Rural Settlement in Roman Britain Project Beyond the domestic: a review of the evidence for the exploitation of wild resources in the Roman South-West Dr Martyn Allen University of Reading
Hunting, fowling and fishing in Roman Britain Understudied area of late Iron Age/Roman Britain Small proportions of wild fauna suggest their diminutive economic importance which understates potential social/cultural significance of hunting, fowling and fishing Interpretations of wild animal remains of late Iron Age/Romano-British sites have altered and developed over time: Supplement meat supplies (Grant 1989), and to protection farmland (Buckland-Wright 1987) Iron Age taboo against consumption (King 1991; Dobney and Ervynck 2007); wild fauna only of symbolic importance (Hill 1995) Consumption of wild resources linked to status or new culinary practices, i.e. fish (Richards and Hodges 1998; Cool 2006; Locker 2006) Control over wild resources reflect changing cultural attitudes towards the natural world (Sykes 2010), or are expressions of landownership and social relations (Allen 2014) Hare and hound hunt relief from Bath, Somerset
percentage of assemblages (n = 152) Wild fauna in the south-west region (1stC BC-4thC AD) Distribution of animal bone assemblages with >100 identified mammal and bird specimens Proportion of assemblages with wild fauna 100.00 75.00 wild mammals present wild fowl present 50.00 25.00 0.00 % sites
No. assemblages No. assemblages Proportions of wild fauna by size of assemblage 80 70 60 50 40 30 20 NISP >100 NISP >300 NISP >500 NISP >1000 10 0 0% 0.1% - 1.0% 1.1% - 3.0% 3.1% - 5.0% 5.1% - 10.0% %NISP wild mammals 80 70 60 50 40 30 20 10 NISP >100 NISP >300 NISP >500 NISP >1000 0 0% 0.1% - 1.0% 1.1% - 3.0% 3.1% - 5.0% 5.1% - 15.0% %NISP wild fowl
Variation in the ubiquity of different wild mammal taxa percentage of assemblages present 0.00 10.00 20.00 30.00 40.00 red deer hare roe deer deer sp.ᶧ fox mustelid sp.ᶧᶧ badger fallow deer* wild boar** cetacean Wild mammals found on rural settlement sites include both hunted and trapped animals, from red deer and hare to fur-bearing species, such as badger, weasel and polecat Wild boar are suggested on some sites with large pig bones Some records include identifications of fallow deer an imported exotica in the Roman period (game parks?) otter bear wild cat ᶧantler removed though some may remain if unspecified in report ᶧᶧincludes weasel, stoat, polecat and pine marten *fallow deer also recorded in 1 non-quantified assemblage **wild boar also recorded in 2 non-quantified assemblages
mean percentage NISP Chronological variation in wild mammal and bird exploitation 1.40 1.20 wild mammal wild fowl 1.00 0.80 0.60 0.40 0.20 0.00 late Iron Age (n = 14) 1stC BC/AD (n = 19) 1st-3rdC AD (n = 19) 2nd-4thC AD 3rd-4thC AD (n = 37) (n = 38)
mean percentage NISP Inter-site type variation in the wild animal frequencies 2.00 1.80 1.60 wild mammal wild fowl 1.40 1.20 average wild mammal 1.00 0.80 0.60 0.40 average wild fowl 0.20 0.00 villa (n=22) nucleated (n=30) farm (n=80) religious (n=9) other rural (n=11) Villas stand out as sites tending to produce higher frequencies of wild mammals and birds Nucleated settlements also have comparative high frequencies Little variation between farms, religious and other rural sites all include low frequencies
%NISP %NISP %NISP Intra-site variation in wild mammal frequencies Frequencies generally minimal compared to domesticates Can reflect natural deaths or occasional exploitation Slightly higher frequencies may reflect more regular hunting and trapping of wild mammals c.50% of villa assemblages include wild mammal frequencies above 2% NISP Only outlier assemblages from farms and nucleated settlements produce >2% wild mammals 8.00 7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 8.00 7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 8.00 7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 %wild mammals on villas sites 0 5 10 15 20 25 no. sites %wild mammals on farm sites 0 20 40 60 80 no. sites %wild mammals on nucleated settlements 0 5 10 15 20 25 30 no. sites
mean percentage NISP Varying proportions of deer bone and antler in assemblages between different types of site 1.40 1.20 deer bone antler 1.00 0.80 0.60 0.40 0.20 0.00 villa (n=13) nucleated settlement (n=20) farm (n=39) religious (n=4)
http://finds.org.uk/database/artefacts/record/id/419674 number of sites Distribution of spear and arrowhead finds across the South-West 20 18 16 14 12 10 8 6 4 2 0 villa farm nucleated settlement other rural religious
Variation in wild fauna frequencies by site Minimum NISP = 15 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% deer group hare group wildfowl group mixed group deer hare fur-bearing mammals wetland fowl other wildfowl
Presence and frequency of different wild fowl taxa percentage of assemblages (n=74) 0.0 20.0 40.0 60.0 mean %NISP where present 0.0 1.0 2.0 3.0 4.0 5.0 ducks small waders geese sea birds cranes and herons galliforms birds of prey pigeons passerines corvids ducks small waders geese sea birds cranes and herons galliforms birds of prey pigeons passerines corvids Ducks and waders: commonly recovered high frequencies Corvids: commonly recovered low frequencies
Geographic variation of different taxa Corvid sp. Anser/Branta sp.
Restricted geographic distribution of seabirds Guillemot Great northern diver Cormorant Great auk (now extinct)
Distribution and frequency of fish remains 61 assemblages with fish bones 39 from assemblages with >100 identified specimens
percentage of assemblages Impact of wet sieving on the recovery of fish remains in south-west faunal assemblages 80 70 60 http://www.ucl.ac.uk/archaeology/research/projects/noviod unum/2005 50 40 https://sites.google.com/site/fishlabyork/home 30 20 10 0 sieved (n = 49) sieving not specified (n = 159)
carp perch trout roach common eel salmon stickleback flat fish seabream wrasses elasmobranch seabass gadid conger eel herring mullets scad/horse mackerel ling john dory mackerel no. assemblages Ubiquity of different fish taxa, grouped by habitat 8 7 6 5 4 3 2 1 0 freshwater /brackish migratory marine
no. assemblages No. of faunal assemblages with fish bones by phase 18 16 14 12 10 8 6 4 2 0 late Iron Age 1stC BC/AD 1st-2ndC AD 1st-4thC AD 2nd-4thC AD 3rd-4thC AD
Evidence for aquaculture? Halstock villa, Dorset Managed water-systems utilising natural spring from mid-2ndc AD System developed in 3rdC AD to supply an ornamental pond and control tanks within the courtyard No environmental sieving carried out; unknown whether fish adorned the gardens
1stC AD Romano-British fish-hook (photo M. Allen) no. of sites Distribution of fishing equipment finds across the South-West 5 4 3 Finds include: Fish-hooks?fishing weights Fish trap 2 1 0 villa farm nucleated settlement other rural religious
Proportion of sites with marine shells by type and taxon no. sites with marine shell (n=142) villa farm nucleated settlement religious coastal production oyster mussel limpet cockle periwinkle whelk scallop carpet shell topshell other % assemblages present 0 20 40 60 80 100
no. assemblages No. of faunal assemblages with marine shell by phase 50 40 30 20 10 0 late Iron Age 1stC BC/AD 1st-3rdC AD 1st-4thC AD 2nd-4thC AD 3rd-4thC AD
Geographic distributions of sites with oyster and mussels Oyster and mussel widespread in east Restricted to coast in west Mussel distribution more thinly spread than oyster, but follows same pattern
Geographic distributions of sites with periwinkle, cockle and limpet Clear coastal distribution periwinkles restricted to coastal sites Restricted inland distribution compared to oyster and mussel Distribution focused upon the Ilchester-Dorchester access route between the Somerset levels and Portland/Poole Harbour
% presence % presence % presence 100 80 oyster Inter-site variation in marine molluscs taxa frequency 60 40 20 0 100 80 60 villa (n=53) farm (n=44) nucleated settlement (n=28) mussel religious (n=11) coastal production (n=6) All marine shell taxa relatively well represented at coastal production sites Proportions of each taxon at consumer sites may suggest how widely each is selected for trade and consumption away from coastal settlements 40 20 0 villa (n=53) farm (n=44) nucleated settlement (n=28) religious (n=11) coastal production (n=6) Oyster widely recovered and well represented at all types of site Mussel widely distributed, but less well represented 100 80 60 periwinkle Periwinkle rarely recovered away from coastal sites 40 20 0 villa (n=53) farm (n=44) nucleated settlement (n=28) religious (n=11) coastal production (n=6)
% presence % presence 40 Inter-site variation in marine molluscs taxa frequency 30 20 10 Scallops found at a few villas - Highly prized? 0 villa (n=53) farm (n=44) nucleated settlement (n=28) religious (n=11) coastal production (n=6) 40 carpet shell 30 topshell other 20 10 Minor taxon recovered from a few farmsteads and villages - Low value? 0 villa (n=53) farm (n=44) nucleated settlement (n=28) religious (n=11) coastal production (n=6)
Summary points Large faunal dataset from the south-west, but severely biased to the east of the region Call for continued efforts in generating statistically-viable samples through considered excavation and recovery strategies, and reporting of standardised data Analysis suggests local and regional variation in wild animal exploitation practices, as well as social differentiation Deer hunting and venison consumption appears to be most commonly associated with villa settlements: protection of agricultural land, diversification of diet, and/or demonstrations of landownership? Evidence for the development of a coastal culture, expressed through exploitation of seabirds, marine fish and molluscs partly economically influenced