We'll have what they're having, cultural identity through diet in the English Saxon Period

Transcription

1 Environmental Archaeology The Journal of Human Palaeoecology ISSN: (Print) (Online) Journal homepage: We'll have what they're having, cultural identity through diet in the English Saxon Period Matilda Holmes To cite this article: Matilda Holmes (2016) We'll have what they're having, cultural identity through diet in the English Saxon Period, Environmental Archaeology, 21:1, 59-78, DOI: / Y To link to this article: Published online: 18 Mar Submit your article to this journal Article views: 265 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at Download by: [University of London] Date: 10 June 2016, At: 02:01

2 We ll have what they re having, cultural identity through diet in the English Saxon Period Matilda Holmes University College London, London The roots of the English population lie in diverse cultural origins. Within the second half of the first millennium AD, there were two major migrations, firstly the Saxons in the immediate post-roman period, and later following the Viking incursions of the mid-9th century. This paper considers the visibility of these migrations in the animal economy. Zooarchaeological data from 141 Saxon sites within England were analysed for evidence of diet and animal husbandry. Results indicate that there were signatures specific to native British rural populations in the early Saxon phase, relying largely on a self-sufficient economy, consuming very few wild species or domestic birds. Saxon cultural differences were implied from a number of sites, dependant on the increased importance of cattle in the diet and as culturally symbolic animals signifying status and utilised for trade. There was also evidence for the earliest settlers to have used pigs as a mobile larder, to provide meat while they established herds and flocks of cattle and sheep. Regional differences were visible in the late Saxon phase, within Wessex and the Danelaw, related to agricultural changes in the midlands and central southern region, and cultural identity, particularly within the heartlands of the northern Danelaw, towards an increase in cattle production. Keywords: Saxon, Viking, Diet, Culture, Identity, zooarchaeology Introduction Much work has been done to identify new cultures as they appeared in the archaeological record during the Saxon period, including investigations into settlement patterns, burial practices, material culture, ornamentation, documentary sources, DNA and isotope analysis (Hills 2011). However, little research has been carried out into the possible reflection of the cultural identity through the animal economy of Britons, Saxons and Vikings within England. This study aims to consider what impact the removal of Roman control, and subsequent influx of Angles, Saxons and Jutes in the early Saxon phase and Scandinavians in the late Saxon phase had on the consumption of animals and their products. Diet has been chosen as a suitable medium to investigate changes in cultural identity, as it largely results from numerous local, quotidian processes of production, distribution, consumption and disposal, all of which may be expected to be influenced by traditional cultural practices, or through necessity resulting from the move to a new land (e.g. Hammon 2011, 278). Correspondence to: Matilda Holmes, Consultant Archaeozoologist, 45, Welland Rise, Sibbertoft, Leics LE16 9UD, UK. matty@archaeozoology.co.uk In the late 5th century the Roman population dwindled as political, military and administrative systems collapsed when Rome failed to support the western edge of the Empire. Various theoretical models exist for the nature of the Roman decline and development of Anglo-Saxon England by the 7th century, and while it is not within the scope of this paper to explore in detail the highly contentious mechanics of this process, several relevant perspectives exist regarding the nature of society and the Saxon migration in the 5th and 6th centuries: Absolute collapse of the Roman state leading to a reversion of the remaining population to small, tribal groups reminiscent of the pre-roman Iron Age with subsequent colonisation by Saxon populations from the east (e.g. Esmonde Cleary 2011). A move of power from the state to a Roman military elite who took regional control through social power and links to the past, with long-term, small-scale diasporas of Saxon settlers (e.g. Moreland 2010, 2011). Comparable land use to that of Roman agriculture across much of western Europe in the early post- Roman phase, although production was increasingly localised, with greater diversification of husbandry dependant on the environment, and a collapse in the market economy (e.g. Crabtree 1991; Esmonde Cleary 1995; Lewit 2009, 82). Association for Environmental Archaeology 2016 DOI / Y Environmental Archaeology 2016 VOL. 21 NO. 1 59

3 Continuity of a dominant Romano-Christian culture and associated supply networks by the remaining native population, with Germanic settlers integrating into eastern regions alongside the Romano-British. A distinction that was gradually to fade, leaving behind a consolidated Germanic language and Anglo-Saxon culture based on a Frankish Christian model (e.g. Dark 2000; Hammon 2011). The De Excidio, composed by the monk Gildas in the 6th century describes the presence of Saxon mercenaries, and the writings of Bede (albeit writing centuries later) indicate that this migration occurred in the eastern region of the country. While there are considerable biases from these early writings, more detailed archaeological information from artefacts and burials have shown that early Saxon settlement took place throughout much of east Yorkshire, East Anglia, Lincolnshire, the midlands, and the south eastern coast, migrating along rivers from coastal areas (Dark 2000, 68, figure 23; Hamerow 2011a, 120; Jones and Mattingly 1990, ). The Saxon migration did not continue into areas of western Britain which may have retained a Romano- Christian culture, with links to the Mediterranean and Europe (Hammon 2011). However, by the end of the early Saxon phase, in the early part of the 7th century, England was largely Anglo-Saxon, greater wealth and social hierarchy became evident and greater from former Roman towns, high-status sites such as Yeavering, and wealthy burials such as those seen at Sutton Hoo (Hamerow 2011b, 159; Holmes 2014a, 61; Moreland 2011, 184; Reynolds 1999, 51). Even within areas of Saxon settlement it is likely that there remained a considerable native British presence. Recent isotope and DNA studies are beginning to show that there is no evidence for a removal of the indigenous population or a mass-migration of whole settlement populations (Budd et al. 2004; Hughs et al. 2014; Montgomery et al. 2005). A study into the origins of the modern British population (Winney et al. 2012) has shown the substantial presence of Anglo-Saxon DNA in those from eastern counties, alongside a considerable amount of ancient-british DNA. Furthermore, the mixing of native and Anglo-Saxon burial customs in eastern and southern areas (Dark 2000, 72) suggests that, even if diet was culturally specific, results from this area are likely to contain both British and Saxon data. The following three centuries saw an increase in international trade and organisation of numerous early Saxon kingdoms into the more politically stable kingdoms of Mercia, Wessex, East Anglia and Northumbria, Essex, Kent and Sussex. The next round of mass migration began with the Viking raids of the 9th century, leading to an established Anglo- Scandinavian population within England by the time of the Norman Conquest. A tentative peace between the Saxon and Viking cultures was established with the division of the country in AD 878 into the Danelaw in the north and east, and the continuation of Saxon kingdoms of Mercia (now reduced) and Wessex (Reynolds 1999, 74, figure 24). As with the Saxon migration, there is much debate regarding the nature of the Danish invasion, but research into isotope and DNA makeup of burials within the Danelaw and Anglo-Saxon England again suggests that there was no mass genocide of the existing population, but an integration with burials showing a mix of Viking and Saxon traits (Budd et al. 2004, 136), a suggestion that is also borne out by other archaeological sources (Hadley 2002, 77; Richards 2007, 2011, 48). Evidence also suggests that, while Scandinavian settlers retained some aspects of identity such as ornamentation (Hadley 2002, 65) they were skilled at appropriating aspects of local language, culture and behaviour (Richards 2011, 58), making their method of integration with the native population largely one of acculturation. Methodology The mixing of native and Saxon; and Saxon and Danish cultures within areas of migration makes understanding the animal economy more difficult, as it can be hard to identify individual settlements as belonging to one or the other cultural population. Nonetheless, broad regions have been specified in Table 1. It must also be remembered that regional boundaries were most likely fluid, particularly given the long chronology between phases (Hooke 1998, 45). As differences between settlements belonging to the secular and ecclesiastical elite and those of lower status have been recognised elsewhere (Crabtree 2012; Holmes 2014a; Sykes 2007), analysis will include only lower-status rural, wic and burh sites. Site classifications were taken from the original site report. Although there was no major migration in the middle Saxon phase, a brief consideration of the archaeozoology of the major regions within this phase will be given to provide background to late Saxon developments. One hundred and fifty animal bone assemblages were included from 141 urban and rural sites throughout England (Table 2). Data were collected from published site reports, grey literature and primary data recorded by the author. Assemblages with over 100 fragments of cattle, sheep and pig bone were included, from sites dated to the early (AD ), middle (AD ) and late (AD ) Saxon phases. Dietary identity will be investigated through differences in the proportions of the major domesticates (cattle, sheep and pig), domestic birds (domestic fowl, ducks and geese) and wild species (birds, deer 60 Environmental Archaeology 2016 VOL. 21 NO. 1

4 Table 1 List of counties within culturally distinct regions for which animal bone reports are included, based on Dark (2000, figure 23) and Reynolds (1999, figures 17 and 24) Phase Region Counties Early Saxon Buckinghamshire, Cambridgeshire, Essex, Hampshire, Kent, Lincolnshire, London, Norfolk, Northamptonshire, Oxfordshire, Suffolk, Surrey, Sussex, east Yorkshire Briton Dorset, Gloucestershire, Leicestershire, Somerset, Wiltshire, Worcestershire Middle Northumbria Durham, Yorkshire Late East Anglia Mercia Wessex Danelaw (north) Danelaw (Mercia) Danelaw (east) Anglo-Saxon Mercia Anglo-Saxon Wessex Norfolk, Suffolk, Essex Buckinghamshire, Cheshire, Gloucestershire, Northamptonshire, Oxfordshire, Lincolnshire Kent*, Berkshire, Hampshire, London, Sussex*, Wiltshire Durham, Yorkshire Buckinghamshire, Cambridgeshire, Northamptonshire Lincolnshire, Norfolk, Suffolk Cheshire, Gloucestershire, Oxfordshire, Staffordshire, Warwickshire Berkshire, Cornwall, Devon, Hampshire, Kent, London, Somerset, Sussex and Wiltshire *Although these counties were independent middle Saxon Kingdoms, they are only represented by isolated sites, so are included within their larger neighbouring Kingdoms. and hare) based on a comparison of the fragment counts antler fragments were not included in counts of deer. Identification of domestic and wild ducks and geese were taken from each site report, and where none was given they were included in the count of wild birds. While analysis of the major domesticates (cattle, sheep/goat and pig) will consider those from all sites, comparison of wild taxa will only include sites with over 300 NISP of the major domesticates in an attempt to eliminate the bias caused by limited sample size on the number of taxa recovered (Casteel 1979; Lyman 2008, Chapter 5). Differences in animal husbandry will be based on mortality profiles of cattle, sheep and pigs from tooth wear data (from Hambleton 1999). Maps were produced using Qgis (Quantum GIS Development Team 2014), and superficial and drift geology taken from the British Geological Survey (2014). This paper will present the results of analysis by phase in the subsequent three sections, and possible causes and implications for the identification of culturally distinct populations will be discussed in the section on cultural change and dietary identity. Native Britons and Early Saxon Saxons Before considering differences in animal economy between regions likely to be culturally different, the role of local geology must be considered. As Fig. 1 shows, all but two sites located on the chalk contained more than 30% of sheep. Sheep are traditionally suited to the chalk downland environment, preferring welldraining soil, which is also the case of the sites on sand and gravel. However, high numbers of sheep were also recorded on clay, and there was no preference for cattle on any geology. So although the environment may have played a small part in the determination of which animals to keep, there was no definitive evidence for environmental determinism in the early Saxon phase. To give a more nuanced understanding of the possible effects of Saxon settlers on the underlying animal economy sites were split into those dated to the early part of the phase (5th 6th centuries), when the Saxon migration began; and those from less well dated or later settlements (Fig. 2). The most notable product of this was that sites with the highest proportion of pig were from within the early Saxon area, most notably: West Stow, Suffolk (22 and 33%); Kings Meadow Lane, Higham Ferrers, Northamptonshire (20%); Higham Ferrers, Northamptonshire (29%); Barton Court Farm, Oxfordshire (27%) and Botolphs, Bramber, Sussex (37%). Saxon County School, Surrey also had a high proportion of pig remains (31%), a site that may be within the early part of the period, although no firm dating was available. The range of species kept at later sites in the Saxon area and those from all western British sites were generally tightly grouped, typically keeping between 35 and 65% cattle, 25 and 55% sheep and 5 and 20% pigs. The small group of outliers with exceptionally high numbers of cattle (>60%) include Hartigans, Milton Keynes (86%) within the British area, and Nettleton Top, Lincolnshire (84%); Spong Hill, Norfolk (87%) and Fossets Farm, Southend (80%) in the Saxon region. The mean values were also plotted to give a better idea of the relationship between the three main domesticates (Fig. 3), which clearly shows the abundance of pigs in the early Saxon phase, alongside fewer cattle, which is not observed at contemporary British sites. At later (or less well dated) settlements the proportion of pig remains falls within the Saxon region, contrasting with an increased number of cattle, while in the British region cattle decrease in line with a rise in sheep numbers. When the minor species were considered (Table 3), nearly all sites in the native, western region had fewer than 3% domestic birds and very few wild species, though at isolated sites unusually high numbers were recorded. A group of sites within the Saxon area recorded over 5% domestic fowl in their assemblages, and further sites with relatively high numbers of domestic birds (i.e. 3 5%) were also Environmental Archaeology 2016 VOL. 21 NO. 1 61

5 Table 2 Sites included within the data set County Area Reference Early Saxon Hartigans, Milton Keynes Buckinghamshire Native (early) Burnett (1993) Pennyland, Milton Keynes Buckinghamshire Native Holmes (1993) Walton vicarage, Aylesbury Buckinghamshire Native Noddle (1976) Poundbury, Dorchester Dorset Native Buckland-Wright (1987) Barnsley Park, Gloucestershire Gloucestershire Native (early) Noddle, 1985a Sherborne House, Lechlade Gloucestershire Native Maltby (2003) Old Down Farm, Andover Hampshire Native (early) Bourdillon (1980) Empingham west, Rutland water Leicestershire Native (early) Morrison (2000) Eye Kettleby, Leicestershire Leicestershire Native (early) Knight (forthcoming) Nettleton Top, Lincolnshire Lincolnshire Saxon Berg (1993) Quarrington, Lincolnshire Lincolnshire Saxon Rackham (2003) Market Lavington, Wiltshire Wiltshire Native (early) Bourdillon (2006) Deansway, Worcester Worcestershire Native (early) Nicholson and Scott (2004) Orton Hall Farm, Cambridgeshire Cambridgeshire Saxon (early) King (1996) Spicer s Warehouse, Sawston Cambridgeshire Saxon Holmes (2009b) Stonea grange, Cambridgeshire Cambridgeshire Saxon Stallibrass (1996a) Fossets Farm, Southend Essex Saxon Grimm (2007) Manston rd, Ramsgate Kent Saxon Hamilton-Dyer (1997) Distillery site, Hammersmith London Saxon Ainsley et al. (2008) Harlington, London London Saxon Grimm (2009) Melford Meadows, Brettenham Norfolk Saxon (early) Powell and Clark (2002b) Mundham, Norfolk Norfolk Saxon Leach and Morris (2008) Redcastle Furze, Thetford Norfolk Saxon Wilson (1995) Spong Hill, Norfolk Norfolk Saxon Bond (1995) Kings Meadow lane, Higham Ferrers Northamptonshire Saxon (early) Albarella and Johnstone (2000) Higham Ferrers, Northamptonshire Northamptonshire Saxon (early) Evans (2007) Aelfric s Abbey, Eynsham, Oxfordshire Oxfordshire Saxon Ayres et al. (2003) Audlett drive, Abingdon Oxfordshire Saxon Levitan (1992) Barton Court Farm, Abingdon Oxfordshire Saxon (early) Wilson et al. (1986) Mill st, Wantage Oxfordshire Saxon Maltby (1996) New Wintles, Oxfordshire Oxfordshire Saxon Noddle (1975) Oxford Science park, Littlemore Oxfordshire Saxon Ingrem (2001) St Helen s Avenue, Benson Oxfordshire Saxon Hamilton-Dyer (2004a) West Stow Suffolk Saxon (early) Crabtree (1989) Saxon County School, Shepperton Surrey Saxon (early) Ayres (2005) Botolphs, Bramber, Sussex Sussex Saxon (early) Stevens (1990) Caythorpe pipeline, North Humberside Yorkshire Native Stallibrass (1996b) Middle Saxon Chalkpit Field North, Sedgeford Norfolk East Anglia Poole (2007) Crow hall park, Downham Market Norfolk East Anglia Curl (2008) Hay Green, Terrington St Clement Norfolk East Anglia Baker (2002) Rose Hall Farm, Walpole St Andrew Norfolk East Anglia Baker (2002) Sedgeford, Norfolk Norfolk East Anglia Clutton-Brock (1976) Ipswich Suffolk East Anglia Jones and Serjeantson (1983) Ipswich Suffolk East Anglia Crabtree, (1994) Site 127 Bury St Edmunds Suffolk East Anglia Murphy (1996) Chicheley, Buckinghamshire Buckinghamshire Mercia Jones (1980) Walton Lodge, Aylesbury Buckinghamshire Mercia Sadler (1989) Hunter Street School, Chester Cheshire Mercia Cartledge (1994) Quarrington, Lincolnshire Lincolnshire Mercia Rackham (2003) Marefair, Northampton Northamptonshire Mercia Harman (1979a) Saxon palaces, Northampton Northamptonshire Mercia Harman (1985b) St Peters Rd, Northampton Northamptonshire Mercia Harman (1979b) Cresswell Field, Yarnton Oxfordshire Mercia Mulville et al. (2004) The Orchard, Walton Rd, Aylesbury Oxfordshire Mercia Hamilton-Dyer (2004c) Worton, Yarnton Oxfordshire Mercia Mulville et al. (2004) Yarnton Oxfordshire Mercia Mulville et al. (2004) Cottam, Yorkshire Yorkshire Northumbria Dobney et al. (1999) Fishergate, York Yorkshire Northumbria O Connor (1991) Site 39, Wharram Yorkshire Northumbria Stevens (1992) Sites 94 and 95, Wharram Yorkshire Northumbria Pinter-Bellows (1992) The south manor area, Wharram Yorkshire Northumbria Pinter-Bellows, 2000 Lot s Hole, Berkshire Berkshire Wessex Powell and Clark (2002b) Riverdene, Basingstoke Hamphsire Wessex Hamilton-Dyer (2003) Anderson s road, Southampton Hampshire Wessex Knight (2006) Cook St, Southampton Hampshire Wessex Bourdillon (1993) Friend s Provident, Southampton Hampshire Wessex Hamilton-Dyer (2005) Melbourne St, Southampton Hampshire Wessex Bourdillon and Coy (1980) Six Dials, Hamwic Hampshire Wessex Bourdillon and Andrews (1997) Church Lane, Canterbury Kent Wessex King (1982) Continued 62 Environmental Archaeology 2016 VOL. 21 NO. 1

6 Table 2 Continued County Area Reference Sandtun, West Hythe Kent Wessex Murray and Hamilton-Dyer (2001) Maiden Lane and 6 7 Exchange Court London Wessex Hamilton-Dyer (2004b) James Street, London London Wessex Armitage (2004a) Jubilee Hall, Covent Garden London Wessex West (1988) Lyceum Theatre, Exeter Street, London London Wessex Rackham and Snelling (2004) Maiden Lane, London London Wessex West (1988) National Gallery Basement London Wessex West (1989b) National Gallery Extension London Wessex Rackham (1989) National Portrait Gallery London Wessex Armitage (2004a, 2004b) Peabody site, London London Wessex West (1989a, 1989b) The Treasury, Whitehall, London London Wessex Ainsley et al. (2008) Friars Oak, Hassocks Sussex Wessex Stevens (2000) Cadley rd, Collingbourne Ducis Wiltshire Wessex Hamilton-Dyer (2001) Late Saxon Brandon Rd, Thetford Norfolk Eastern Danelaw Jones, 1993 Bury Road, Thetford Norfolk Eastern Danelaw Grimm (2006) Chalkpit Field North, Sedgeford Norfolk Eastern Danelaw Poole (2007) Danesgate, Lincoln Lincolnshire Mercian Danelaw Holmes (2003) Flaxengate, Lincoln Lincolnshire Mercian Danelaw O Connor (1982) Ipswich Suffolk Eastern Danelaw Jones and Serjeantson (1983) Ipswich Suffolk Eastern Danelaw Crabtree (1994) Knocker s site, Thetford Norfolk Eastern Danelaw Jones (1984) Lincoln Lincolnshire Mercian Danelaw Dobney et al. (1996) Mill Lane, Thetford Norfolk Eastern Danelaw Albarella (1999) St Nicholas St, Thetford Norfolk Eastern Danelaw Hutton MacDonald (1999) Whitefriars car park, Norwich Norfolk Eastern Danelaw Cartledge (1983) High st, Oxford Oxfordshire Mercia Maltby (2000) Abbey Green, Chester Cheshire Mercia Cartledge (1994) All Saints Church, Oxford Oxfordshire Mercia Wilson (2003) Church Rd, Bishop s Cleeve Gloucestershire Mercia Lovell et al. (2007) Crook Street, Chester Cheshire Mercia Cartledge (1994) Goss Street, Chester Cheshire Mercia Cartledge (1994) Hinxey Hall, Queen st, Oxford Oxfordshire Mercia Wilson et al. (1983) Hunter s Walk, Chester Cheshire Mercia Cartledge (1994) St Aldates, Oxford Oxfordshire Mercia Armour-Chelu (2003) St Ebbes, Oxford Oxfordshire Mercia Wilson et al. (1989) Trill Mill Stream, Oxford Oxfordshire Mercia Wilson (2003) Winchcombe, Gloucestershire Gloucestershire Mercia Levitan (1985) Black Lion Hill, Northampton Northamptonshire Mercian Danelaw Harman (1985a) Chalk Lane, Northampton Northamptonshire Mercian Danelaw Harman (1981) Langham Rd and Burystead, Raunds Northamptonshire Mercian Danelaw Davis (2009) Marefair, Northampton Northamptonshire Mercian Danelaw Harman (1979a, 1979b) Higham Ferrers, Northamptonshire Northamptonshire Mercian Danelaw Evans (2007) Walton vicarage, Aylesbury Buckinghamshire Mercian Danelaw Noddle (1976) West Cotton, Raunds Northamptonshire Mercian Danelaw Albarella and Davis (1994) Blue bridge lane, York Yorkshire Northern Danelaw Rowland (2005) Coppergate, York Yorkshire Northern Danelaw O Connor (1989) Micklegate, York Yorkshire Northern Danelaw O Connor (2004) Ribblehead Yorkshire Northern Danelaw Rackham (1977) Skeldergate, York Yorkshire Northern Danelaw O Connor (1984) St Saviourgate, York Yorkshire Northern Danelaw O Connor (2004) Tenements, Durham City Durham Northern Danelaw Rackham (1979) The south manor area, Wharram Yorkshire Northern Danelaw Pinter-Bellows (2000) Walmgate, York Yorkshire Northern Danelaw O Connor (2004) Wearmouth and Jarrow Durham Northern Danelaw Noddle et al. (2006) Maiden Lane and 6 7 Exchange Court London Wessex Hamilton-Dyer (2004b) 27, Jewry Street, Winchester Hampshire Wessex Bourdillon (2009) Benham s Garage, Taunton Devon Wessex Levitan (1984a) Bristol Castle Somerset Wessex Noddle (n.d.) Canterbury Lane, Canterbury Kent Wessex Marples (1983) Chester Rd, Winchester Hampshire Wessex Bourdillon (2009) Citizen house, Bath Somerset Wessex Grant (1979) Easton Lane, Winchester Hampshire Wessex Maltby (1989) Market field, Steyning Sussex Wessex O Shea (1993) Market Lavington, Wiltshire Wiltshire Wessex Bourdillon (2006) Mary-Le-Port, Bristol Somerset Wessex Noddle (1985a, 1985b) Mawgan Porth, Cornwall Cornwall Wessex Clutton-Brock (1976) Staple Gardens, Winchester Hamphsire Wessex Holmes (2009a, 2009b) Steyning Sussex Wessex Sykes (2007) Ufton Nervet, Berkshire Berkshire Wessex Westley (1974) Victoria Rd, Winchester Hampshire Wessex Bourdillon (2009) Western Suburb, Winchester Hampshire Wessex Coy (2009) Environmental Archaeology 2016 VOL. 21 NO. 1 63

7 Figure 3 Mean proportions of the main domesticates by region and subphase (early = sites dated to the 5th 6th century) for the early Saxon phase. Figure 1 Relative proportions of the major domesticates by underlying geology for the early saxon phase. observed. Wild mammal remains were also slightly more common in the Saxon area. Wild birds were uncommon on all sites, with nothing to suggest that they were targeted by either British or Saxon cultures. Evidence for animal husbandry came from mortality profiles that were largely similar for both regions, with the majority of cattle and all sheep and pigs culled for their meat. It is notable, however, that at all sites a greater number of cattle were kept longer (past wear stage G) in the Saxon region, this is particularly notable at Fossets Farm and Oxford Science Park where the majority were alive before reaching wear stage G (Fig. 4). Middle Saxon Stabilisation In order to set the data from the Germanic and Danish migrations in context, middle Saxon sites were also included in this analysis as a control group from a time where no major influx of populations occurred. Figure 2 Relative proportions of the major domesticates by region and subphase (early = sites dated to the 5th 6th century) for the early Saxon phase. It has previously been observed that middle Saxon settlements were more likely to farm sheep on the chalk downlands of the east and south (Holmes 2012, 58; O Connor 2010, 8), and this is borne out by this data set (Fig. 5), alongside higher numbers of cattle on clay. The role of environmental determinism must therefore be considered as an influential factor on the animal economy of middle Saxon settlements. However, a predominance of cattle at wics and corresponding high number of sheep on rural sites has also been appreciated elsewhere (Holmes 2013; Holmes 2012; O Connor 2010) illustrated in Fig. 6. When considered by region (Fig. 7), there are still patterns to be observed, but these are largely due to the effects of geology and site type described above. In particular, low numbers of pigs (4 11%) were noted on sites in the north, which may reflect relatively low woodland coverage of that area (Roberts and Wrathmell 2000, 31, figure 24), although a recent study of the palaeoecology of the northwest region has suggested that woodland coverage was highly variable in that area (Forster 2010). The predominance of sheep at sites in East Anglia and Northumbria (ca %) may be more a reflection of the local geology and redistribution of animals than regional preferences, as the majority of sites in that area were rural settlements situated on chalk. Correspondingly, greater numbers of cattle were evident in Wessex (ca %), the majority of sites in this case being on clay, and either wic or trading sites, again suggesting that geology or modes of production were the major causal factors rather than regional distinction. There were no regional differences in the proportion of wild birds or mammals or domestic birds (Table 4). Figure 8 shows the mean proportions of major domesticates to help better understand the relationship between them, where it is obvious that the depressed number of sheep in Wessex is even greater, given the higher numbers of both cattle and pigs, again the predominance of both these taxa are more common on urban sites (Holmes 2014a, 100). 64 Environmental Archaeology 2016 VOL. 21 NO. 1

8 Table 3 Proportions of domestic birds, wild birds and wild mammals recorded at early Saxon sites (as a % of the total number cattle, sheep and pig) Site Region Wild bird Domestic bird Wild mammal Caythorpe pipeline, North Humberside Native Pennyland, Milton Keynes Native Poundbury, Dorchester Native Sherborne House, Lechlade Native Walton vicarage, Aylesbury Native Barnsley Park, Gloucestershire Native early Deansway, Worcester Native early Eye Kettleby, Leicestershire Native early Market Lavington, Wiltshire Native early Aelfric s Abbey, Eynsham, Oxfordshire Saxon Fossets Farm, Southend Saxon Manston rd, Ramsgate Saxon Mundham, Norfolk Saxon Nettleton Top, Lincolnshire Saxon 0 New Wintles, Oxfordshire Saxon Oxford Science park, Littlemore Saxon Quarrington, Lincolnshire Saxon Redcastle Furze, Thetford Saxon Saxon County School, Shepperton Saxon Spicer s Warehouse, Sawston Saxon Spong Hill, Norfolk Saxon St Helen s Avenue, Benson Saxon Stonea grange, Cambridgeshire Saxon West Stow Saxon Barton Court Farm, Abingdon Saxon early Melford Meadows, Brettenham Saxon early Higham Ferrers, Northamptonshire Saxon early Orton Hall Farm, Cambridgeshire Saxon early West Stow Saxon early West Stow Saxon early The beginning of specialisation of animal husbandry in the hinterlands of wics, as the demand for food and raw materials became increasingly important, is also a factor in the understanding of regional differences in this phase. Animals kept for secondary products and deliberately marketed for meat to wics from rural producer sites become evident in the areas surrounding the eastern and southern seaboard. Inland areas do not show signs of such demand for specialisation in the absence of dedicated consumer sites in the region (Holmes 2013). Late Saxon Vikings The apparent effect of the environment observed in the middle Saxon phase did not continue at late Saxon settlements (Fig. 9), with the majority of data indicating a wide range of animals kept on both clay and Figure 4 Wear stages of cattle teeth in the early Saxon phase. 1 = Fossets Farm, Southend; 2 = Oxford Science Park. Figure 5 The relative proportions of cattle, sheep and pig at sites on various geologies in the middle Saxon phase. Environmental Archaeology 2016 VOL. 21 NO. 1 65

9 Figure 6 Proportions of the main domesticates at wics and rural sites in the middle Saxon phase. chalk. However, it should be noted that the greatest proportions of cattle came from the settlements of York and Chester, on clay, which is likely to be a reflection of the marketing of cattle to urban sites, rather than the preferential keeping of cattle within the town, although it could indicate a preference of cattle, either socially or economically in the hinterland (see discussion). There was also less distinction in the proportions of major domesticates on urban and rural sites than the preceding phase (Fig. 10), although sites with the greatest numbers of sheep continued to be rural in nature, and the majority of those with over 50% cattle were urban. When regional differences were considered (Fig. 11) Wessex sites contained some of the greatest numbers of sheep (ca %) compared to the Danelaw (ca %) and Mercia (ca. 7 55%) and least cattle. Mercian sites generally had high proportions of pigs, while sites from the Danelaw were more varied, consisting of a wide range of species. To investigate this Figure 7 Relative proportions of the major domesticates by region for the middle Saxon phase. further, the Viking sites were subdivided into those from the north (around the Danish capital of York including Yorkshire and Durham); east (Norfolk, Suffolk and Cambridgeshire) and former Mercia (Lincolnshire, Northamptonshire, Buckinghamshire and London). Results showed a considerable difference between the three areas (Fig. 12), split between cattle in the north and sheep in the midlands, with less distinction apparent in the eastern part of the Danelaw. Pigs were also recorded in some of the lowest numbers on many northern Danelaw sites. Fig. 13 illustrates the relative proportions of cattle, sheep/ goat and pigs to help understand changes between them, from which the lowest numbers of sheep and pigs are seen to be in the northern Danelaw, offset by very high numbers of cattle, from rural and urban sites alike. Wessex was the only region where sheep were predominant, with some of the lowest numbers of cattle and pigs again coming from both rural and urban sites. Proportions of the main domesticates were similar at midland and eastern sites, the greatest difference observed in the Mercian Danelaw sites, where sheep were more common at the expense of both cattle and pigs. There was little to suggest any regional patterns in the distribution of domestic birds or wild animals (Table 5), and too few sites provided mortality data to make any reliable comparison of regional differences. Cultural Change and Dietary Identity As already noted, the investigation of cultural identity on the scale attempted here is made more difficult given the disparate nature of both Saxon and Scandinavian groups. The Saxons included Franks, Angles, Jutes, Saxons and southern Scandinavians; and the Vikings originated from various areas of Scandinavia (Denmark, Norway, Sweden, Jutland and Finland) all of whom would have seen themselves as culturally independent (Richards 2011, 48). Furthermore, it is likely that Saxons and Vikings did not take over a region wholesale, but that pockets of existing settlement remained, influencing the settlers to varying degrees (Loveluck and Laing 2011, 12). This discussion will make some broad interpretations of the data presented above, with particular reference to the modes of Saxon migration summarised in the introduction. There is nothing from this, or other studies to suggest a wholesale collapse of the Roman economy. Continuity between Roman and early Saxon features has been observed at Wroxeter (Hammon 2011), and higher numbers of cattle at Eye Kettleby and Hartigans, Melford Meadows and Orton Hall Farm in the earlier part of the period implies some continuation of the Roman economy (King 2001, 219), which 66 Environmental Archaeology 2016 VOL. 21 NO. 1

10 Table 4 Proportions of domestic birds, wild birds and wild mammals recorded at middle Saxon sites (as a % of the total number cattle, sheep and pig) Site name Area Wild mammal Domestic bird Wild bird Ipswich East Anglia Ipswich East Anglia Rose Hall Farm, Walpole St Andrew East Anglia Sedgeford, Norfolk East Anglia 0 Cresswell Field, Yarnton Mercia Hunter Street School, Chester Mercia Marefair, Northampton Mercia Quarrington, Lincs Mercia Saxon palaces, Northampton Mercia St Peters Rd, Northampton Mercia The Orchard, Walton Rd, Aylesbury Mercia Walton Lodge, Aylesbury Mercia 0.26 Yarnton Mercia Cottam, Yorkshire Northumbria 0.51 Fishergate, York Northumbria Site 39, Wharram Northumbria Sites 94 and 95, Wharram Northumbria The south manor area, Wharram Northumbria Maiden Lane and 6 7 Exchange Court Wessex Anderson s road, Southampton Wessex Cadley rd, Collingbourne Ducis Wessex Cook St, Southampton Wessex Friars Oak, Hassocks Wessex Friend s Provident, St Mary s Stadium, Southampton Wessex James Street, London Wessex Jubilee Hall, Covent Garden Wessex Lot s Hole Wessex Lyceum Theatre, Exeter Street Wessex Maiden Lane Wessex Melbourne St, Southampton Wessex National Gallery Basement Wessex National Gallery Extension Wessex National Portrait Gallery Wessex Peabody site Wessex Only sites with a NISP > 300 cattle, sheep and pig recorded were included. may also be observed at certain eastern sites in the later phase. Other trends such as the maintenance of larger animals in the early Saxon phase at some sites in proximity to Roman settlements (Albarella et al. 2008; Holmes 2014b), and the presence of a highstatus elite at Roman towns (Gerrard 2007; Holmes 2014a, 32) also go some way to imply the persistence of either a Roman elite or continuation of Roman culture on at least a regional scale in Britain as hypothesised in the introduction. However, certain trends typical of Romanised sites do seem to disappear, such as the hunting of wild animals, keeping of chickens (Allen 2014; Maltby 1997), and the move to sheep-based farming at some sites. This may relate to the suggestion that farming became more localised, albeit within previous Roman landscapes (Crabtree 1991; Lewit 2009) and also feeds into the model that much of post-roman Britain returned to a largely Figure 8 Mean proportion of the main domesticates by region for the middle Saxon phase. Figure 9 Relative proportions of the major domesticates by underlying geology for the late Saxon phase. Environmental Archaeology 2016 VOL. 21 NO. 1 67

11 Figure 10 Proportions of the main domesticates at burhs, Danish towns and rural sites in the middle Saxon phase. self-sufficient economy, in keeping with a culture maintaining its traditional ties as far back as the pre- Roman Iron age. However, it is improbable that this represents a direct link to Iron Age communities of some 400 years previously, it is more likely to reflect husbandry best supported by local conditions for a largely self-sufficient society. Furthermore, comparisons of Roman and Saxon human isotope studies have shown a considerable reduction in the amount of marine resources consumed in the early Saxon period compared to that of the Roman population (Müldner and Richards 2007, 690), again reflecting a change to a locally sourced food-base. Therefore, although there is evidence for continuity of an elite at some former Roman Towns and isolated sites, it seems that the former supply networks did not last long into the early Saxon period. As a final evaluation, the available butchery data were compared against the most common butchery methods associated with Roman techniques, i.e. filleting marks, removal of Figure 12 Relative proportions of the major domesticated from sites within the Danelaw. Danelaw East = sites from within the former East Anglia; Danelaw Midlands = sites from within the former Mercia; Danelaw North = sites in Yorkshire and Durham. the spine from the scapula, holes in scapulae, axial splitting of upper limb bones and transverse splitting of the metapodial shaft (Maltby 2007). Results show that at none of the early Saxon rural sites were any of these typical marks observed, and at West Stow, Old Down Farm and Quarrington, the presence of butchered horse bones within domestic waste deposits also implies a non-romanised population, as the Romans notably abstained from eating horse meat (Harris 1986, 94; Simoons 1978, 181). As with the zooarchaeological evidence, isotope studies have also found considerable variation between regions in the Saxon period (Hull and O Connell 2011, 679). However, individuals recovered from 22 native British graves (identified by burial position) throughout England showed similar carbon and nitrogen levels, indicating that a culturally distinct diet was a feature of the British population, suggesting that they maintained a diet separate to that of their Saxon neighbours (Hull and O Connell 2011, 678). The most compelling evidence for the presence of a Saxon culture in the eastern region comes from early sites where cattle were less common, proportional to the Figure 11 Relative proportions of the major domesticates by region for the late Saxon phase. Figure 13 Mean proportion of the main domesticates by region for the late Saxon phase. 68 Environmental Archaeology 2016 VOL. 21 NO. 1

12 Table 5 Proportions of domestic birds recorded at late Saxon sites (as a % of the total number cattle, sheep and pig) Site Area Wild mammal Domestic bird Wild bird Brandon Rd, Thetford Danelaw East Bury Road, Thetford Danelaw East Chalkpit Field North, Sedgeford Danelaw East Ipswich Danelaw East Ipswich Danelaw East Ipswich Danelaw East Knocker s site, Thetford Danelaw East Mill Lane, Thetford Danelaw East Maiden Lane and 6 7 Exchange Court Danelaw Mercia Chalk Lane, Northampton Danelaw Mercia Danesgate, Lincoln Danelaw Mercia Flaxengate, Lincoln Danelaw Mercia Langham Rd and Burystead, Raunds Danelaw Mercia Lincoln Danelaw Mercia Marefair, Northampton Danelaw Mercia Walton vicarage, Aylesbury Danelaw Mercia Coppergate, York Danelaw North Coppergate, York Danelaw North Coppergate, York Danelaw North Micklegate, York Danelaw North Skeldergate, York Danelaw North St Saviourgate, York Danelaw North 0 The south manor area, Wharram Danelaw North Wearmouth and Jarrow Danelaw North High st, Oxford Mercia Abbey Green, Chester Mercia All Saints Church, Oxford Mercia Crook Street, Chester Mercia Goss Street, Chester Mercia Hinxey Hall, Queen st, Oxford Mercia Hunter s Walk, Chester Mercia St Aldates, Oxford Mercia St Ebbes, Oxford Mercia Trill Mill Stream, Oxford Mercia Winchcombe Mercia , Jewry Street, Winchester Wessex Benham s Garage, Taunton Wessex Bristol Castle Wessex 0.61 Chester Rd, Winchester Wessex Citizen house, Bath Wessex 0.27 Market field, Steyning Wessex Mary-Le-Port, Bristol Wessex Mawgan Porth, Cornwall Wessex Portchester Castle Wessex Staple Gardens, Winchester Wessex Steyning Wessex Victoria Rd, Winchester Wessex Western Suburb, Winchester Wessex Only sites with a NISP > 300 cattle, sheep and pig recorded were included. high numbers of pigs kept. A predominance of pigs has been recognised previously at West Stow and explained as the utilisation of a fast-maturing animal to provide meat for a population establishing new herds and flocks from other livestock (Crabtree 1989, 210). This study has been able to show that this effect was more widespread, and also occurred away from East Anglia. Low numbers of cattle consumed, combined with the keeping of cattle at older ages in this area also implies the importance of cattle in the economy, suggesting that they would not be culled in the first instance until herds became well established, as they were presumably vital to the creation of the arable economy. It is therefore possible that West Stow, Barton Court Farm, Higham Ferrers and Botolphs, Bramber were home to Saxon settlers in the 5th 6th centuries, as they exhibit this high pig, low cattle signature. The subsequent decrease in pig numbers at later early Saxon settlements may have come about when meat was more plentiful from cattle, as pigs were replaced by a greater number of cattle. While it is likely that cattle continued to be important for traction, it is also possible that this was linked to a deeply embedded social meaning to the Saxon population. Cattle were highly valued in the German migration period, the old German word for cattle, fehu, also meaning money or possessions, and they were important for the maintenance of social relations through exchange (Hamerow 2002, 129). This is Environmental Archaeology 2016 VOL. 21 NO. 1 69

13 exemplified at the contemporary German settlement at Feddersen Wierde, the economy of which suggests that German animal husbandry at this time was based predominantly on cattle (Todd 1996), a trend mirrored at other north west European sites (Hamerow 2002, 148). As symbolic icons large animals such as cattle would be easily visible to other elites and the peasant classes alike (DeMarrais et al. 1996, 18). Certainly within England the predominance of high numbers of cattle in the Saxon-populated region (Fig. 14) suggests that there was some preference for cattle to be kept that may be related to a link with a traditional Germanic husbandry. If cattle were associated with wealth and could be used to maintain relationships between tribes, it would not be unrealistic to suggest that they would be of great importance to new settlers as a link to the ancestors and traditions left behind as well as a means of bargaining and exhibiting wealth and power. A greater number of wild mammals within the Saxon region were also occasionally observed (Fig. 14). The general dearth of wild species in early Saxon domestic assemblages implies that wild animals were valued for their symbolism rather than their place in the diet (Sykes 2011, 331). In Bede s Historia Ecclesiastica he mentions that the Britons only resorted to hunting once the whole country was left destitute of food (in Whitelock 1996, 644), implying a reluctance to otherwise procure food from wild animals. However, the early Germanic tribes were described by Caesar as living largely on cow s milk and hunted species, and although this is likely an exaggerated picture given the archaeozoological record (Hamerow 2002, 134), it does suggest that the Germanic population were not reticent of hunting. Two examples of hunting scenes were incised on cremation urns from Spong Hill dating to the 6th century, which differ from other grave goods of wild animals more commonly included as bones, skins or teeth as amulets, charms or symbols of status or power, (Hicks 1993, 24) as they represent hounds bringing down stags and hinds, indicating that the makers were familiar with the act of chasing and catching deer. The scarcity of such depictions in the archaeological record for this period remains consistent with the paucity of wild animals in the domestic lives of much of the population, the association of Saxons with hunting (e.g. at Saxon County School and Barton Court Farm) contrasts strongly with sites in the west, where such animals were even more poorly represented, and may be indicative of a difference in cultural identities. At Aelfric s Abbey; Figure 14 Map to show early Saxon trends in the data set. S 40% sheep; C 60% cattle; P 20% pig; B 3% domestic bird; W 1% wild mammal; * = poor burials with high nitrogen isotopes; +=other burials. 70 Environmental Archaeology 2016 VOL. 21 NO. 1

14 Manston Rd, Ramsgate and Poundbury, they were associated with greater numbers of sheep, which have already been suggested as indicating native sites and it may be possible that at these sites wild mammals were present in more of a symbolic context. When taken as a whole, the consideration of early Saxon animal human relationships with the major models of early Saxon change suggest the following: There was no universal collapse of all things Roman or return to a wholesale pre-roman Iron Age economy in the immediate early Saxon phase. There was evidence for a continuation of Roman economy in some areas in the consumption of cattle, prolonged use of former Roman towns and maintenance of morphologically similar animals, which may be related to a continuing Romano-Christian culture or Roman military elite. The later part of the early Saxon phase saw a return of some areas to a locally determined economy in the rise of sheep farming and decline of hunting and fishing similar to that occurring across western Europe following the retreat of the Roman empire. Direct evidence for a high-pig, low cattle diet exists at some eastern sites in the immediate post-roman phase that is likely to represent a Saxon culture. It is further likely that cattle were important to these populations economically for ploughing, socially to bestow status and wealth, and symbolically as a tie to the homeland. One further point remains to be made, relating to the high numbers of domestic fowl at settlements in Oxfordshire and Suffolk in the Saxon region that may have connections with a recent study into differences between status and diet. A considerable number of low-status burials (based on an absence of, or very few grave goods) have recently been established with greater values of nitrogen isotopes associated with terrestrial animal protein than those buried in wealthy graves (Hull and O Connell 2011, 675). The discrepancy lies in the expectation that wealthier individuals would be able to procure and therefore consume more meat than those of lower status. This has been explained by several theories, the most pertinent of which links the consumption of eggs to the low-status burials, as eggs have an elevated nitrogen isotope signature (Hull and O Connell 2011, 677). There is certainly close physical proximity between some cemeteries with poor burials exhibiting high nitrogen levels and sites with high numbers of domestic birds (Fig. 14), most notably at Berinsfield, Oxfordshire (Sherborne House, Lechlade; Aelfric s Abbey; Barton Court Farm) and Westgarth Gardens, Suffolk (Melford Meadows). If this does represent an underclass in early Saxon society; it is not unreasonable to suggest that they may be of a native British origin, as early Saxon laws imply that the majority of Britons were, of low legal status (Härke 1997, 148). There was relative homogeneity observed in the middle Saxon phase, where differences in the relative proportions of the major taxa can be explained in terms of geology or site type, and no perceivable difference in the exploitation of wild resources was observed. This is consistent with the consolidation of the Christianised, Anglo-Saxon culture in the study area from the 7th century. As with the Saxon migrants, in Scandinavian culture cattle were traditionally highly regarded (McCormick 1991, 42 43). It is well documented that the group of Vikings that colonised Iceland in the 870s successfully raised cattle, despite the poor quality of the land for arable production (Adalsteinsson 1991, 285; Wickham 1994, ). However, data from Viking-age sites in northern Europe record proportions of cattle ranging from 12 to 82% (data from O Connor 2010, 7), suggesting such an onus on cattle was not universal, and did not constitute a cultural signature (O Connor 2010, 13). Within the English data set; however, greater numbers of cattle were observed at many sites within the northern and eastern Danelaw (Fig. 15), made all the more notable given the active move away from sheepbased husbandry that predominated in the middle Saxon phase (Fig. 16), with a 25% drop in the mean proportion of sheep in the northern region, and similar increase in cattle. The presence of high numbers of cattle in the Danelaw has been noted previously, and described variously as being of little impact (Poole 2008, 111); representing cultural identity (O Connor 2010, 13) and an effect caused by an increase in sheep in the midlands and south (Sykes 2007, 34). All three conclusions are valid, but the regional detail considered here helps refine the argument. Elsewhere in Europe there was a considerable move towards surplus production, requiring communal farming and a greater intensification or specialisation of agriculture and husbandry from the 9th century at the latest, a marked intensification of bread-cereal production and a decline in the economic importance of cattle-rearing is apparent (Hamerow 2002, 147). Data from Wessex represents just such an active change from cattle-based farming of the previous phase, to an agriculture where sheep were more important, reflected in a 20% increase in the mean proportion (Fig. 16). This phenomenon was not only identified in Wessex, but also midlands sites in Mercia and the Danelaw (Fig. 11), consistent with the suggestion that agricultural change occurred in the midlands and central southern area, associated with larger, nucleated settlements surrounded by fields practicing a more intensive agriculture and a decrease in cattle husbandry from the late 9th century (Holmes 2014a, 78; Hooke 1998, 114; Environmental Archaeology 2016 VOL. 21 NO. 1 71

15 Figure 15 Map to show late Saxon trends in the data set. S 30% sheep; C 50% cattle; P 20% pig. Roberts and Wrathmell 2000, 28). Archaeobotanical evidence for an intensification of cereal production is scarce, although there is some early indication that cereals became more abundant in later Saxon phases (Moffett 2011, 357). Sheep would have been valued in the new agricultural regime for their manure, and an increase in the number of sheep was associated with expanding intensification of agriculture in Ireland and Iron Age England (McCormick 1991, 46). The evidence so far validates Sykes findings, that there was a move to sheep production in the late Saxon midlands area, associated with open field farming. Furthermore, it is likely that this change originated in the middle Saxon period, which has been seen in a small region in Cambridgeshire Figure 16 Mean proportion of cattle, sheep and pigs in various regions between middle and late Saxon phases. (Oosthuizen 2005). This can be observed in Fig. 17, where a cluster of sites in central Mercia, the area at the heart of open field farming in the late Saxon phase, contained a high number of sheep. It is also possible that the abundance of sheep outside this area may indicate the burgeoning importance of wool to the economy of the southern region (Holmes forthcoming 2.4.2). Although data showing high proportions of cattle in the Danelaw are somewhat skewed in favour of urban sites, where higher numbers of cattle may be expected (Fig. 9), they were also better represented at rural sites (Fig. 18). The high number of cattle recovered from the northern Danelaw may relate to an increase in ploughing as more land came into production yet, as in the early Saxon phase, there is also a strong possibility that this occurred alongside a cultural element. As suggested by O Connor, an increase in cattle may have been a way of assuring visible change to the Saxon population. The association of sheep with a new agriculture in the midlands included many sites in the Mercian Danelaw, at odds with the predominance of cattle in the rest of the Viking area. This suggests that either Scandinavian influence was less strong in this region than the north and east, or that they willingly took up agricultural innovation as the chance arose. Research into documentary, linguistic and archaeological evidence pertaining to the 72 Environmental Archaeology 2016 VOL. 21 NO. 1