the Ninth to the Thirteenth Century, 355-72. BAR International Series 255. Graham, B.J. 1993. The High Middle Ages: c. 1100 to c. 1350. In B.J. Graham and L.J. Proudfoot (eds), A Historical Geography of Ireland, 58-98. London: Academic Press. 143 Graham, B.J. 2000. Urbanisation in Ireland during the High Middle Ages. In Barry, T.B. (ed.) A History of Settlement in Ireland, 124-139. London. Graham-Campbell, J. 1976. The Viking Age silver hoards of Ireland. London. Viking Society for Northern Research. In B. Almquist and D. Greene (eds) The Proceedings of the Seventh Viking Congress. London: Dundalk Press. Gramsci, A. 1971. Selections from the Prison Notebooks, Hoare, Q. (trans.). New York: International Publishers. Gwynn, A. & Hadcock, N.D. 1970. Medieval Religious Houses: Ireland. London: Longman. Hamlin, A. & Hughes, K. 1997. The Modern Traveller to the Early Irish Church, Second Edition. Dublin: Four Courts Press. Higham, R. & Barker, P. 2000. Hen Domen, Montgomery - A Timber Castle on the English-Welsh Border: A Final Report. Exeter: University of Exeter Press. Hodkinson, B. 2004. The topography of pre-Norman Limerick. North Munster Antiquarian Journal, 42, 1-6. Hodkinson, B. 2004. A reinterpretation of the Castle at Clarecastle. The Other Clare 28: 55-8. Holm, P. 1986. The slave trade in Dublin, ninth to twelfth centuries. Peritia 5: 317- 45. Hughes, K. 1966. The Church in Early Irish Society. London: Methuen. Hughes, K. 1972. Early Christian Ireland: An Introduction to the Sources. London: Hodder & Stoughton. 144 Hughes, K. 2005a. The Church in Irish Society 400 – 800. In Ó Cróinín, D. (ed.) A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 301-30. Oxford: University Press. Hughes, K. 2005b. The Irish Church 800-c.1050. In Ó Cróinín, D. (ed.) A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 635-655. Oxford: University Press. Hurley, M.F, O.B.M. Scully & S.J. McCutcheon (eds) 1997. Late Viking Age and Medieval Waterford. Waterford: Waterford City Council. Hurley, M.F. 1998. Viking Age towns: archaeological evidence from Waterford and Cork. In Monk & Sheehan (eds) Early Medieval Munster: Archaeology, History and Society. Cork: Cork University Press. Hurley, M.F. 2010. Viking elements in Irish towns: Cork and Waterford. In Sheehan & Ó Corráin (eds) The Viking Age: Ireland and the West – Papers from the Proceedings of the Fifteenth Viking Congress, Cork, 18-27 August 2005, 154-164. Dublin: Four Courts. Jenkins, K. 1991. Rethinking History. London: Routledge. Jobling, M.A. In the name of the father: surnames and genetics. Trends in Genetics 17.6: 353-357. Jones, S. 1997. The Archaeology of Ethnicity. London: Routledge. Kelly, E.P. & O’Donovan, E. 1998. A Viking longphort near Athlunkard, Co. Clare. Archaeology Ireland 34: 13-16. Kelly, E. P. & Maas, J. 1999. The Vikings and the Kingdom of Laois. In Lane & Nolan (eds.) Laois History and Society, 123–59. Dublin: Geography Publications. 145 Kelly, E.P. 2010. The Vikings in Connemara. In Sheehan & Ó Corráin (eds) The Viking Age: Ireland and the West – Papers from the Proceedings of the Fifteenth Viking Congress, Cork, 18-27 August 2005, 174-187. Dublin: Four Courts. Kelly, F. 1988. A Guide to Early Irish Law. Dublin: Institute for Advanced Studies. Kelly, F. 1997. Early Irish Farming. Dundalk: Dundalgan Press. Kenny, M. 1987. The geographical distribution of Irish Viking-age coin hoards. Proceedings of the Royal Irish Academy 87C: 507-25. Kenny, M. 2005. Coins and coinage in pre-Norman Ireland. In Ó Cróinín, D. (ed.) A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 842-851. Oxford: University Press. Kenward, H. K. & E. P. Allison 1994. A preliminary view of the insect assemblages from the early Christian rath site at Deer Park Farms, Northern Ireland, in J. Rackham (ed.) Environment and Economy in Anglo-Saxon England: Proceedings of a conference held at the Museum of London, 9-10 April, 1990. London: Council for British Archaeology. Kerr, Thomas R. 2007. Raths and Ringforts in the landscape of northwest Ulster. Oxford: BAR 430. Kuhn, T.S. 1970. The Structure of Scientific Revolutions, Second Edition. Chicago: University of Chicago Press. Leask, H.G. 1941. Irish Castles and Castellated Houses. Dundalk: Dundalgan Press. Leask, H.G. 1955. Irish Churches and Monastic Buildings, I. Dundalk: Dundalgan Press. Leask, H.G. 1966. Irish Churches and Monastic Buildings, II. Dundalk: Dundalgan Press. 146 Loeber, R. 2001. An Architectural History of Gaelic Castles and Settlement. In Duffy, Edwards & FitzPatrick (eds) Gaelic Ireland c.1250-1650, 271-314. Dublin. Limbert, D. 1996. Irish ring-forts: a review of their origins. Archaeological Journal, 153: 243–89. Lukaćs, G. 1968. History and Class Consciousness, Revised Edition. London: Merlin. Lydon, J. 1972. Ireland in the Later Middle Ages. Dublin: Gill & Macmillan. Lydon, J. 1987a. The expansion and consolidation of the colony, 1215-54. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 156-178. Oxford: Oxford University Press. Lydon, J. 1987b. The years of crisis, 1254-1315. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 179-204. Oxford: Oxford University Press. Lydon, J. 1987c. A land of war. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 240-274. Oxford: Oxford University Press. Lydon, J. 1987d. The impact of the Bruce invasion, 1315-27. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 275-302. Oxford: Oxford University Press. Lynn, C J 1975a. The dating of raths: an orthodox view. Ulster Journal of Archaeology, 38: 45–7. Lynn, C J 1975b. The medieval ring-fort: an archaeological chimera. Irish Archaeological Research Forum 2: 29–36. Maas, J. 2008. Longphort, dún and dúnad in the Irish annals of the Viking period. Peritia 20: 257-275. 147 Macalister, R. A. S. 1928. The archaeology of Ireland. London: Methuen. McCormick, F. 1995. Cows, ringforts and the origins of Early Christian Ireland. Emania 13: 33–7. McCormick, F. 2008. The decline of the cow: agricultural and settlement change in early medieval Ireland. Peritia 20: 209-224. MacCotter, P. 2008. Medieval Ireland: Territorial, Political and Economic Divisions. Dublin: Four Courts Press. McEvoy et al 2006. The scale and nature of Viking settlement in Ireland from Y- chromosome admixture analysis. European Journal of Human Genetics, 14: 1288- 1294. McMahon, M. 1993. The charter of Clare Abbey and the Augustinian ‘Province’ in Co. Clare. The Other Clare 17: 21-28. McMahon, M. 1997. On a fertile rock: the Cistercian abbey of Corcomroe. The Other Clare 21: 22-31. McNeill, T 1975. Medieval raths? An Anglo-Norman comment. Irish Archaeological Research Forum, 2.1: 37–9. McNeill, T. 1980. Anglo-Norman Ulster: The History and Archaeology of an Irish Barony, 1177-1400. Edinburgh: John Donald. McNeill, T. 1997. Castles in Ireland: Feudal Power in a Gaelic World. London: Routledge. McNeill, T. 2006. Castles. London: Batsford. MacNiocaill, G. 1985. The colonial town in Irish documents. In Clarke & Simms (eds) The Comparative History of Urban Origins in Non-Roman Europe: Ireland, 148 Wales, Denmark, Germany, Poland and Russia from the Ninth to the Thirteenth Century, 373-8. BAR International Series 255. Martin, F.X. 1987a. Introduction: Medieval Ireland. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, xlix-lxii. Oxford: Oxford University Press. Martin, F.X. 1987b. Diarmait mac Murchadha and the coming of the Anglo-Normans. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 43-66. Oxford: Oxford University Press. Martin, F.X. 1987c. Allies and an overlord, 1169-72. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 67-97. Oxford: Oxford University Press. Martin, F.X 1987d. Overlord becomes feudal lord, 1173-85. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 98-126. Oxford: Oxford University Press. Martin, F.X 1987e. John, Lord of Ireland. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 127-155. Oxford: Oxford University Press. Martin, F.X. 1994. The Normans: Arrival and settlement. In T.W. Moody and F.X. Martin (eds) The Course of Irish History, Revised and Enlarged Edition. Cork: Mercier Press. Mills, C.W. 1956. The Power Elite. New York: Oxford University Press. Monk, M. 2007. The Plant Remains. In Clyne, M. 2007. Kells Priory, Co. Kilkenny: Archaeological Excavations by T. Fanning & M. Clyne, 383-4. Dublin: Government of Ireland. 149 Mooney, C. 1955. Franciscan architecture in pre-Reformation Ireland I. Journal of the Royal Society of Antiquaries of Ireland 85: 133-73. Mooney, C. 1956. Franciscan architecture in pre-Reformation Ireland II. Journal of the Royal Society of Antiquaries of Ireland 86: 125-69. Mooney, C. 1957. Franciscan architecture in pre-Reformation Ireland III. Journal of the Royal Society of Antiquaries of Ireland 87: 103-24. Moore, M. 1996. Archaeological Inventory of County Wexford. Dublin: Government of Ireland Stationary Office. Moore. M. 1999. Archaeological Inventory of County Waterford. Dublin: Government of Ireland Stationary Office Murphy, C. 1980. The Norman Manor of Bunratty. The Other Clare 4: 19. Mytum, H. 1992. The Origins of Early Christian Ireland. London: Routledge. Mytum, H. 2003. The Vikings in Ireland: Ethnicity, identity and culture change. In J. Barrett (ed.) Contact, Continuity and Collapse: The Norse Colonization of the North Atlantic, 113-137. Turnhout: Brepols. Naum, M. 2010. Re-emerging frontiers: Postcolonial theory and historical archaeology of the borderlands. Published online by the Journal of Archaeological Method and Theory. Nelson, J. 2003. England and the Continent in the ninth century: II, the Vikings and others. Transactions of the Royal History Society 13: 1-28. Nicholls, K.W. 1987. Gaelic society and economy. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 397-438. Oxford: Oxford University Press. 150 Nicholls, K.W. 2003. Gaelic and Gaelicized Ireland in the Middle Ages. Dublin: Lilliput. O’Brien, J. 1999. An outline of the history of the civil parishes of Bunratty and Drumline, c.1200-1700. The Other Clare 23: 34-43. O’Brien, R. & Russell, I. 2005. The Hiberno-Scandinavian site of Woodstown 6, County Waterford. In J. O’Sullivan and M. Stanley (eds) Recent Archaeological Discoveries on National Road Schemes 2004, Archaeology and the National Roads Authority Monograph Series 2, 111-124. Bray: Wordwell. Ó Carragáin, T. 2009. Cemetery settlements and local churches in pre-Viking Ireland in light of comparisons with England and Wales. In Ryan & Graham-Campbell (eds) Anglo-Saxon/Irish Relations Before the Vikings, Proceedings of the British Academy 157: 329-366. O’Conor, Kieran D. 1998. The Archaeology of Medieval Rural Settlement in Ireland. Discovery Programme Monograph 3. Dublin: Discovery Programme & Royal Irish Academy. O’Conor, K. 2002. Motte castles in Ireland: permanent fortresses, residences and manorial centres. Château Gaillard 20: 173-182. O’Conor, K. 2003. A reinterpretation of the earthworks at Baginbun, Co. Wexford. In Kenyon, J. & O’Conor, K. (eds.) The Medieval Castle in Ireland and Wales, 17-31. Dublin: Four Courts Press. Ó Corráin, D. 1972. Ireland before the Normans. Dublin: Gill & Macmillan. Ó Corráin, D. 1981. The early Irish churches: some aspects of organisation. In Ó Corráin, D. (ed.) Irish Antiquity, 328-341. Cork: Tower. Ó Corráin, D. 1989. Prehistoric and early Christian Ireland. In Foster, R. (ed.) The Oxford Illustrated History of Ireland, 1-52. Oxford: Oxford University Press. 151 Ó Corráin, D. 2005. Ireland c.800: Aspects of society. In Ó Cróinín, D. (ed.) A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 549-608. Oxford: University Press. Ó Cróinín, D. 1995. Early Medieval Ireland 400-1200. London: Longman. Ó Dálaigh, B. 1989. A history of Clare Castle, 1248-1891. The Other Clare 13: 40-48. Ó Floinn, R. 1998. The archaeology of the early Viking Age in Ireland. In Clarke, Ní Mhaonaigh & Ó Floinn (eds) Ireland and Scandinavia in the Early Viking Age, 131- 165. Dublin: Four Courts Press. Ó Floinn, R. 2006. Bishops, liturgy and reform: some archaeological and art historical evidence. In Bracken & Ó Riain-Raedel (eds) Ireland and Europe in the Twelfth Century: Reform and Renewal. Dublin: Four Courts Press. O’Keeffe, Tadhg 2000. Medieval Ireland – An Archaeology. Stroud: Tempus. O’Keeffe, Tadhg 2003. Romanesque Ireland. Dublin: Four Courts Press. Ó Ríagáin, R. 2008. People and symbols: A return to the Great Cat Massacre controversy. Carnival: The Journal of the International Students of History, 10: 27- 33. Ó Ríagáin, R 2010a. The rectilinear houses of the Irish Early Neolithic: The introduction of new identities, ideologies and economies. The Undergraduate Journal of Ireland & Northern Ireland, 1, 53-68. Ó Ríagáin, R 2010b The ecclesiastic round towers of Ireland: Date, origins, functions and symbolism. Trowel XII, forthcoming. Orpen, G.H. 1911-20. Ireland Under the Normans. Oxford: Clarendon. 152 O’Sullivan, A. 1998. The Archaeology of Lake Settlement in Ireland. Discovery Programme Monograph 4. Dublin: Discovery Programme & Royal Irish Academy. O’Sullivan, A. 2000. Crannogs: Lake Dwellings of Early Ireland. Dublin: Country House. O’Sullivan, A. 2001. Crannogs in Late Medieval Gaelic Ireland, c1350-c.1650. In Duffy, Edwards & FitzPatrick (eds), Gaelic Ireland c.1250-1650, 397-417. Dublin. Othway-Ruthven, A.J. 1951. The character of Norman settlement in Ireland. Reproduced in Crooks, P. (ed.) 2008.Government, War and Society in Medieval Ireland, Essays by Edmund Curtis, A.J. Othway-Ruthven and James Lydon, 263-274. Dublin: Four Courts Press. Othway-Ruthven, A.J. 1968. A History of Medieval Ireland. London: Benn. Platt, C. 1978. Medieval England: A Social History from the Conquest to 1600 AD, London: Routledge. Pinker, S. 1994. The Language Instinct. Harmondsworth: Penguin. Popper, K. 2002[1959]. The Logic of Scientific Discovery. London: Routledge. Pounds, N.J.G. 1990. The Medieval Castle in England and Wales: A Social and Political History. Cambridge: Cambridge University Press. Proudfoot, V. B. 1961. The economy of the Irish rath. Medieval Archaeology 5: 94– 122. Renfrew, C. 1986. Introduction: Peer polity interaction and socio-political change. In C. Renfrew, and J.F. Cherry (eds) Peer Polity Interaction and Socio-political Change, pp. 1-18. Cambridge: Cambridge University Press. 153 Roche, R. 1987. Forth and Bargy – a place apart. In Whelan, K. (ed.) Wexford: History and Society, 102-121. Dublin: Geography Publications. Roche, R. 1995. The Normans Invasion of Ireland. Dublin: Anvil. Roesdahl, E. 1998. The Vikings. Harmondsworth: Penguin. Russell, I. 2003. Archaeological Excavation of Woodstown 6. Unpublished excavation report. Russell, J.C. 1966. Late thirteenth-century Ireland as a region. Demography 3: 500- 512. Ryan, G. 1977. The Normans in Thomond. The Other Clare 1: 10-11. Said, E. 1978. Orientalism. Harmondsworth: Penguin. Said, E. 1993. Culture and Imperialism. London: Vintage. Sharpe, R. 1984. Some Problems Concerning the Organisation of the Church in Early Medieval Ireland. Peritia 3: 230-70. Sheehan, J. 1998. Early Viking Age silver hoards from Ireland and their Scandinavian elements. In Clarke, Ní Mhaonaigh & Ó Floinn (eds) Ireland and Scandinavia in the Early Viking Age, 166-202. Dublin: Four Courts Press. Sheehan, J. 2000. Ireland’s early Viking-Age silver hoards: Components, structure, and classification. Acta Archaeologica, 71: 49–63. Sheehan, J. 2004. Social and economic integration during the Viking Age – the evidence of the hoards. In J. Hines, A. Lane and M. Redknap (eds.) Land, Sea and Home, Society for Medieval Archaeology Monograph 20, pp.177 – 188. 154 Simms. K. 1989. The Norman invasion and the Gaelic recovery. In Foster, R. (ed.) The Oxford Illustrated History of Ireland, 53-103. Oxford: Oxford University Press. Stephenson, A. 2009. Wining and dining in a medieval village at Mullaghmast, Co. Kildare. In Stanley, Danaher & Eogan (eds) Dining and Dwelling, Archaeology and the National Roads Authority Monograph Series 6, 143-154. Bray: Wordwell. Stalley R.A. 1987. The Cistercian Monasteries of Ireland. London and New Haven: Yale University Press. Stalley R.A. 2005. Ecclesiastical architecture before 1169. In Ó Cróinín, D. (ed.), A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 714-743. Oxford: Oxford University Press. Stout, Matthew 1997. The Irish Ringfort, Dublin: Four Courts. Stout, G. & Stout, M. 2008. Excavation of an Early Medieval Secular Cemetery at Knowth Site M, County Meath. Bray: Wordwell. Swan, L 1983. Enclosed ecclesiastical sites and their relevance to settlement patterns of first millennium AD. In Reeves-Smyth & Hammond (eds) Landscape Archaeology in Ireland, 269-94. BAR British series 116 Swan, L. 1985. Monastic proto-towns in early medieval Ireland: the evidence of aerial photography, plan analysis and survey. In Clarke & Simms (eds) The Comparative History of Urban Origins in Non-Roman Europe: Ireland, Wales, Denmark, Germany, Poland and Russia from the Ninth to the Thirteenth Century, 77-102. BAR International Series 255. Sweetman, P.D. 1978a. Archaeological excavations at trim Castle, Co. Meath, 1971- 4. Proceedings of the Royal Irish Academy, 78C: 127-98. Sweetman, P.D. 1978b. Excavation of medieval “field boundaries” at Clonard, County Meath. Journal of the Royal Society of Antiquaries of Ireland, 108: 10-22. 155 Sweetman, P.D. 1979. Archaeological excavations at Ferns Castle, County Wexford. Proceedings of the Royal Irish Academy, 79C: 217-45. Sweetman, P.D. 1980a. Archaeological excavations at Adare Castle, County Limerick. Journal of the Cork Historical and Archaeological Society, 85: 1-6. Sweetman, P.D. 1980b. Archaeological excavations at King John’s Castle, Limerick. Proceedings of the Royal Irish Academy, 80C: 207-29. Sweetman, P.D. 1999. The Medieval Castles of Ireland. Cork: Collins Press. Swift, C. 1998. Forts and fields: a study of ‘monastic towns’ in seventh and eight century Ireland. Journal of Irish Archaeology, 9: 105-125. Tilly, C. 1992. Coercion, Capital, and European States, AD 1990-1990, Revised Edition. Oxford: Blackwell. Trigger, B. 1984. Alternative archaeologies: Nationalist, colonialist, imperialist. Man, New Series, 19.3: 355-370. Valente, M. 1998. Reassessing the Irish ‘monastic town. Irish Historical Studies 31.121: 1-18. Valente, M. 2008. The Vikings in Ireland. Dublin: Four Courts Press. Veblen, T. 1925. The Theory of the Leisure Class. Oxford: Oxford University Press. Wallace, P.F. 1985a. The archaeology of Viking Dublin. In Clarke & Simms (eds) The Comparative History of Urban Origins in Non-Roman Europe: Ireland, Wales, Denmark, Germany, Poland and Russia from the Ninth to the Thirteenth Century, 103-146. BAR International Series 255. Wallace, P.F. 1985b. The archaeology of Anglo-Norman Dublin. In Clarke & Simms (eds) The Comparative History of Urban Origins in Non-Roman Europe: Ireland, 156 Wales, Denmark, Germany, Poland and Russia from the Ninth to the Thirteenth Century, 379-410. BAR International Series 255. Wallace, P.F. 1992. The Viking age buildings of Dublin. Part 1, Text. Dublin: Royal Irish Academy. Wallace, P.F. 2005. The archaeology of Ireland’s Viking Age towns. In Ó Cróinín, D. (ed.) A New History of Ireland, Vol. I: Prehistoric and Early Ireland, 814-841. Oxford: University Press. Walsh, A. 1998. A summary classification of Viking Age swords in Ireland. In Clarke, Ní Mhaonaigh & Ó Floinn (eds) Ireland and Scandinavia in the Early Viking Age, 222-235. Dublin: Four Courts Press. Watt, J.A. 1972. The Church in Medieval Ireland, Dublin: Gill & Macmillan. Watt, J.A. 1987a. Approaches to the history of fourteenth century Ireland. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 303-313. Oxford: Oxford University Press. Watt. J.A. 1987b. Gaelic polity and cultural identity. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 314-351. Oxford: Oxford University Press. Watt. J.A. 1987c. The Anglo-Irish colony under strain, 1327-99. In Cosgrove, A. (ed.) A New History of Ireland, Vol. II: Medieval Ireland 1169-1534, 352-396. Oxford: Oxford University Press. Wiggins, K. 1990. Excavations at King John’s Castle, Limerick. In Bennet, I. (ed.) Excavations 1990, available at http://www.excavations.ie/Pages/Details.php?Year=&County=Limerick&id=3373 Accessed 14/06/2010 157 Wiggins, K. 1994. Excavations at King John’s Castle, Limerick. In Bennet, I. (ed.) Excavations 1994, available at http://www.excavations.ie/Pages/Details.php?Year=&County=Limerick&id=2684 Wiggins, K. 1998. Excavations at King John’s Castle, Limerick. In Bennet, I. (ed.) Excavations 1998, 135-6. Bray: Wordwell. Wilson, D. 1985. Moated Sites. Aylesbury: Shire Publications. Young, R.J.C 2003. Postcolonialism: A Very Short Introduction. Oxford: Oxford University Press. 158