AMS dating of the Manx Ronaldsway Neolithic.

Burrow, Stephen ; Darvill, Timothy

Introduction

The Ronaldsway Culture of the Isle of Man is a clearly discernible and geographically focused late Neolithic regional grouping 'well defined in all its major aspects' (Piggott 1954: 346). Yet Ronaldsway material remains poorly dated and of unknown duration, an unfortunate obstacle to understanding the Manx cultural sequence, Neolithic social relations and cultural change in the coastlands of the Irish Sea region.

This paper reports a programme of AMS dating carried out in association with the Oxford University Radiocarbon Accelerator Unit. The aim was comprehensively to date the distinctive ceramic Ronaldsway jars.

The Ronaldsway Culture and its affinities

Ronaldsway Culture material, in particular the distinctive thick-walled jars, was initially thought to be of ultimate Bronze Age date/Clark 1935: 85). However, Second World War excavations by J.R. Bruce and E.M. Megaw at Ronaldsway Airport in the south of the Island (Bruce et al. 1947), and by G. Bersu at Ballateare in the north (Bersu 1947), showed beyond doubt that Ronaldsway material was earlier and should be placed within the later Neolithic.

The main features of the Ronaldsway Culture as defined by Basil Megaw (in Bruce et al. 1947: 157), subsequently affirmed and comprehensively illustrated (Piggott 1954: figures 59-61), were: deep baggy ceramic jars with overhanging rims, stone axes with roughened butts, engraved stone plaques, thick hump-backed flint scrapers, polished flint knives, hollow scrapers, and lozenge-shaped arrowheads. Single houses rather than villages were represented, as were cremation cemeteries. Most distinctive of all was the preponderance of ceramic jars placed upright in the ground with their tops just below ground-level and covered by a stone slab.

Piggott's pre-radiocarbon short chronology for the British Neolithic allowed the Ronaldsway Culture a duration of three centuries immediately prior to 1500 Be, making it contemporary with the Rinyo-Clacton, Peterborough and Dorchester Cultures (Piggott 1954: table opposite 380). By the most recent review of the Ronaldsway material (Moffatt 1978), five radiocarbon determinations were available from two relevant sites, but all are problematic and their associations with Ronaldsway material poor (TABLE 1). Of the three dates from Killeaba, Ramsey, one related to a 'burning pit' and was essentially of Mesolithic age (6310[+ or -]72 b.p.: BM-838). Two later dates, 3309-2905 BC(1) (4381[+ or -]58 b.p.: BM-839) and 3089-2704 BC (4300[+ or -]52 b.p.: BM-840) result from determinations on oak charcoal from what were interpreted as timber linings (?planked) within two separate pits, TI and TII, both containing cremated bone. The latter was loosely associated with a Ronaldsway jar which cut the upper fill (Cubbon 1978: pot 2). Stratigraphic position and the likelihood that the timber was older than the feature in which it lay means [TABULAR DATA FOR TABLE 1 OMITTED] that these dates only provide a very crude terminus post quem for the pottery.

The dates from Ballaharra, St John's, 30272600 BC (4225[+ or -]67 b.p.: BM-768) and 3026-2616 BC (4233[+ or -]59 b.p.: BM-769) both derive from charcoal found in a large cremation pit containing the remains of 30-40 individuals, four burnt lozenge-shaped flint arrowheads, and the base of a small ceramic vessel similar to a miniature vessel found Ballateare, Jurby (Cregeen 1978: 147). Moffatt attempted to reconcile all these dates, but concluded that the determinations served only to complicate rather than clarify the situation (1978: 216).

Little changed between 1978 and 1994. A single charcoal sample from Ballavarry, Andreas, provided a date of 2887-2584 BC (4140[+ or -]50 b.p.: GU-2696) for one of a series of interconnecting pits containing a scatter of fragmentary Ronaldsway jars and a decorated slate plaque (Garrad 1984: 162-5). Recent attempts to date animal bone from war-time excavations of the Ronaldsway 'House' were frustrated by contaminated or insecurely provenanced samples. Two dates were obtained: 2274-1451 BC (3490[+ or -]150 b.p.: GU-2694) and AD 452-986 (1310[+ or -]130 b.p.: GU-2695) but neither can be considered reliable (see Darvill 1996: 52-4 for listing of other prehistoric radiocarbon determinations from the Isle of Man).

In the absence of secure dates for the Ronalds-way Neolithic various cultural associations for the tradition have been suggested through analogy. These include links with Grooved Ware (Piggott 1954: 351), with Irish passage-graves through the presence of barbell bone pins (Herity 1982: 284), and with middle/late Neolithic Irish pottery styles in Sheridan's Phase 3 provisionally dated to 3100-2600 BC (summarized in Sheridan 1995: 6). However, none of these cross-cultural similarities is clear-cut or well-established.

AMS dating

The presence of carbonaceous deposits on the lower inside walls of Ronaldsway jars has been noted several times (Kewley 1907; Bruce et al. 1947: 158) and explained as carbonized food residues of some kind (Bruce et al. 1947: 158). During the examination of the pottery in the Manx Museum it was found that sufficient carbonized material for AMS dating still adhered to a large number of sherds from several sites on the island, including the Ronaldsway 'House', Glencrutchery, and Ballateare. Accordingly, with the support of the Oxford University Radiocarbon Accelerator Unit, a dating programme commenced in January 1995.

Twenty samples from seven sites were selected. Choice of vessel was dictated by the availability of sherds with adequate carbonized material adhering.(2) Wherever possible, samples were taken from rim or decorated sherds [ILLUSTRATION FOR FIGURE 1 OMITTED], although some samples had to be taken from plain body sherds. In all cases care was taken to select sherds of differing thickness, textures and temper types in order to minimize the possibility that two sherds from the same vessel were being sampled. Tests on the samples revealed that they comprised elemental carbon, although in this study no residue analysis as such was carried out.

Chemical pretreatment, target preparation and AMS measurement followed standard procedures (Bronk & Hedges 1989; Hedges et al. 1989; 1992). The resulting determinations were calibrated at two standard deviations using the OxCal computer program (version 2.01, 1996) with the bi-decadal calibration curve (Pearson et al. 1986). TABLE 2 summarizes the determinations and calibrations made; FIGURE 2 shows the results in graphical form. The following notes describe the archaeological context of the material studied. FIGURE 3 shows the distribution of sites dated in relation to other recorded Ronaldsway pottery finds.

Ballacottier NX 435005

Trial trenching over a surface scatter located sherds representing a minimum of 28 vessels. No clear archaeological features were found (Garrad 1984).

Ballalheaney SC 426986

Small-scale excavation of a surface scatter was carried out and sub-soil hollows defined (Garrad 1985). Sherds from a minimum of 5 vessels were recovered.

Ballateare SC 344970

Cemetery site excavated by G. Bersu (1947). Ronaldsway jars were found either holding, or besides, cremation deposits; two had sufficient carbon deposit for dating (Bersu's vessels CVI and CXII).

Ballavarry SC 409681

Three inter-connecting pits discovered through excavation. No evidence was found for stratigraphy within the fills (Garrad pers. comm. 1995). Parts of at least 17 vessels was recovered, along with lithics, and a inscribed slate plaque (Garrad 1984).

[TABULAR DATA FOR TABLE 2 OMITTED]

Glencrutchery SC 382776

A possible settlement mainly examined during quarrying in the 1890s, although finds continued to be made after the turn of the century (Clark 1935). Over 450 sherds have been recovered, representing a minimum of 48 vessels.

Ronaldsway 'House' SC 291686

Type-site of the Ronaldsway Culture with the largest ceramic assemblage on the Island: c. 550 sherds representing a minimum of 44 vessels (Bruce et al. 1947). It was not possible to distinguish between sherds from the basal deposit of the pit and those from upper fills.

West Kimmeragh NX 437004

A surface scatter excavated to reveal a cobbled surface into which cremated remains had been placed. Sherds of at least 6 vessels were recovered (Garrad 1987).

Discussion

All the samples provided what appear to be reliable dates. Using carbonized material from vessel walls the dates should relate closely to the use of the pots; there is no evidence that the deposits result from re-use of sherds, residual contamination from secondary activities, or post-depositional processes. The dates are remarkably consistent and clearly show that the Ronaldsway ceramic tradition spans the 3rd millennium BC and should be seen as a fairly long-lived late Neolithic phenomenon. Although the determinations from some of the sites studied appear to bunch together while others are more spread, the total number of dates available is insufficient to provide an internal chronology for the tradition as a whole.

In relation to other British and Irish late Neolithic ceramic traditions, the dates show considerable overlaps with Peterborough wares, especially the Mortlake and Fengate components (Gibson & Kinnes 1997), and also Grooved Ware (Wainwright & Longworth 1971: 265; MacSween 1995; Sheridan 1995: 15). The latter is represented on the Isle of Man by vessels with incised decoration at the Ronaldsway 'House' and Glencrutchery (e.g. Bruce et al. 1947: figure 7.8). The later centuries of the Ronaldsway ceramic tradition overlap with the currency of Beaker pottery (Case 1993), which is notably sparse on the Isle of Man.

Perhaps the most important questions relating to Ronaldsway pottery concern its origins. There are no obvious sources of inspiration on the Isle of Man. Neolithic communities of the 4th millennium BC constructed long barrows and passage-graves, and used plain bowl-style pottery that finds good parallels all around the Irish Sea, especially in Ireland to the west and Dumfries and Galloway, Scotland, to the north. However, no middle Neolithic jars are known on the Island and nothing that has been found to date provides an obvious ancestry for the Ronaldsway ceramic tradition. Two possible explanations present themselves. First, Ronaldsway ceramics represents a purely local development with no external stimuli. Second, that some more distant influences were involved.

Local development is supported by the restricted distribution of stone implements of quartz diorite/tonalite (Petrological Group XXV) identified to a source on the southeastern side of the Island (Coope & Garrad 1988). Implements in this stone, including the distinctive axes with roughened butts, are apparently absent from Ireland (Cooney & Mandal 1995), southwest Scotland (Ritchie & Scott 1988), Cumbria (Fell & Davies 1988) and North Wales (Houlder 1988).

Against this must be set some observable similarities between Ronaldsway ceramics and both earlier and contemporary deep baggy vessels from elsewhere in the British Isles. Piggott (1954: 351), for example, drew attention to vessels from as far afield as Icklingham, Suffolk, and more recent studies have confirmed the widespread yet proportionately small-scale presence of deep jars,(3) sometimes with lugs and cordons, in both British and Irish middle Neolithic assemblages (Smith in Mercer 1981: figure 72; Cleal 1992: 293; Sheridan 1995). Morphologically, the closest similarities are with the deep vessels with overhanging rims in the Hebridean style of the Western Isles (Henshall 1972: 153-6; Kinnes 1985: 22). The domestic site of Eilean an Tighe, North Uist, provides interesting parallels for the Ronaldsway forms, although more heavily decorated and significantly different in their fabric and vessel size. Moreover, on the basis of the rimsherds quantified, deep jars represented less than 20% of the Hebridean assemblage at Eilean an Tighe (Scott 1951: especially [ILLUSTRATION FOR FIGURE 6 OMITTED]). The passage-grave on Unival, North Uist (UST 34) also has a small proportion of similar vessels (Scott 1948).

In the light of the new dating for the Ronaldsway ceramic tradition proposed here possible contributions from British impressed wares may also be considered. Although the exact forms cannot be paralleled, deep jars are certainly a feature of the Peterborough repertoire (e.g. Piggott 1962: figure 12, P15), as too is the extensive use of overhanging rims (Gibson 1995 for material from Wales). As with Ronaldsway jars, coarsely tempered fabrics are a distinctive feature of Mortlake wares (Gibson 1995: 24). Decoration is extensively applied on Peterborough vessels, but is confined to the rim on Ronaldsway pots. However, the means and style of decoration, for example the use of incised lines and impressed designs using fingernails, points and bird bones, is common to both traditions.

It would be easy to overestimate possible contributions to the development of the Ronaldsway ceramic tradition either from Hebridean communities or from users of impressed wares to the south and east. A high degree of parallel development is evident in the ceramic traditions of many late Neolithic communities in the British Isles, but the Ronaldsway material is unusual in being geographically isolated without obvious local prototypes. On the basis of the pottery evidence it would seem reasonable to propose limited contact between relatively autonomous island-based communities around the north Irish Sea basin and western Scotland, links northward towards the Western Isles perhaps being the strongest. Indeed, other general similarities can be seen in this same area, for example the poverty of Grooved Ware, the occasional presence of stone balls, the extensive use of white quartz pebbles, and the apparent absence of large developed passage-graves of the type found in the Boyne Valley of Ireland, Anglesey, and Orkney (Henshall 1972: 111-57; Armit 1992). Together, these characteristics set the area apart from the adjacent mainland regions of eastern Ireland, Wales, northern England, and western Scotland to the extent that it may be possible to glimpse the development of a linear, island-based interaction sphere contemporary with, and thus socially parallel to, those polities using developed passage-graves identified by Bradley & Chapman (1986).

Acknowledgements. The authors would like to thank Stephen Harrison, Wendy Horn, Andrew Johnson, Larch Garrad, and Hazel Simons of Manx National Heritage for their help and assistance with the location and sampling of the vessels dated through the programme of research reported in this paper. The potsherds used in the study were variously sampled in the Manx Museum, Douglas, and under the terms of a licence issued by Manx National Heritage on 19 December 1994 in accordance with Section 21 of the Manx Museum and National Trust Acts 1959 to 1986. Rupert Housley, Paul Pettitt, and Robert Hedges of the Radiocarbon Accelerator Unit at the Oxford University Research Laboratory for Archaeology and the History of Art kindly facilitated the processing of samples, the production of the dates, and helpful comments on the interpretation of the results. The study was undertaken as an in-house project funded through a NERC Research Grant. Louise Geall kindly undertook preliminary analysis of the samples to determine carbon content. Funding for research into Manx Neolithic ceramics by one of the authors (SB) is provided by a British Academy postgraduate studentship.

1 All calibrated radiocarbon dates in this paper are expressed at two standard deviations and were determined using the OxCal computer program (1996) and the 1986 bi-decadal calibration curve.

2 Approximately 5-30 mg is required. This can typically be obtained from a surface deposit c. 0.5 mm thick over an area of 2 sq. cm.

3 Defined here as having a mouth diameter to depth ratio of 1:1.5 or greater.

References

ARMIT, I. 1992. The Hebridean Neolithic, in Sharples & Sheridan (ed.): 307-21.

BERSU, G. 1947. A cemetery of the Ronaldsway Culture at Ballateare, Jurby, Isle of Man, Proceedings of the Prehistoric Society 13: 161-9.

BRADLEY. R, & R. CHAPMAN. 1986. The nature and development of long-distance relations in later Neolithic Britain and Ireland, in C. Renfrew & J. Cherry (ed.), Peer polity interaction and socio-political change: 127-36. Cambridge: Cambridge University Press.

BRONK, C.R. & R.E.M. HEDGES. 1989, Use of the CO2 source in radiocarbon dating by AMS, Radiocarbon 31(3): 298-304.

BRUCE, J.R., E.M. MEGAW & B.R.S. MEGAW. 1947. A Neolithic site at Ronaldsway, Isle of Man, Proceedings of the Prehistoric Society 13: 139-60.

CASE, H. 1993. Beakers: deconstruction and after, Proceedings of the Prehistoric Society 59: 241-68.

CLARK, J.G.D. 1935. The prehistory of the Isle of Man, Proceedings of the Prehistoric Society 1: 70-92.

CLEAL, R. 1992. Significant form: ceramic styles in the earlier neolithic of southern England, in Sharples & Sheridan (ed.): 286-304.

CLOUGH, T.H.McK. & W.A. CUMMINS (ed.). 1988. Stone axe studies 2. London: Council for British Archaeology. Research report 67.

COONEY, G, & S. MANDAL. 1995. Petralogical results from the Irish Stone Axe Project, Antiquity 69: 969-80.

COOPE, G.R. & L.S. GARRAD. 1988. The petralogical identification of stone implements from the Isle of Man, in Clough & Cummins (ed.): 67-70.

CREGEEN, S. 1978. Ballaharra excavations 1971. A summary of work and results, in P.J. Davey (ed.), Man and environment in the Isle of Man: 141-64. Oxford: British Archaeological Reports. British series 54.

CUBBON, A.M. 1978. Excavation at Killeaba, Ramsey, Isle of Man, Proceedings of the Prehistoric Society 44: 69-96.

CUBBON, W. 1934. Bronze Age cemetery site at Knocksharry, Proceedings of the Isle of Man Natural History and Antiquarian Society 3: 446-54.

DARVILL, T. 1996. Billown Neolithic Landscape Project, Isle of Man, 1995. Bournemouth and Douglas: Bournemouth University and Manx National Heritage. School of Conservation Sciences Research Report 1.

FELL, C.I. & DAVIS, R.V. 1988. The petrological identification of stone implements from Cumbria, in Clough & Cummins (ed.): 71-84.

GARRAD, L.S. 1984. Rescue excavations, 1980-83, Proceedings of the Isle of Man Natural History & Antiquarian Society 9.2: 157-68.

1985. Rescue trials at Ballaheaney, Andreas. Manx Museum: unpublished typescript report.

1987. A Ronaldsway Neolithic site on West Kimmeragh, Bride, Isle of Man, Proceedings of the Isle of Man Natural History & Antiquarian Society 9.3: 421-6.

GIBSON, A. 1995. First impressions: a review of Peterborough Ware in Wales, in Kinnes & Varndell (ed.): 23-40.

GIBSON, A. & KINNES, I. 1997. On the urns of a dilemma: radiocarbon and the Peterborough problem, Oxford Journal of Archaeology 16.1: 65-72.

HEDGES, R.E.M., I.M. LAW, C. BRONK RAMSEY & R.A. HOUSLEY. 1989. The Oxford mass spectrometer facility: technical developments in routine dating, Archaeometry 31.2: 99-114.

HEDGES, R.E.M., M.J. HUMM, J. FOREMAN, G.J. VAN KILKEN & C.R. BRONK. 1992. Developments in sample combustion to carbon dioxide, and in the Oxford AMS Carbon Dioxide ion source system, Radiocarbon 34.3: 306-11.

HENSHALL, A.S. 1972. The chambered tombs of Scotland 2. Edinburgh: Edinburgh University Press.

HERITY, M. 1982. Irish decorated Neolithic pottery, Proceedings of the Royal Irish Academy 82C: 247-404.

HOULDER, C.H. 1988. The petrological identification of stone implements from Wales, in Clough & Cummins (ed.): 133-6.

KEWLEY, J. 1907. On a cinerary urn from Ballahot, Proceedings of the Isle of Man Natural History and Antiquarian Society 1.2: 33-4.

KINNES, I. 1985. Circumstance not context: the Neolithic of Scotland as seen from outside. Proceedings of the Society of Antiquaries of Scotland 115: 15-57.

KINNES, I. & G. VARNDELL (ed.). 1995. 'Unbaked urns of rudely shape'. Essays on British and Irish Pottery for Ian Longworth. Oxford: Oxbow. Monograph 55.

MACSWEEN, A. 1995. Grooved Ware from Scotland: aspects of decoration, in Kinnes & Varndell (ed.): 41-8. Oxford: Oxbow. Monograph 55.

MERCER, R.J. 1981. Excavations at Carn Brea, Illogan, Cornwall, 1970-73, Cornish Archaeology 20: 1-204.

MOFFATT, P.J. 1978. The Ronaldsway Culture, in P.J. Davey (ed.), Man and environment in the Isle of Man: 177-218. Oxford: British Archaeological Reports. British series 54.

PEARSON, G.W., J.R. PILCHER, M.L. BAILLIE, D.M. CORBETT & F. QUA. 1986. High-precision 14C measurement of Irish Oaks to show the natural 14C variation from AD 1840-5210 DC, Radiocarbon 28: 911-34.

PIGGOTT, S. 1954. The Neolithic Cultures of the British Isles. Cambridge: Cambridge University Press.

1962. The West Kennet Long Barrow. Excavations 1955-56. London: HMSO. Ministry of Works Archaeological Reports 4.

RITCHIE, P.R. & J.G. SCOTT. 1988. The petrological identification of stone axes from Scotland, in Clough & Cummins (ed.): 85-91.

SCOTT, W.L. 1948. The chambered tomb of Unival, North Uist, Proceedings of the Society of Antiquaries of Scotland 82:1-49

1951. Eilean an Tighe: a pottery workshop of second millennium BC, Proceedings of the Society of Antiquaries of Scotland 85:1-37

SHARPLES, N. & A. SHERIDAN (ed.). 1992. Vessels for the ancestors. Edinburgh: Edinburgh University Press.

SHERIDAN, A. 1995. Irish Neolithic pottery: the story in 1995, in Kinnes & Varndell (ed.): 3-22.

WAINWRIGHT, G.J., & I.H. LONGWORTH. 1971. Durrington Walls: Excavations 1966-1968. London: Society of Antiquaries. Research Committee Report 29.