South Wales Gas Pipeline Project Site 38.17 Land North of Llwyn-Meurig Trecastle Powys Archaeological Excavation
for Rhead Group on behalf of National Grid CA Project: 9150 CA Report: 13315 Event No.: CPAT102846
August 2014
South Wales Gas Pipeline Project Site 38.17
Archaeological Excavation
CA Project: 9150 CA Report: 13315 Event No.: CPAT102846
prepared by Jonathan Hart, Senior Publications Officer
date 7 August 2014
checked by Karen E Walker, Post-Excavation Manager
date 17 July 2015
approved by Martin Watts, Project Director, Head of Publications
signed
date
issue 01
This report is confidential to the client. Cotswold Archaeology accepts no responsibility or liability to any third party to whom this report, or any part of it, is made known. Any such party relies upon this report entirely at their own risk. No part of this report may be reproduced by any means without permission.
© Cotswold Archaeology
Cirencester Milton Keynes Andover Building 11 Unit 4 Office 49 Kemble Enterprise Park Cromwell Business Centre Basepoint Business Centre Kemble, Cirencester Howard Way, Newport Pagnell Caxton Close, Andover Gloucestershire, GL7 6BQ MK16 9QS Hampshire, SP10 3FG t. 01285 771022 t. 01908 218320 t. 01264 326549 f. 01285 771033 e. [email protected] © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
CONTENTS
SUMMARY ...... 3
1. INTRODUCTION ...... 4
2. RESULTS (FIGS 2–12) ...... 6
3. PROJECT TEAM ...... 15
4. REFERENCES ...... 16
APPENDIX A: CONTEXT DESCRIPTIONS ...... 19
APPENDIX B: THE FINDS INCLUDING HALBERD DESCRIPTION BY STUART NEEDHAM 23
APPENDIX C: THE HUMAN REMAINS BY ANNA FOTAKI AND MALIN HOLST ...... 28
APPENDIX D: THE PALAEOENVIRONMENTAL EVIDENCE BY JAMES RACKHAM .... 31
APPENDIX E: RADIOCARBON-DATING EVIDENCE BY SEREN GRIFFITHS ...... 44
LIST OF ILLUSTRATIONS Fig. 1 Site location plan (1:25,000) Fig. 2 Plan of archaeological features (1:50) Fig. 3 Photograph: pre-excavation view of ring-ditch, looking south-west Fig. 4 Photograph: working shot showing ring-ditch, looking south-west Fig. 5 Photograph: post-excavation view of ring-ditch, looking south-west Fig. 6 Sections through ring-ditch and associated features (1:20) Fig. 7 Photograph: working shot of halberd being excavated Fig. 8 Photograph: detail shot showing halberd in situ Fig. 9 Photograph: detail shot of halberd following conservation Fig. 10 Photograph: detail shot of halberd following conservation Fig. 11 Drawing of halberd following conservation (1:2) Fig. 12 Plans and sections of easternmost features (1:20 and 1:10) Fig. 13 A Bayesian statistical chronological model for Site 38.17 Fig. 14 A more interpretative Bayesian chronological model for Site 38.17
1 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
GLOSSARY CA – Cotswold Archaeology CAP – Cambrian Archaeological Projects CPAT – Clwyd Powys Archaeological Trust DAT – Dyfed Archaeological Trust GGAT - Glamorgan Gwent Archaeological Trust FTP – Felindre to Brecon gas pipeline HER – Historic Environment Record MHA – Milford Haven to Aberdulais gas pipeline NLMJV – Nacap Land & Marine Joint Venture UPD – Updated Project Design
2 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
SUMMARY
Project Name: South Wales Gas Pipeline Project Location: Site 38.17, Land North of Llwyn-Meurig, Trecastle, Powys NGR: SN 8588 2917 Type: Excavation Date: 27 March–24 April 2007 Location of Archive: To be deposited with RCAHMW (original paper archive) and Brecknock Museum (material archive and digital copy of paper archive; accession number BRNCM 2013.19.08) Site Code: MHA06
An archaeological excavation was undertaken by Cotswold Archaeology during groundworks associated with construction of gas pipelines (part of the South Wales high pressure gas pipeline scheme) between Milford Haven and Aberdulais, and Felindre and Brecon, which were conducted between 2005 and 2007.
A ring-ditch was found on top of a ridge overlooking the River Usk. The ridge adjoins Mynydd Bach Trecastell and Mynydd Myddfai and is the location of previously recorded prehistoric activity. Radiocarbon-dating and Bayesian analysis suggests that the ring-ditch dated to the Early Bronze Age/Beaker period and was in use between at least 2590 and 2200 cal. BC. The ring-ditch enclosed pits, one of which contained a rare copper halberd and was perhaps a founder deposit associated with the first use of the ring-ditch. The halberd is a find of considerable importance as it is only the second from Britain to have been recovered during archaeological excavation. Part of its wooden haft survived and the presence of organic material offered the first opportunity to obtain a radiocarbon date in direct association with a halberd in north-western Europe. Human remains were absent from these features, and it is possible that the ring-ditch was regarded as a cenotaph rather than as a grave. Later on in the Early Bronze Age, the ring-ditch was revisited and was used for the burial of assemblages of at least two separate cremated human remains, at which time the monument may have been marked with an upright post or totem pole. This activity occurred between at least 2590 and 2200 cal. BC. For each of these periods, it is possible that single events are represented by the dated features. Some 100m to the east, three Middle Bronze Age pits were found. The function of these is not clear, although they contained a range of finds suggestive of domestic debris either deposited within a former settlement or within a funerary/ritual setting.
3 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
1. INTRODUCTION
1.1 NACAP Land and Marine Joint Venture (NLMJV), on behalf of National Grid, commissioned RSK Environment (part of the RSK Group) to manage the archaeological works (non-invasive surveys, desk based assessment, evaluation, watching brief, and open area excavation) on a 216km-long section of pipeline from Milford Haven (Pembrokeshire) to Brecon (in Powys). The high pressure gas pipeline (part of the 316km-long pipeline route from Milford Haven to Tirley in Gloucestershire) was required to reinforce the gas transmission network. The archaeological work performed in advance of this pipeline was undertaken in a number of sections by a number of archaeological companies. The westernmost section of 122km, from Milford Haven to Aberdulais, was investigated by CA (then Cotswold Archaeological Trust) during 2005–2007 with some additional excavation work carried out by CAP. The section of 89km, from Felindre to Brecon was investigated by CA during 2006–2007 and CAP during 2007. Assessment reports on the works were completed in January 2012 (NLM 2012a, 2012b) and the current reporting stage was commissioned in February 2013.
1.2 In March and April 2007 CA carried out an archaeological excavation at Site 38.17, Land North-east of Llwyn-Meurig, Llywel, Powys (centred on NGR: SN 8588 2917; Fig. 1). The objective of the excavation was to record all archaeological remains exposed during the pipeline construction.
1.3 The excavation was carried out in accordance with professional codes, standards and guidance documents (EH 1991; IfA 1999a, 1999b, 2001a, 2001b and IfA Wales 2008). The methodologies were laid out in an Archaeological Framework Document (RSK 2007) and associated Written Statements of Investigation (WSIs) and Method Statements.
The site 1.4 The site is located within a field on top of a north-west/south-east aligned ridge of high ground between the Nant Gwydderig to the north and the River Usk to the south (Fig. 1). It is located at approximately 350m AOD towards the eastern end of the ridge, the western end of which adjoins Mynydd Bach Trecastell.
4 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
1.5 The underlying solid geology of the area is mapped as the St Maughans Formation (Interbedded Argillaceous Rocks and Sandstone) of the Devonian Period overlain by in the northern part of the site by superficial deposits of Quaternary Till (BGS 2013).
Archaeological background 1.6 No archaeological remains were identified within the site during the preliminary Archaeology and Heritage Survey (CA 2006). However, the wider landscape is populated with prehistoric monuments, particularly along the ridgeline on which the site lies.
1.7 Tyle Mawr, a round cairn, is 600m east of the site at the eastern end of the ridge (CPAT3429; BR340). Pant Madog Cairn, the remains of a round barrow, also lies on the ridge crest, 1.6km north-west of the site (CPAT3102; BR341). Further prehistoric monuments recorded by the HER form a cluster 3.3km north-west of the site and include at least two, possible three, stone circles (CPAT12386; BR069; CPAT12387), two possible stone rows, a standing stone and two possible cairns (CPAT12386; CPAT743; BR142). Whilst the HER records suggest that all of these features date to the Bronze Age, the dating of these is largely untested and it is possible that some have earlier origins.
1.8 During the pipeline construction works archaeological remains were found in the near vicinity of Site 38.17 (Fig. 1). These included undated road deposits found along the line of a Roman road at Site 38.14 and an alignment of three undated pits with charcoal-rich fills at Site 38.20.
Archaeological objectives 1.9 The objectives of the archaeological works were:- • to monitor groundworks, and to identify, investigate and record all significant buried archaeological deposits revealed on the site during the course of the development groundworks; and • at the conclusion of the project, to produce an integrated archive for the project work and a report setting out the results of the project and the archaeological conclusions that can be drawn from the recorded data.
Methodology 1.10 The fieldwork followed the methodology set out within the WSI. An archaeologist was present during intrusive groundworks comprising stripping of the pipeline
5 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
easement to the natural substrate (Fig. 1). The limits of excavation were closely centred on the limits of the exposed features and it was not possible to examine the wider environs beyond these.
1.11 Where archaeological deposits were encountered written, graphic and photographic records were compiled in accordance with CA Technical Manual 1: Fieldwork Recording Manual (2005). The features exposed were 100% excavated. The discovery of a prehistoric halberd resulted in a meeting between CA, RSK and a team of conservators. The latter block-lifted the halberd and it was taken to a lab for excavation and conservation (see Appendix B for details).
1.12 The post-excavation analysis and reporting was undertaken following the production of the UPD (GA 2012) and included re-examination of the original site records. Finds, environmental and radiocarbon-dating evidence was taken from the assessment reports (NLM 2012b) except where the UPD recommended further work, in which case the updated reports were used. The archaeological background to the site was assessed using the following resources:- • the Archaeology and Heritage Survey which was undertaken in advance of the pipeline construction and which examined a 1km-wide corridor centred on the pipeline centre line, including the then existing HER record (CA 2006); • Clwyd-Powys Archaeological Trust HER data (received May 2014); and • other online resources, such as Google Earth and Ordnance Survey maps available at http://www.old-maps.co.uk/index.html. All monuments thus identified that were relevant to the site were taken into account when considering the results of the fieldwork.
1.13 The archive and artefacts from the watching brief are currently held by CA at their offices in Kemble. Subject to the agreement of the legal landowner the artefacts will be deposited with Brecknock Museum under accession number BRNCM 2013.19.08, along with a digital copy of the paper archive. The original paper archive will be deposited with the RCAHMW.
2. RESULTS (FIGS 2–12)
2.1 This section provides an overview of the excavation results; detailed summaries of the recorded contexts, finds, human remains, palaeoenvironmental evidence and
6 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
radiocarbon dates are to be found in Appendices A–E. Full, original versions of the specialist reports are contained within the archive. The radiocarbon date ranges quoted in italics are posterior estimates based on Bayesian analysis; those not in italics are calibrated dates (see Appendix E for methodologies relating to these dates).
2.2 The geological silty clay substrate (38.17.002) was cut by a ring-ditch, a hearth, pits and a posthole (Fig. 2). The presence of fairly deep features and of stone capping above some of these suggests that any horizontal truncation in antiquity has been limited. The ring-ditch fill and the majority of the discrete features were dated through radiocarbon determinations and, based on these and on subsequent Bayesian analysis (Appendix E), three distinct phases of activity (Phases A–C) have been identified.
Phase A: Early Bronze Age/Beaker c. 2590–2200 cal. BC 2.3 Phase A comprised the construction of the ring-ditch and the excavation of at least two internal pits. Ring-ditch 38.17.011 was 8m–9m in diameter internally and consisted of a steep-sided cut with a flat-based profile, typically 0.8m wide and 0.3m deep (Figs 2–6). In places the ditch contained a thin primary silt deposit but it was mainly filled with homogenous grey-brown to red-brown silty clays which included frequent small to medium-sized stones. Finds from the ditch fills were restricted to occasional charcoal flecks, a small quantity (0.1g) of burnt bone unidentifiable to species, and four flints of which the only one to be closely dateable was diagnostic of Mesolithic/Early Neolithic technology. Radiocarbon-dating on charcoal from the ditch fill produced an Early Bronze Age/Beaker period date of 2440–2200 cal. BC (SUERC-52593; 95% probable).
2.4 Pit 38.17.060 was located just east of the central point of the ring-ditch. It comprised a steep-sided, flat-based cut 1.2m wide and 0.5m deep (Fig. 6, section DD). This was filled with stony clay silt 38.17.057 which included occasional flecks of wood charcoal. Towards the uppermost surviving part of this fill, a copper halberd was found, that is paralleled within the Irish halberd corpus (Needham et al., Appendix B; Figs 7–11; http://www.cotswoldarchaeology.co.uk/highlight-11/; Needham 2015). Attached to this was a surviving part of the wooden haft which produced a radiocarbon date of 2450–2200 cal. BC (Beta-240338; 95% probable). The infilled pit was truncated by a somewhat larger pit, 38.17.056, which had a similar profile
7 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
but which contained no finds but which has been assigned to Phase B on the basis of this stratigraphic relationship.
2.5 Immediately north of pit 38.17.060 was pit 38.17.064. This was also a steep-sided, flat-based cut and was 1.3m wide and 0.8m deep (Fig. 6, section EE). It contained a thin primary clay silt fill, 38.17.065, overlain by a red-brown pebbly clay silt fill, 38.17.066 which was capped with stone rubble 38.17.067. The uppermost fill, 38.17.068, consisted of stones within clay silt. Charcoal from the primary fill was radiocarbon dated to 2460–2280 cal. BC (SUERC-52590; 92% probability) whilst charcoal from fill 38.17.066 was dated to 2470–2280 and 2460–2280 cal. BC (SUERC-52591–2; 93% and 94% probability respectively).
Phase B: Early Bronze Age c. 1970–1680 cal. BC 2.6 During Phase B, at least eight further features were dug within the ring-ditch interior or were cut into its ditch fills, which by this time had filled to the depth at which the feature survived during the excavation (0.3m). Almost all of these features were found within the eastern half of the ring-ditch, mostly within the north-eastern quadrant.
2.7 The easternmost of these features was pit 38.17.058 which had been cut into the ring-ditch fill. It was a steep-sided, flat-based cut 0.5m wide and 0.1m deep (Fig. 6, section FF) and contained 223.8g of cremated human remains. Osteological analysis revealed that all of the cremated bone assemblages from the site were relatively well burnt, and there was a loss of bone surface detail, a high degree of fragmentation and some cracking. The remains from pit 38.17.058 included skull fragments, tooth fragments, and some mostly unidentified long bone fragments. No duplicate bone elements were found from all of the contexts in the pit, so this probably only represents one individual, who was aged at least 16 at the time of death. The sex could not be determined (Fotaki and Holst, Appendix C). The cremated remains were found alongside wood charcoal, probably the remains of pyre fuel, and the burial had been capped by flat-laid stones. The charcoal was dominated by oak but included significant quantities of birch. A fragment of the cremated human bone was radiocarbon dated to 1890–1740 cal. BC (SUERC- 52598; 95% probability). Adjacent to this, within the ring-ditch interior, was a smaller pit, 38.17.023. This pit was 0.4m wide and 0.1m deep and contained sandy silt 38.17.024 which included patches of charcoal, a sample of which was radiocarbon dated to 1890–1740 cal. BC (SUERC-52580; 95% probability). The proximity of
8 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
these two features to one another, the presence in both of charcoal and the similarity of the radiocarbon dates obtained from them suggests that they may represent a pair of contemporaneous features excavated during a single burial rite.
2.8 This pairing of two pits associated with a cremation was also found along the northern edge of the ring-ditch, with pits 38.17.026 and 38.17.080. Pit 38.17.026 had been excavated just inside the ring-ditch interior and was a steep-sided cut, 0.4m wide and 0.3m deep, with a rounded base. It contained a small quantity of cremated human bone (fill 38.17.047) from a single individual, although neither the age nor sex of this person could be determined as the majority of identifiable bone consisted of unidentified long bone fragments. This burial was sealed by upper fill 38.17.027 which comprised wood charcoal with small burnt clay fragments. The charcoal was again dominated by oak with significant quantities of birch. Human bone from this pit was radiocarbon dated to 1980–1770 cal. BC (SUERC-52594; 95% probability) whilst charcoal from the same fill was dated to 1900–1740 cal. BC (SUERC-52583; 95% probability). Pit 38.17.080 was immediately north of the cremation burial and had been cut into the ring-ditch fill, providing further evidence that the ditch had largely infilled by this time. It was circular in plan with steep sides and a flat base and was 0.7m wide and 0.2m deep. This pit contained a thin primary fill (38.17.082) overlain by silty clay 38.17.081. Samples from these fills yielded wood charcoal which produced a radiocarbon date of 1880–1690 cal. BC (SUERC-52588; 95% probability) and they were capped by stone layer 38.17.072.
2.9 Hearth 38.17.028 had been cut into the northern part of the infilled ditch at an equidistant point between the two cremation graves and comprised a circular cut 0.25m wide and 0.3m deep with steep sides and a flat base (Fig. 6, section GG). The surrounding substrate and ditch fills had been scorched revealing that burning had occurred within this feature and its lower fill, 38.17.029, was rich in wood charcoal and included fragments of burnt bone unidentifiable to species. The remainder of the hearth was filled with silty clay 38.17.030 which included a considerable quantity of wood charcoal, again, largely from oak, but including a significant quantity of birch, as well as a few worked flints, burnt clay and small quantities of burnt bone (again, not identifiable to species). Charcoal from the lower fill was radiocarbon dated to 1880–1680 cal. BC (SUERC-52581; 95% probability) and 1740–1520 cal. BC (Beta-253577; 95% confidence) whilst charcoal from the upper fill returned a radiocarbon date of 1890–1690 cal. BC (SUERC-52582; 95% probability).
9 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
2.10 Within the ring-ditch interior, and located between the two cremations, was a large posthole, 38.17.052. This was circular in plan, 1m wide and 0.4m deep (Fig. 6, section HH). It contained a rubbly post-packing fill surrounding a post-pipe which was 0.2m in diameter. The post pipe included two fills (38.17.054 and 38.17.055) and samples from these yielded wood charcoal and vitrified clay, in addition to which the uppermost fill (38.17.055) contained a very small amount of burnt bone, unidentifiable to species. Charcoal from the upper fill produced a radiocarbon date range of 1880–1690 cal. BC (SUERC-52584; 95% probable).
2.11 Within the southern part of the ring-ditch interior, pit 38.17.100 had been excavated (Fig. 6, section II). This was an irregular but steep-sided cut 2.2m long, 1.2m wide and 0.3m deep and contained a single stony fill which included wood charcoal and a few hazelnut shell fragments, one of which was radiocarbon-dated to 1780–1620 cal. BC (Beta-253578; 89% probability). Pit 38.17.056 (Fig. 6, section DD) was found to the north of pit 38.17.100 and, whilst technically undated, truncated the Phase A pit containing the halberd and may belong to Phase B.
Unphased pit within the ring-ditch 2.12 Pit 38.17.096 was located within the ring-ditch interior but was not dated.
Phase C: Middle Bronze Age c. 1500-1190 2.13 Three circular pits (38.17.003, 38.17.005 and 38.17.007; Fig. 12) were found 100m east of the ring-ditch. They were up to 0.5m wide and 0.1m deep with bowl-shaped profiles and contained charcoal-rich silty clay fills. Samples from these pits yielded large quantities of wood charcoal. Pit 38.17.003 also yielded a small quantity of burnt bone (not identifiable to species) and burnt clay whilst pit 38.17.005 contained a few sherds of early prehistoric pottery (fill 38.17.006). Although this pottery was tentatively identified as Neolithic or Early Bronze Age (See Appendix B), radiocarbon dates of 1500–1380 and 1420–1270 cal. BC (SUERC-52573 and -52574; 85% and 95% probability respectively) were obtained from charcoal within pit 38.17.005 and radiocarbon dates of 1420–1270 and 1400–1190 cal. BC (SUERC-52578 and - 52579; 95% probability) were obtained from charcoal within pit 38.17.007.
Discussion Introduction 2.14 The discovery of a prehistoric ring-ditch along the ridge of high ground occasions little surprise as early monuments are well attested on Mynydd Myddfai which
10 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
adjoins the ridge, and along the ridge itself. What makes the ring-ditch at Site 38.17 significant is the fact that it has been excavated to modern standards, including the production of radiocarbon dates examined through Bayesian analysis, and of course the presence of the halberd and haft fragment. The results are discussed below by phase. It should be noted that although the phases identified are based on spot dates, and might not represent the full temporal ranges of activity on the site, the fact that the majority of the features have been radiocarbon-dated suggests that the impression suggested below of episodic rather than continuous use of the site is correct (see Appendix E).
Phase A: Early Bronze Age/Beaker c. 2590–2200 cal. BC 2.15 The Trecastle ring-ditch at 8–9m in diameter is a relatively small example; that found further to the west along the pipeline at Site 513 was 11.5m in diameter (CA 2014). The exact form of both of these monuments is not certain, although there was some evidence that the ring-ditch at Site 513 may have included either an internal bank or a mound. Some ring-ditches surrounding Beaker burials have been interpreted as palisade trenches (Stuart Needham pers. comm. to Jonathan Hart) and the flat- based profile of the ditch at Site 38.17 lends itself to this interpretation. However, the ditch was shallow (0.3m deep) and probably fairly untruncated and so seems too slight to have held substantial posts. This, taken with the absence of packing stones or postholes along the ditch base, tends to suggest that the ditch was not a palisade trench. Instead, as with the ring-ditch at Site 513, there are hints that a mound may have been present at Site 38.17, although clear evidence for this was absent. The ditch fills were stony and could have derived from an internal mound whilst the erection of a post within the ring-ditch interior during Phase B had necessitated the excavation of a broad pit, despite the post having been no more than 0.2m in diameter, perhaps indicating that this posthole had been excavated through soft mound deposits and that the post therefore required packing with large stones to keep it from tilting. Slightly more convincing is the fact that the ring-ditch was re- used at least two centuries after Phase A, at a time when the ditch had at least partially filled in, suggesting that the monument remained visible, perhaps as a mound and/or earthwork ditch.
2.16 No dates were obtained for the construction of the ring-ditch. However, the radiocarbon dates from the Phase A pits were statistically consistent, suggesting that these depositions may represent a single event and perhaps, given the almost central location of the halberd and its significance, these can be seen as founder
11 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
deposits contemporary with the origin of the ring-ditch. The radiocarbon date from the charcoal within the ring-ditch fill was comparable to those from the pits although this must be regarded as redeposited material and can only provide a terminus post quem for the infilling of the ditch. Certainly, the ditch had infilled by Phase B, which is estimated to have begun between 1970–1810 cal. BC (95% probable). Although the date ranges for Phase A cover a broad span of up to 390 years, the limited stratigraphical evidence suggests that activity may have been more temporally restricted, perhaps to a single episode of construction (of the ring-ditch and internal Phase A pits) and a broadly contemporary episode of deposition (within the Phase A pits), following which the monument was left alone (or was used for ceremonies that left no archaeological trace) during which time the ditch infilled, either naturally or as the result of slighting.
2.17 No human remains were dated to this phase, but this is not unknown for ring- ditches, even for examples surviving as barrows (for example, Jones and Quinnell 2011, 214). It is possible that unburnt bones could have entirely eroded due to the local soil conditions or that no human remains were ever interred. Given this, the burial of the halberd (perhaps as a founder deposit as suggested above) may have been the significant event associated with the Phase A ring-ditch. What this deposition represented is unknown of course, but the halberd may have been symbolic of an individual and perhaps the ring-ditch during Phase A was regarded as a cenotaph, a monument to a dead person without the physical presence of a body. In light of this, it is of note that the halberd may have been deliberately damaged prior to burial and that other halberds do not seem to have been used as grave goods, but instead provided their own contextual meaning (see Appendix B).
Phase B: Early Bronze Age c. 1970–1680 cal. BC 2.18 All of the certain human remains and all of the unidentifiable burnt bone belonged to Phase B. By the time Phase B is estimated to have begun, the monument may have been untouched for up to 230 years (approximately nine generations), although it may have course have been visited or the scene of events which have left no archaeological trace. A distinct burial rite seems to be represented by the two certain cremation burials, each of which was paired with another pit. In each of these pairings, one of the pits (although not necessarily the cremation pit) was sealed with stone capping, and these capping deposits may have served as grave markers. The post erected within the ring-ditch interior is perhaps comparable to the possible totem pole suggested at the Buckskin Barrow, Hampshire (Darvill 2010, 199).
12 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Although the radiocarbon date for this posthole post-dates the removal of the post, there was no evidence that the post had not rotted in situ and it seems likely that it was instead deliberately removed, and then a token amount of burnt bone (not identified to species, but potentially human) was deposited within the resulting void. Although the posthole has been assigned to Phase B, this is on the basis of a radiocarbon date from this disuse deposit, and it is possible that it was a Phase A feature, dismantled in Phase B. If so, this could suggest that this posthole was paired with adjacent pit 38.17.064 whilst pit 38.17.056 could have been paired with the pit containing the halberd, in which case it would have been cut through the halberd pit in Phase A, not Phase B.
2.19 The cremation burials represented at least two individuals, perhaps more if the unidentified bone was human, although the unidentified bone could equally have come from either of the two identified burials. Only token amounts of the cremated individuals were deposited; this finds parallels with other Bronze Age cremation sites where analysis of bone weight clearly shows that the cremated remains were rarely buried in their entirety (McKinley 1997, 130), at least within a single location. The remaining parts of the cremations might have been buried elsewhere, scattered or curated, although unexpected uses are also possible (modern practices for example use cremation ashes in tattoos and jewellery). The small hearth might represent either the site of a pyre, or of a burning event associated with a burial rite or commemorative event.
2.20 The large proportion of birch charcoal within the cremation fills is notable (see Appendix D). This is a poor fuel for cremations, other than for use as kindling and so the reasons for its use are not readily apparent. It may be that it was simply readily available since birch is a primary coloniser of open landscapes, which may have been the case for the landscape at the time of the cremations (Rackham et al. forthcoming), or it may be that the birch came from coppiced trees and represents the remains of poles used to support the cremation pyre (indeed, roundwood was relatively common amongst the birch charcoal, as compared to amongst other species present). An alternative explanation is that birch was perhaps selected for other properties; for example, if silver birch was used, the colour of its bark may have been resonant with the colour of pyre smoke or burnt bone.
2.21 Statistical analysis of the radiocarbon dates from the Phase B features suggests that all features so dated may have been of the same actual age. However, this does not
13 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
make it certain that a single event is represented by Phase B and an alternative possibility is that the phase extended over a period of up 90 years (between 1970– 1680 cal. BC at 95% probability), a period of c. four generations. The overall dated lifespan of the ring-ditch, from the start of Phase A to the end of Phase B was 2590– 1680 cal. BC (95% probability). However, it is notable that if the monument’s use for cremation burials is considered alone (ie, Phase B), the date range is closely comparable to that of the ring-ditch found at pipeline Site 513, the dated use of which was between 2020–1700 cal. BC (95% probability; CA 2014). At Site 513, the majority of the cremations were found within urns, although un-urned examples were also present. However, too few data exist to allow for statements to be made regarding temporal relationships between the assemblage of urned and un-urned cremation burials from these two sites and the burial rites between the two sites seem to have followed different traditions.
Phase C: Middle Bronze Age c. 1500–1190 2.22 The three circular pits to the east of the ring-ditch date to at least 180 years after the last dated activity at the ring-ditch. It is unclear what these pits represent: the material within them could equally be domestic debris or, if the burnt bone was human, suggest that the pits were funerary features. In either instance, they were located away from the ring-ditch, perhaps suggesting that it remained visible or was remembered as a feature into this period. The pottery tentatively identified as Neolithic or Early Bronze Age may have been curated items but, more convincingly on the basis of the radiocarbon dates, was probably Middle Bronze Age.
2.23 These pits were located within 150m east of the three undated pits found at pipeline Site 38.20 (CA 2013) and although these might represent different areas of activity and indeed different periods, the overall impression is that archaeological remains are widely distributed across this ridge of high ground. The plant remains from the site provided little evidence as to the nature of the environment along the ridge at this time (Appendix C); what evidence there was indicated a mixture of woodland, hedgerows and scrub with some damp, possibly disturbed grassland.
14 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
3. PROJECT TEAM
The fieldwork was directed in the field by Mark Brett (Cotswold Archaeology). This report was written by Jonathan Hart with illustrations prepared by Daniel Bashford. The archive has been compiled by Jonathan Hart and prepared for deposition by Hazel O’Neill. The fieldwork was managed for CA by Clifford Bateman and the post- excavation was managed for CA by Karen Walker.
Stuart Needham’s assistance in co-ordinating the programme of analysis and reporting work on the halberd by specialists (including the team from the National Museum of Wales) is much appreciated. Adam Gwilts assisted in this. The line illustration of the halberd is by Anthony Daly of the National Museum Wales.
15 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
4. REFERENCES
Ashmore, P. 1999 ‘Radiocarbon dating: avoiding errors by avoiding mixed samples’, Antiquity 73, 124–30 Bell, D. 2014 ‘A radiocarbon date for an Irish Bronze Age halberd’, PAST 76, 15-6 Bourke, L. 2001 Crossing the Rubicon: Bronze Age Metalwork from Irish Rivers. Galway, National University of Ireland. Bronze Age Studies 5 Britton, D. 1963 ‘Traditions of metal-working in the later Neolithic and Early Bronze Age of Britain: part 1’, Proc. Prehist. Soc. 29, 258-325 BGS (British Geological Survey) 2013 Geology of Britain Viewer. Online resource at http://mapapps.bgs.ac.uk/geologyofbritain/home.html accessed 2 September 2013 Coles, J. M. 1968-9 ‘Scottish Early Bronze Age metalwork’, Proc.Soc. Antiq.Scotland 101, 1- 110 CA (Cotswold Archaeology) 2006 Felindre to Tirley Gas Pipeline: Archaeology and Heritage Survey. CA typescript report 05140 CA (Cotswold Archaeology) 2013 South Wales Pipeline Project, Site 38.20. Land North of Ffos-ddu, Llywel, Powys: Archaeological Watching Brief. CA typescript report 13316 CA (Cotswold Archaeology) 2014 South Wales Pipeline Project, Site 513. Land North-West of Steynton, Milford Haven: Archaeological Excavation. CA typescript report 13261 Darvill, T. 2010 Prehistoric Britain. 2nd Edition. Routledge, Abingdon EH (English Heritage) 1991 The Management of Archaeological Projects 2 Gale, R. and Cutler, D. 2000 Plants in Archaeology. Otley/London, Westbury Publishing and Royal Botanical Gardens, Kew Gibson, A. 2013 Milford Haven to Aberdulais and Felindre to Brecon Gas Pipeline: The Neolithic and Bronze Age Pottery from the Archaeological Investigations, Gibson typescript report no. 121 GA (Groundwork Archaeology) 2012 Milford Haven to Aberdulais and Felindre to Brecon High Pressure Gas Pipelines: Updated Project Design Hall, R. 2008a ‘Assessment for Ceramic Building Material’, in NLM 2012b Hall, R. 2008b ‘Assessment for Fired Clay and Daub’, in NLM 2012b IfA (Institute for Archaeologists) 1999a Guidelines for Finds Work. IfA, Birmingham IfA (Institute for Archaeologists) 1999b Standard and Guidance for Finds and Ecofact Studies and Curation. IfA, Reading IfA (Institute for Archaeologists) 2001a Standard and Guidance for the Collection, Documentation, Conservation and Research of Archaeological Materials. IfA, Reading
16 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
IfA (Institute for Archaeologists) 2001b Standard and Guidance for Archaeological Excavation IfA Wales (Institute for Archaeologists of Wales/Cymru) 2008 Introducing a Research Framework for the Archaeology of Wales, online resource at http://www.archaeoleg.org.uk/intro.html accessed December 2008 Jones, A. M. and Quinnell, H. 2011 ‘The Neolithic and Bronze Age in Cornwall, c 4000 cal BC to c 1000 cal BC: an overview of recent developments,’ Cornish Archaeology 50, 197–230 Lelong, O. and MacGregor, G. 2007 The Lands of Ancient Lothian: Interpreting the Archaeology of the A1. Edinburgh, Society of Antiquaries of Scotland Martin, G. 2008 ‘Assessment Report for the Archaeobotanical Remains’, in NLM 2012b McKinley, J. 1997 ‘Bronze Age ‘Barrows’ and Funerary Rites and Rituals of Cremation,’ in Proc. Prehist. Soc. 63, 129–145 NLM (Nacap Land and Marine) 2006 Milford Haven to Aberdulais Natural Gas Pipeline: Scheme of investigation for a programme of archaeological works NLM (Nacap Land and Marine) 2012a Milford Haven to Aberdulais High Pressure Gas Pipeline: Archaeology Assessment of Potential for Analysis NLM (Nacap Land and Marine) 2012b Felindre to Brecon High Pressure Gas Pipeline: Archaeology Assessment of Potential for Analysis Needham, S. 1988 ‘Selective deposition in the British Early Bronze Age’, World Archaeology 20, 229-48 Needham, S. P. 2002 ‘Analytical implications for Beaker metallurgy in North-west Europe’, in Bartelheim, M., Pernicka, E. and Krause, R. (eds), Die Anfänge der Metallurgie in der Alten Welt, 99-133. Freiberg, Forschungen zur Archäometrie und Altertumswissenschaf 1 Needham, S. P. with contributions by Mary Davis, Adam Gwilt, Mark Lodwick, Phil Parkes and Peter Reavill 2015 ‘A hafted halberd excavated at Trecastell, Powys: from undercurrent to uptake – the emergence and contextualisation of halberds in Wales and north-west Europe’, Proceedings of the Prehistoric Society 81 O’Brien, W. 2004 Ross Island: Mining, Metal and Society in Early Ireland. Galway, National University of Ireland. Bronze Age Studies 6 O’Flaherty, R. 1995 ‘An analysis of Irish Early Bronze Age hoards containing copper or bronze objects’, J. Soc. Antiq. Ireland 125, 10-45 Ó Ríordáin, S. 1937 ‘The halberd in prehistoric Europe: a study in prehistoric origins, evolution, distribution, and chronology’, Archaeologia 86, 195-321 Pannett, A. 2009 ‘Assessment Report for Lithics’, in NLM 2012b Rackham, J. 2009 ‘Assessment Report for Charcoal Remains’, in NLM 2012b
17 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Rackham, D. J., Challinor, D., Langdon, C. and Scaife, R. forthcoming Palaeoenvironmental studies along the Milford Haven to Aberdulais and Felindre to Tirley Natural Gas Pipeline. Southampton University Archaeology Research Series RSK (RSKENSR) 2007 Felindre to Tirley Natural Gas Pipeline: Archaeological Framework Document, v7. Nacap Land and Marine Final, RSKENSR Environmental Ltd Sheridan J. A. 2007 ‘From Picardie to Pickering and Pencraig Hill? New Information on the ‘Carinated Bowl Neolithic’ in northern Britain’, in Whittle, A. W. R. and Cummings, V. (eds) Gowing Over: the Mesolithic – Neolithic Transition in North-West Europe, 441- 92. Oxford, Oxford University Press (Proceedings of the British Academy 144) Vince, A. and Steane, K. 2008 ‘Assessment Report for Stone’, in NLM 2012b Whittle, A., Healy, F. and Bayliss, A. 2011 Gathering Time. Dating the Early Neolithic Enclosures of Southern Britain and Ireland. Oxford, Oxbow Books Wood, J. 2009 ‘Assessment Report for Faunal Remains’, in NLM 2012b
18 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
APPENDIX A: CONTEXT DESCRIPTIONS
Context Fill of Interpretation Description W Depth Spot date No. (m) (m) 38.17.001 Topsoil 38.17.002 Natural 38.17.003 Pit Circular, bowl-shaped profile 0.5 0.1 38.17.004 38.17.003 Pit fill Charcoal and burnt stone in 0.5 0.1 silty clay matrix 38.17.005 Pit Circular, bowl-shaped profile 0.35 0.1 38.17.006 38.17.005 Pit fill Charcoal and burnt stone in 0.35 0.1 1500–1380 cal BC silty clay matrix 1420–1270 cal BC 38.17.007 Pit Circular, bowl-shaped profile 0.4 0.05 38.17.008 38.17.007 Pit fill Charcoal and burnt stone in 0.4 0.05 1420–1270 cal BC silty clay matrix 1400–1190 cal BC 38.17.009 Posthole Modern posthole 0.4 0.1 Modern 38.17.010 38.17.009 Posthole fill Dark grey clay silt 0.4 0.1 38.17.011 Ring-ditch Ring ditch: steep edges and 0.7- 0.2- generic flat base. 1.3 0.6 number 38.17.012 Part of 38.17.011 0.85 0.2 38.17.013 38.17.012 Ring-ditch fill Dark grey-brown silty clay with 0.85 0.2 occasional small stones 38.17.014 Part of 38.17.011 0.7 0.3 38.17.015 38.17.014 Ring-ditch fill Red-brown silty clay with 0.7 0.3 occasional small stones 38.17.016 Part of 38.17.011 0.8 0.25 38.17.017 38.17.016 Ring-ditch fill Red-brown silty clay with 0.8 0.25 occasional small stones 38.17.018 Part of 38.17.011 0.8 0.35 38.17.019 38.17.016 Ring-ditch fill Upper fill: Red-brown silty clay 0.8 0.1 38.17.020 38.17.016 Ring-ditch fill 3rd fill: Red-brown silty clay 0.8 0.1 with charcoal 38.17.021 38.17.016 Ring-ditch fill Lower fill: Red-brown silty clay 0.8 0.05 38.17.022 38.17.016 Ring-ditch fill 2nd fill: stones along base of 0.8 0.05 ring ditch, above lower fill 38.17.023 Pit Within ring ditch circuit. 0.4 0.1 Circular in plan with gently sloping sides and flat base 38.17.024 38.17.023 Pit fill Orange-brown sandy silt with 0.4 0.1 1890–1740 cal BC charcoal 38.17.025 38.17.011 Ring-ditch Generally red-brown silty clay 0.7- 0.2- generic fill single fill with occasional small 1.3 0.6 stones 38.17.026 Pit Pit within ring ditch cut but 0.4 0.3 relationship not established. Circular, steep, undercut sides, rounded base 38.17.027 38.17.026 Pit fill Upper fill: Charcoal and small 0.4 0.15 burnt clay fragments 38.17.028 Hearth Circular, vertical sides, flat 0.25 0.3 base. Surrounding substrate was scorched 38.17.029 38.17.028 Hearth fill Lower fill: brown-black 0.4 0.15 1880–1680 cal BC charcoal-rich clay silt with burnt bone 38.17.030 38.17.028 Hearth fill Upper fill: grey-brown silty clay 0.4 0.15 1890–1690 cal BC with charcoal and burnt stones 38.17.031 Scorched substrate around hearth 38.17.028
19 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
38.17.032 Part of 38.17.011 1.3 0.3 38.17.033 38.17.032 Ring-ditch fill Mid brown silty clay with 1.3 0.3 occasional small stones and charcoal flecks 38.17.034 Part of 38.17.011 0.9 0.2 38.17.035 38.17.034 Ring-ditch fill Upper fill: mid orange-brown 0.9 0.15 sandy clay with occasional small stones and charcoal flecks 38.17.036 38.17.034 Ring-ditch fill Lower fill: mid brown silty clay 0.9 0.05 38.17.037 Part of 38.17.011 0.9 0.25 38.17.038 38.17.037 Ring-ditch fill Mid grey-brown silty clay with 0.9 0.25 2440–2200 cal BC occasional small stones and charcoal flecks; four flints found along inner edge of ring ditch cut 38.17.039 Context not used 38.17.040 Context not used 38.17.041 = 38.17.038 38.17.042 = 38.17.024 38.17.043 Part of 38.17.011 38.17.044 38.17.043 Ring-ditch fill Lower fill: mid red-brown silty 0.25 clay with occasional small stones 38.17.045 38.17.043 Ring-ditch fill 2nd fill: mid red-brown silty 0.15 clay with frequent charcoal 38.17.046 38.17.043 Ring-ditch fill 3rd fill: dark grey-brown sandy 0.05 silt 38.17.047 38.17.026 Pit fill Lower fill: Charcoal in dark 0.4 0.15 1980–1770 cal BC grey-brown clay silt matrix with 1900–1740 cal BC burnt bone 38.17.048 Part of 38.17.011 0.65 0.4 38.17.049 38.17.048 Ring-ditch fill Upper fill: mid red-brown silty 0.65 0.35 clay with occasional small stones 38.17.050 38.17.048 Ring-ditch fill Lower fill: pale silt primary fill 0.4 0.05 38.17.051 38.17.052 Posthole Large stones in orange-brown 0.4 packing fill clay silt 38.17.052 Posthole Circular with steep sides and 1.0 0.4 flat base 38.17.053 Context not used 38.17.054 38.17.052 Post pipe Lower part: yellow-brown silty 0.3 0.1 clay 38.17.055 38.17.052 Post pipe Upper part: orange-brown silty 0.3 0.15 1880–1690 cal BC clay 38.17.056 Pit Rectangular with steeply 1.8 0.75 sloping sides and flat base 38.17.057 38.17.060 Pit fill Brown-red clay silt with 0.7 0.5 2450–2200 cal BC frequent small and medium- sized stones and charcoal. Contained halberd 38.17.058 Pit Oval in plan, steep sides, flat 0.55 0.1 base. Truncated ring ditch 0.85 38.17.059 = 38.17.025 38.17.060 Pit Sub-circular with moderate 1.2 0.5 sides and flat base. Fill contained halberd 38.17.061 Context not used 38.17.062 Number assigned for sample of 38.17.025
20 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
38.17.063 Number assigned for sample of 38.17.025 38.17.064 Pit Circular in plan with steep 1.3 0.8 sides and rounded base 38.17.065 38.17.064 Pit fill Lower fill: pale grey clay silt 0.5 0.05 2460–2280 cal BC with occasional stones and charcoal 38.17.066 38.17.064 Pit fill 2nd fill: mid red-brown clay silt 1.2- 0.2 2470–2280 cal BC with small stones 1.3 2460–2280 cal BC 38.17.067 38.17.064 Pit fill 3rd fill: stone rubble 1.2- 0.2 1.3 38.17.068 38.17.064 Pit fill Upper fill: pale brown clay silt 1.2- 0.35 with small stones and charcoal 1.3 38.17.069 Context not used 38.17.070 Number assigned for sample of 38.17.025 38.17.071 38.17.052 Posthole fill Lower fill: pale brown clay silt 0.02 primary fill 38.17.072 38.17.080 fill Stone layer capping pit 080. 1.3 No structural form but limited in extent 38.17.073 Number assigned for sample of 38.17.025 38.17.074 Context not used 38.17.075 Context not used 38.17.076 Part of 38.17.011 0.6 0.6 38.17.077 38.17.076 Ring-ditch fill Mid brown silty clay with 0.6 0.6 occasional small stones and charcoal 38.17.078 Generic Geological feature or animal number burrow truncated by ring ditch 38.17.079 38.17.088 Fill Fill of geological feature 38.17.080 Pit Circular with steep sides and 0.7 0.2 rounded base 38.17.081 38.17.080 Pit fill Upper fill: mid orange-brown 0.7 0.15 1880–1690 cal BC clay silt with occasional charcoal 38.17.082 38.17.080 Pit fill Lower fill: pale brown clay silt 0.7 0.05 primary fill with occasional charcoal 38.17.083 38.17.058 Pit fill Dark brown clay silt with burnt 1890–1740 cal BC bone and charcoal 38.17.084 Part of 38.17.078 38.17.085 38.17.084 Fill Fill of geological feature 38.17.086 Part of 38.17.011 0.9 0.3 38.17.087 38.17.086 Ring-ditch fill Mid red-brown silty clay with 0.9 0.9 occasional small stones 38.17.088 Part of 38.17.078 38.17.089 = 38.17.083 38.17.090 38.17.056 Pit fill Upper fill: grey-brown silty clay 1.8 0.70 38.17.091 38.17.056 Pit fill Lower fill: pale grey silty clay 0.05 38.17.092 Context not used 38.17.093 Context not used 38.17.094 Part of 38.17.078 38.17.095 38.17.094 Fill Fill of geological feature 38.17.096 Pit Circular in plan with 0.3 0.05 moderately sloping sides and flat base 38.17.097 38.17.096 Pit fill Mid brown silty clay with 0.3 0.05 frequent stones and occasional charcoal
21 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
38.17.098 = 38.17.083 38.17.099 = 38.17.083 38.17.100 Pit Irregular in plan with steep 2.2- 0.3 sides and flat base 1.2 38.17.101 38.17.100 Pit fill Mid brown silty clay with 0.2 0.3 1780–1620 cal BC frequent stones and occasional charcoal 38.17.102 Number assigned for sample of 38.17.025 38.17.103 Part of 38.17.011 0.8 0.25 38.17.104 38.17.103 Ring-ditch fill Upper fill: mid red-brown silty 0.8 0.25 clay with occasional small stones 38.17.105 38.17.103 Layer Stone layer on inner edge of ring ditch, partially slumping into 38.17.103 38.17.106 Layer Buried soil above natural and sealed by 38.17.105: pale grey silt 38.17.107 38.17.103 Ring ditch fill Lower fill: mid red-brown silty 0.8 0.05 clay with occasional small stones 38.17.108 Natural feature 38.17.109 38.17.108 Natural feature 38.17.110 Part of 38.17.011 0.5 0.25 38.17.111 38.17.110 Ring-ditch fill Mid grey-brown silty clay with 0.5 0.25 occasional small stones and charcoal 38.17.112 Part of 38.17.011 (partially 0.3 excavated) 38.17.113 38.17.110 Ring-ditch fill Pale silt (partially excavated) 0.3 38.17.114 38.17.011 Ring-ditch fill Pale silt (partially excavated) 38.17.115 Part of 38.17.011 (partially 0.35 excavated) 38.17.116 38.17.115 Ring-ditch fill Pale silt (partially excavated) 0.35
22 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
APPENDIX B: THE FINDS INCLUDING HALBERD DESCRIPTION BY STUART NEEDHAM
The halberd by Stuart Needham, with contributions by Mary Davis, Rowena Gale and Phil Parkes The halberd (Figs 8-12) was lifted on a block of soil by Mary Davis at the request of CA and transferred to Phil Parkes, conservator at the Department of Archaeology, University College Cardiff. The metal edges and surface are very corroded. Cleaning resulted in the full exposure of a copper blade, the butt end of which was concealed by the remains of a wooden haft-grip attached by three thick rivets. The haft-grip has suffered in various ways and only extends a short distance beyond the limits of the metal hafting-plate. Both the alignment of the wood grain and the morphology of blade and rivets make it clear that this object is a halberd, a dagger-like blade set more-or-less perpendicularly into a long-handled haft, similar to the hafting of an axe (O’Ríordáin 1937). A full report on the halberd will be published in the Proceedings of the Prehistoric Society (Needham 2015).
Dimensions Total length 187mm; total weight 262.5g. Blade length (established from radiographs): 183mm; maximum width: 76mm; maximum thickness: 9.2mm. Haft-grip maximum length (with grain): 85.3mm; maximum breadth: 44mm; maximum thickness: 21mm.
General condition The object was buried in an adverse environment for copper but the thickness of the central blade has saved it from extensive internal corrosion. The penetration of surface corrosion has had a disproportionate effect on the thin cutting edges and much of these have crumbled away and t virtually none of the original sharp edges are intact, an exception being a stretch of 8mm along the upper edge. Much of the surface has a powdery appearance but roughly half of the surface presents a smooth ‘patina’ of olive- to fir-green colour. This ‘patina’ is evidently not sufficiently pristine to preserve much evidence of working and use: only two small patches of striations or grain-like impressions were noted and these are presumed to be grinding marks. Some of the damage to the wooden haft-grip and enclosed hafting-plate is believed to be recent, deriving from both the first impact of an excavation tool and subsequent distortions during drying out. The haft-grip is now badly split.
Blade section and form The consolidated object has a very ragged outline. Comparative evidence from well preserved halberds from north-west Europe shows that blade edges could be straight, gently curved or gently sinuous. The best deduction of shape draws on soil stains as well as the extant projections on the blade. Assuming relatively constant decay along the length of the object, the outlines of the soil stains indicate that the blade edges had very little curvature. It is certain that the midrib’s longitudinal axis passes through the central rivet hole.
The blade has a thick and broad midrib, its profile having a gently convex top and sides sloping down to an angle where they meet the blade wings. While the midrib is up to 9.2mm thick, the blade wings are only a maximum of 2.6 and 3.1mm thick; on average therefore the midrib stands proud of the wings by some 3mm on either face; this is not atypical. As seen on well preserved halberds, the final thinning of the blade to sharp edges is normally achieved by means of a neat bevel inset by a few millimetres; on more ornate weapons this is sometimes further backed by shallow furrows, sometimes called ‘blood grooves’. Very little of this part of the edge survives on the Trecastell blade and the most intact stretch of 8mm is bent up towards face 1 by about 30°. This stretch has a neat crease along the bend internally and it is possible that it bent along the line of a blade-edge feature such as a groove or step; on the opposite, stretched face, there are instead hints of two facets running parallel to the
23 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
edge and these too might be vestiges of former edge-backing features. The more truncated stretch of edge immediately tip-wards has a slight crease which is inset from and parallel to the features just noted. No other traces of morphological features have been noted on the blade.
The maximum width of the midrib is 37.5mm (both faces) and it tapers to about 8mm close to the tip. It tapers steadily with straight sides for most of its length, but becomes convex close to the tip; this end shape is likely to have been echoed by the cutting edges prior to corrosion loss. Viewed in long section, the midrib is remarkably constant in thickness for 90mm of its length, between 8.5 and 9.2mm. From a point a little outside the hafting line, it thins steadily, ultimately coming to the very thin butt that had become rotated through the above described damage. The profile of the hafting-plate faces appears to be slightly very slightly concave. The midrib thins more gradually towards the tip over a distance of 55mm and it is still nearly 4mm thick at the extant tip; slight undulations on face 1 are due to differential corrosion although this was downward in the ground.
Rivets The rivets are of thick ‘plug’ form but have been differentially eroded by corrosion; this is most acute for the lower rivet which is markedly atrophied towards face 2. The diameter of one rivet shank can be estimated from the radiographs to be around 9mm. The better preserved heads of the lower and central rivets on face 1 do not appear to the naked eye to be domes, but the profile radiographs do show modest convexity.
Butt/hafting plate form The butt is concealed inside the haft-grip and its shape has to be deduced from the face-view radiographs. These show the shape as fairly deeply arched with a somewhat tri-lobate tendency. The ‘lobes’ are rounded angles which curve round the rivet positions; straighter stretches run in between and probably also down towards the shoulders.
Hafting-plate/blade relationship Although we have the wood itself, there are various problems with its movement, and the haft-lines in the patinated surface are the best indication of original alignment; these are approximately 2° off the perpendicular to the long axis through the blade. They are also roughly parallel to a line through the lower two rivet holes and it may be deduced that the blade was intentionally mounted so as to tilt a little downwards.
The wooden haft-grip Most of the haft-grip, the part of the wooden haft directly in contact with either side of the metal hafting-plate, survives in a somewhat desiccated state. The wood grain is aligned perpendicular to the blade’s axis, the expected orientation for a halberd’s haft. The haft may have had some significant growth age at the time of manufacture, but the species concerned would not give rise to particularly old wood.
Metal composition by Mary Davis Analysis was carried out using a CamScan Maxim 2040 scanning electron microscope (SEM) fitted with an Oxford Instruments wavelength dispersive spectrometer (WDS) and ISIS energy dispersive spectrometer (EDS). Operating conditions employed a 30o take-off angle, and a 20kV accelerating voltage. All minor and trace elements were analysed using WDS. Three samples were taken from the halberd using a 0.9mm drill bit: one from the hafted end, one from near the tip and the third from one of the rivets. These were mounted in resin, polished flat and carbon coated. Each sample was analysed at least three times both by WDS and EDS, and the results were averaged for each sample. Most readings were generally consistent; the exception was antimony,
24 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
which appeared to be very unevenly dispersed within the metal. The analysis confirms that both blade and rivet are of unalloyed copper; no tin was detected. The principal minor elements present in both are arsenic (As), silver (Ag) and antimony (Sb), arsenic being by far the largest impurity, and it is significant that no nickel was detected. Although the antimony level in the blade is at the low end of the distribution, this impurity signature still conforms well to ‘A-metal’ (Needham 2002, fig 4) and it can be taken with some confidence that the metal derives from Ross Island, Co Kerry (O’Brien 2004).
Wood identification by Rowena Gale The sample included five fragments of desiccated wood. Four fragments were too degraded for sectioning; these were prepared for examination by exposing freshly fractured surfaces. The remaining fragment was in a better state of preservation and thin sections were prepared using standard techniques (Gale and Cutler 2000). The wood structure was examined using both transmitted and incident light on a Nikon Labophot-2 compound microscope at magnifications up to x400 and matched to prepared reference slides of modern wood.
The wood structure was consistent with that of the Pomoideae, a subfamily of the Rosaceae. Members of this group include Malus sp. (apple), Pyrus sp. (pear), Crataegus sp. (hawthorn) and Sorbus sp. (rowan, whitebeam and service). These taxa are anatomically similar and it is not possible to identify the sample more positively. The wood of these species is hard, close-grained and was traditionally used for tool handles.
Radiocarbon determination A sample of the haft-grip was submitted for radiocarbon assay. The result returned was calibrating to 2470–2200 cal. BC (Beta-240338; 95% probability). The species concerned makes it unlikely that there would be a large discrepancy between the date of the growth rings and the date of hafting the halberd. It had been hoped to obtain at least one additional radiocarbon determination, but unfortunately the wood had all been consolidated by the time the present study began and a sample submitted to the Oxford Laboratory yielded insufficient carbon for dating.
The state of the object on deposition Little can be ventured about the ancient condition of the object. Virtually nothing of the cutting edges survives in good enough condition to assess ancient damage except to note that the one limited stretch with intact sharp edge is bent towards face 1 and this must be pre-corrosion damage. The whole of this 8mm stretch is bent and is not just a notch formed by having been struck by another sharp edge so might imply more systematic de- commissioning. This might be reinforced by indications that an adjacent stretch of edge seems to have been bent in the opposite direction.
Discussion The halberd is a find of considerable importance as it is only the second from Britain to have been recovered during archaeological excavation (the other being from Eweford, East Lothian; Lelong and MacGregor 2007, 235- 6), since part of its wooden haft has survived and since the presence of organic material offered the first opportunity to obtain a radiocarbon date in direct association with a halberd in north-western Europe. More recently, a small fragment of haft on a halberd found in Lough Ree, Ireland, has allowed a second independent date to be obtained (Bell 2014) and these have important implications for the still debated issue of the absolute chronology of halberds. This discussion will begin by assessing the typological place of the halberd within the wider halberd repertoire and, conversely, the implications of this find for the wider series.
25 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Despite the fact that most of its features taken individually are very familiar amongst European halberds, there are relatively few close matches for the combination of features found on the current example. Amongst the rarer features are the short and squat blade, and the broad midrib. Ó Ríordáin’s (1937) recognised the potential importance of size, without making it a wholly defining attribute; his types 1, 2 and 3 were predominantly made up of short-bladed weapons, whereas types 4, 5 and 6 comprised longer ones. For Ó Ríordáin, the shorter types were seen early. The combination of the three rivets and deeply arched butt for the halberd from Site 38.17 is paralleled within the Irish halberd corpus. The smaller of these, Type Roscrea, are very close to Ó Ríordáin’s type 3 and includes examples with broad midribs. Only six Roscrea halberds have any kind of provenance and these are widely spread from Co Antrim to Co Cork. Overall, the broader-midribbed halberds are all from the southern half of the Ireland and it be significant that the example exported to Site 38.17 is a broad-midrib weapon. In contrast to Ireland, very few halberds from Britain approach the small size of that from Site 38.17. The smallest, a three-rivet halberd from Falkland, Fife (Ó Ríordáin 1937, fig 55.2), is already about 40mm longer. As in Britain, squat halberds are not a regular feature anywhere in continental north-west Europe.
Halberds were in use in Britain only during the Chalcolithic (2450/2400 – 2200/2150 BC) and the earlier part of the Early Bronze Age (c. 2200/2150 – 2050 BC), a period of perhaps only three or four centuries. The halberd from Site 38.17 may represent an early example of Irish influence, given the calibrated date from the haft, but we should be wary of assuming that this represents Britain’s first acquaintance with the metal halberd, given the influential influx of people and concepts coming up from the south during the third quarter of the third millennium BC.
While the handle does not survive, it is clear from the shaping of the butt and torn wood fibres at the bottom that it entered the grip in line with the front with the butt forming a significant backward projection. This is paralleled by Irish and continental examples. The second point to draw from the haft is that the grip effectively comprises two separate plates of wood which sandwich the metal hafting-plate. This should not be too surprising since the hafting-plates of halberds are not only fairly deep (along the longitudinal axis), but are also rather thin and it would be incredibly difficult to carve out a recess in a single piece of wood that fitted snugly round the hafting- plate. The function of the rivets in this arrangement would thus be twofold; securing the blade against twisting and wrenching, but at the same time tightly clamping the ‘sandwich’.
It has long been apparent that halberds were not seen as fitting accompaniments for burials and they were instead usually deposited in their own self-contained contextual world (Needham 1988). They occur either singly or in small hoards (Ó Ríordáin 1937; Britton 1963; Coles 1968-9) and often in wet places (e.g. O’Flaherty 1995; Bourke 2001). The only halberd in a secure context on a funerary site was excavated at Eweford, East Lothian. It was deposited in the eastern edge of cairn material which had been added in the Early Bronze Age to an earlier Neolithic monument (Lelong & MacGregor 2007, 235-6); it did not accompany any human skeletal remains. The regions yielding halberds frequently lack obvious contemporary burials.
26 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
The Prehistoric Pottery (Gibson 2013) Weight (g) No of Contexts Periods Represented 32 1 Early Neolithic ?, Neolithic/Bronze Age?
Early Neolithic 38.17.006 Six sherds + crumbs (22g). Undecorated sherds in a medium-hard black fabric. The sherds average 5mm thick and contains crushed quartz inclusions up to 3mm cross. Early Neolithic ? It is only the fineness of the fabric and the crushed quartz inclusions that suggest that this material may be Early Neolithic. The absence of rim or other diagnostic formal or decorative traits do not allow the vessel to be further classified. It must suffice to identify the sherd as coming from Carinated Bowl or Developed/Modified Carinated Bowl as discussed by Sheridan (2007). Dates for this material in Wales span the whole of the earlier Neolithic from c.4000-3400 cal BC (Whittle et al. 2011).
Neolithic or Early Bronze Age 38.17.006 Four sherds (10g) in a soft black fabric averaging 7mm thick. In the absence of formal or decorative features, this sherd cannot be ascribed to a specific tradition however the grog-filled fabric combined with the thickness of the sherd may suggest that it is either Late Neolithic Grooved Ware or Early Bronze Age Collared Urn. This permits only a broad date range from c.3000 – 1500 BC.
Flint (Pannett 2009) Nine struck lithics were recovered. These included four flakes, two flake fragments and two chunks which were not closely dateable, and a single flake from a blade core (fill 38.17.038) which is diagnostic of Mesolithic/Early Neolithic technologies.
Stone (Vince and Steane 2008) Three fragments (13g) of burnt stone were recovered from pits 38.17.026, 38.17.064 and posthole 38.17.052. These included a fragment of burnt shale from pit 38.17.026 which retained soot.
Ceramic building material (Hall 2008a) Eight small fragments of ceramic building material (amounting to 1g) were recovered from fill 38.17.004 (pit 38.17.003). These are not dateable but are presumably intrusive.
Fired clay (Hall 2008b) 11 fragments (8g) of fired clay were recovered from posthole 38.17.051 (fill 38.17.051). The fabrics are all oxidised, sandy fabrics, with sparse organic temper on the few diagnostic fragments.
27 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
APPENDIX C: THE HUMAN REMAINS BY ANNA FOTAKI AND MALIN HOLST
Contexts containing cremated bone are summarised in Table 1. The assemblages were split into two groups; those containing identifiable human bones (Table 2) and those containing only unidentifiable bones (Table 3). The identifiable human bone was found to be moderately well to badly preserved. Much bone surface detail had been lost, with many fragments exhibiting worn edges. Bone cracking was also evident, but warping was limited or non-existent. Most burials were highly fragmented, with few containing large amounts of larger fragments (Table 2). Five of the cremated bone assemblages contained bone fragments that were 10mm in size or larger (see Table 2). Burial 38.17.026 contained mostly fragments larger than 10mm, while the majority of the bone from the contexts in pit 38.17.058 was recovered from the 5mm sieve. There were three smaller bone assemblages from the site. They were minute in weight and due to their poor preservation and small quantity it was not possible to determine species (Table 3). In the case of all burials it is unclear whether post-depositional or post-burning disturbance of the bone caused the fragmentation.
The quantity of cremated bone recovered from the five contexts containing identifiable human bone varied in weight from 16g to 111g (see Tables 1 and 2), with an overall mean weight of 43.14g. All of the burials contained much less than the quantity of bone expected from modern cremations, suggesting that only a portion of the individual’s remains were necessary for interment, or that later disturbances resulted in the truncation of the burials. The majority of the bones were well burnt, causing the complete loss of the organic portion of the bone and producing a grey or white/off white colour in most of the burials (Table 2). Burial 28.17.026 was the only burial with different colour (red/brown), likely due to soil staining. It was possible to identify between 25.5% and almost 65% of the skeletal elements in the cremation burials (Table 4), with an average of 46% of bone being identifiable. All of the burials in Table 4 could be positively identified as human.
The majority of identifiable bone consisted either of skull elements or of long bone, including a number of tooth and tibia fragments. The majority of identifiable bone from Burial 38.17.026 consisted of unidentified long bone fragments. By contrast, deposit 38.17.083 in pit 38.17.058 contained more skull fragments, including tooth fragments, parts of the temporal and parietal skull vault. The remainder of contexts in the pit contained mostly unidentified long bone fragments.
MNI For pit 38.17.58 (containing contexts 38.17.083, 38.17.089, 38.17.098 and 38.17.099), no duplicated elements were identified, which suggests that only single individual was buried here. Burial 38.17.026 also appears to be of a single individual.
Age Age could not be accurately determined from the remains of any of the burials, because the ageing criteria that are normally used, did not survive. Deposit 38.17.099 in pit 38.17.058 contained a fragment of a hand phalanx, as well as a vertebral facet and tooth fragments suggesting an age of adolescent or older (over 16 years of age). Deposit 38.17.099 in the same pit contained two complete incisor tooth roots, which also hint to an older individual (adolescent or older). None of the other burials contained any age specific markers to provide an idea of age (Table 5).
Sex
None of the assemblages contained any skeletal elements which were sexually dimorphic.
28 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Pathology Pathology was not identified in any of the skeletal assemblages.
Dentition A few dental fragments survived. In particular contexts 38.17.083 and 38.17.099 in Burial 38.17.058 contained a total of eight tooth fragments.
Non-Metric Traits Non-metric traits were not identified in any of the burials.
Summary and Funerary Ritual The osteological analysis has revealed that all of the cremated bone assemblages were relatively well burnt, with a loss of bone surface detail, a high degree of fragmentation and some cracking. No duplicate bone elements were found from all of the contexts from Pit 38.17.58, so this probably only represents one individual. If this were the case, remains from contexts 38.17.083 and 38.17.099 indicate that the individual was older than 16 at the time of death. Sex could not be estimated for any of the individuals.
Table 1 Summary of cremated bone assemblages Context Feature Bone Colour Preservation Weight (g) % of Expected Quantity of Bone 38.17.030 Hearth 38.17.028 Grey/White Very Poor <0.0 - 38.17.047 Pit 38.17.026 Red/Brown Moderate 25.0 1.5 38.17.055 Pit 38.17.052 Red Very Poor <0.0 - 38.17.062 Ring ditch White/Brown Very Poor <0.0 - 38.17.083 Pit 38.17.58 Grey/White Moderate 111.0 6.9 38.17.089 Pit 38.17.58 White/Off white Poor 18.7 1.2 38.17.098 Pit 38.17.58 White/Off white Poor 16.0 1 38.17.099 Pit 38.17.58 Grey/White Poor 45.0 2.8
Table 2 Summary of cremated bone fragment size from pit 38.17.058 Context Feature 10mm 10mm 5mm 5mm 2mm 2mm < 2mm < 2mm Weight (g) (%) (g) (%) (g) (%) (g) (%) (g) 38.17.047 38.17.026 10.8 43.2 7.7 30.8 4.7 18.8 1.8 7.2 25 38.17.083 38.17.058 25.7 23.1 44.4 40 38.2 34.4 3.2 2.9 111 38.17.089 38.17.058 2.2 11.7 7.3 39 6.9 36.9 2.3 12.3 18.7 38.17.098 38.17.058 1.8 11.2 7.2 45 5.6 35 1.4 8.8 16 38.17.099 38.17.058 4.6 10.2 20 44.4 18.1 40.2 2.3 5.1 45
Table 3 Summary of smaller cremations found on site Context Feature Max frag length (mm) Total frags Weight (g) 38.17.030 38.17.028 3.1 3 <0.0 38.17.055 38.17.052 3.9 1 <0.0 38.17.062 38.17.025 2.1 5 <0.0
29 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Table 4 Summary of identifiable elements in the cremation burials Context Feature Skull Skull Axial Axial UL UL LL LL UIL UIL Tot. Tot. Tot. Tot. (g) (%) (g) (%) (g) (%) (g) (%) (g) (%) ID (%) UID UID (g) (%) 38.17.047 38.17.026 4.1 16.4 1.2 4.8 3.5 15 7.4 29.6 16.2 64.8 8.7 34 38.17.083 38.17.058 30.8 28 0.4 0.4 0.3 0.3 4.5 4 22.2 20 58.2 52.4 53.4 48 38.17.089 38.17.058 6.9 37 2.8 15 9.7 52 10.6 56 38.17.098 38.17.058 1.2 7.5 0.4 2.5 1.8 11.2 2.6 16.3 6 37.5 10 62 38.17.099 38.17.058 3.2 7.1 0.9 2 7.4 16.4 11.5 25.5 31.1 69
Table 5 Summary of osteological results Context Feature Preservation MNI Species Age Sex Weight Period (g) 38.17.030 38.17.028 Very Poor - Unknown - - <0.0 BA 38.17.047 38.17.026 Moderate 1 Human - - 25 BA 38.17.055 38.17.052 Very Poor - Unknown - - <0.0 BA 38.17.062 38.17.025 Very Poor - Unknown - - <0.0 BA 38.17.083 38.17.058 Moderate 1 Human Older than adolescent - 111 BA (16 years +) 38.17.089 38.17.058 Poor 1 Human - - 18.7 BA 38.17.098 38.17.058 Poor 1 Human - - 16 BA 38.17.099 38.17.058 Poor 1 Human Older than Adolescent - 45 BA
References Browne, D. and Hughes, S. 2003. The Archaeology of the Welsh Uplands, Royal Commission on the Ancient and Historical Monuments of Wales Margary, I.D. 1973. Roman Roads in Britain 3rd edition (London) McKinley, J.I. 1994. ‘Bone fragment size in British cremation burials and its implications for pyre technology and ritual’, Journal of Archaeological Science 21: 339-342
30 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
APPENDIX D: THE PALAEOENVIRONMENTAL EVIDENCE BY JAMES RACKHAM
Bone Small assemblages of burnt bone were recovered from three environmental samples – 38.17.001 (context 38.17.004), 38.17.005 (context 38.17.029) and 38.17.027 (context 38.17.083) - and were not classified as possible cremations so are considered here. Sample 38.17.001 produced 97 (+uncountable very tiny fragments) very small fragments of indeterminate burnt bone (2g). Sample 38.17.005 produced 8 very small fragments of indeterminate burnt bone (0.2g). Sample 38.17.027 produced over 200 very small fragments of indeterminate burnt bone (15g). All of these fragments could derive from the human cremation assemblages discussed above, particularly those from sample 38.17.027.
Environmental soil samples Thirty nine environmental samples were taken at this site from a range of the features. Three samples were taken from pits 38.17.003, 38.17.005 and 38.17.007 which lay some 100m east of the barrow site illustrated in Fig. 1. Seven samples were taken from the fills of the barrow ditch, from various sections across it, while the remainder were taken from features cut into the interior, and three features cut into the ditch fills, a hearth, 38.17.028 and two cremation pits, 38.17.026 and 38.17.058, (Table 6 and Fig. 1). Radiocarbon dates were obtained from material from fifteen of the samples (Table 6) indicating a date range for activities associated with the barrow throughout the early Bronze Age, approximately a six hundred years, the later features, pits 38.17.052 and 38.17.100, must have been excavated through the barrow mound to be visible on the truncated floor of the barrow (Fig. 1). Several of the features clearly represent activity much later than the original barrow construction, including an episode of features cut into the barrow ditch fills including a cremation pit, and at least two episodes of excavation of features through the barrow mound (see Griffiths 2014). The three isolated pits to the east of the barrow clearly indicate a much later phase of activity in the mid-late Bronze Age (Table 6).
The samples from the pits from the earliest phase of activity on the barrow site are lacking any archaeological finds other than a magnetic fraction that was recovered from the residue and this may have been no more than baked or concreted mudstone. No cremated bone was recovered from any of these features so thjeir function remains obscure. The deposits from the barrow ditch have produced little more with a little burnt bone from context 38.17.062, and burnt stone and ‘clinker’ from 38.17.036.
Pits 38.17.026 and 38.17.023 represent a second phase of activity within the barrow ditch. Both pits were excavated just inside the ditch (Fig. 1) and pit 26, on the basis of the identified human bone within its fills, is a cremation pit, but unfortunately no finds were recovered from the two samples from pit 23. Pit 26 had a ‘rich’ magnetic component composed of concreted mudstone crumb but no evidence for burning in situ.
The next group of features were cut into the barrow ditch fills and included pits 80 and 58 and hearth 28 (Fig. 1). Pit 58 is clearly a cremation pit, and produced the bulk of the identifiable human bone recovered from the site plus the largest quantity of unidentified burnt bone, a little material recorded during processing as ‘clinker’ was also found in three of the samples from this feature. Pit 80 produced no finds (Table 7), just a little magnetised material. Hearth 28 produced burnt bone, a ;little worked flint, ‘slag’ and a magnetic fraction composed largely of concreted mudstone crumb which may have been heat effected. The lower fill of the hearth, 38.17.029, included a very large charcoal component (Table 7).
31 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
The latest phase of activity within the barrow ditch is indicated by the fills of pits 52 and 100. A very small amount of unidentifiable burnt bone was recovered from postpipe fill, 055, in pit 52 and a little ‘vitrified clay’, while pit 100 produced no finds other than a little magnetic material. The date obtained from the upper fill of post-pipe 38.17.053 may reflect later collapse or intrusion after decomposition of the post and this feature may relate to an earlier phase of activity, while its fills reflect the latest deposits in the barrow. There is no evidence that pit 100 was a cremation pit, although it was large enough for an inhumation burial and must have been dug through the barrow mound and represents a significant intrusion into the barrow
Of the features excavated within the barrow and its ring ditch only two can confidently be identified as cremation pits, 26 and 58, and unless the calcined human bone has failed to survive in the soils of the other features, the bone rich deposits have been truncated, or the samples were taken from areas of the fill with no bone (which is unlikely given that several of the samples represented between 50% and 100% of the sampled deposit) these features would appear not to have been cremation burials.
Three pits some 100m east of the barrow were sampled (Table 6), and of these undated pit 003 produced a little unidentifiable burnt bone, a little fired clay, a small magnetic fraction and a very large charcoal component (Table 7). Pit 005 produced a few pieces of pottery, while pit 007 no archaeological finds, other than a little magnetic material.
The environmental assemblages were a little more productive and the charred plant remains and several charcoal assemblages have been studied in detail.
32 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Location of samples taken at the barrow site. Three other samples, were taken 100m to the east. Pit 108 is not located
33 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Table 6. The environmental samples from site 38.17
sample feature vol proc. context description wt kg proc.
38.17.0001 38.17.004 38.17.003 Pit fill 15 17 38.17.0002 38.17.006 38.17.005 Pit fill 38.17.0003 38.17.008 38.17.007 Pit fill 4 4 38.17.0004 38.17.024 38.17.023 Pit fill 2 2 38.17.0005 38.17.029 38.17.028 Hearth lower fill 26 22 38.17.0006 38.17.030 38.17.028 Hearth upper fill 8 11 38.17.0007 38.17.027 38.17.026 Pit upper fill 7 6 38.17.0008 38.17.036 38.17.034 Ring ditch fill 11 14 38.17.0009 38.17.042 38.17.023 Pit fill 7 8 38.17.0010 38.17.047 38.17.026 Pit lower fill – cremation? 6 5 38.17.0011 38.17.047 38.17.026 Pit lower fill – cremation? 7 7 38.17.0012 38.17.027 38.17.026 Pit upper fill 2 1 38.17.0013 38.17.055 38.17.052 Post pipe upper fill 24 29 38.17.0014 38.17.054 38.17.052 Post pipe fill 8 9 38.17.0015 38.17.057 38.17.060 Pit fill (halberd dated) 15 20 38.17.0016 38.17.059 38.17.011 ring ditch fill 32 44 38.17.0017 38.17.062 38.17.011 ring ditch upper fill 32 42 38.17.0018 38.17.061 38.17.060 Pit fill 5 7 38.17.0019 38.17.063 38.17.011 ring ditch upper fill 32 40 38.17.0020 38.17.070 38.17.011 ring ditch lower fill 28 37 38.17.0021 38.17.051 38.17.052 Post hole packing fill 28 36 38.17.0022 38.17.071 38.17.052 Post hole fill 38.17.0023 38.17.073 38.17.011 ring ditch lower fill 30 38 38.17.0024 38.17.057 38.17.060 Pit fill (halberd dated) 32 42 38.17.0025 38.17.081 38.17.080 Pit fill 12 12 38.17.0026 38.17.082 38.17.080 Pit fill 6 7 38.17.0027 38.17.083 38.17.058 Pit fill – cremation? 11 13 38.17.0028 38.17.089 38.17.058 Pit fill – cremation? 4* nd 38.17.0029 38.17.065 38.17.064 Pit fill 12 11 38.17.0030 38.17.066 38.17.064 Pit fill 38 42 38.17.0031 38.17.066 38.17.064 Pit fill 40 49 38.17.0032 38.17.098 38.17.058 Pit fill – cremation? 5 6 38.17.0033 38.17.099 38.17.058 Pit fill – cremation? 5 1 38.17.0034 38.17.101 38.17.100 Pit fill (hazelnut dated) 32 48 38.17.0035 38.17.090 38.17.056 Pit fill 21 28 38.17.0036 38.17.091 38.17.056 Pit fill 6 10 38.17.0037 38.17.102 38.17.011 ring ditch fill 28 36 38.17.0038 38.17.109 38.17.108 Pit fill 6 4 38.17.0039 38.17.106 Buried soil beneath 105 28 32 * volume recorded on site – not accurate
34 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Table 7. Data for the environmental samples from Site 38.17 in broad chronological order
1st 2nd char- vol. flot flot coal crem. proc vol. Burnt Fired Worked Magnetic sample context human slag coal sample comments bone clay flint material bone
Barrow ditch 38.17.0008 38.17.036 11 5 1 C burnt stone-D; clink E 38.17.0016 38.17.059 32 500 15 +0.2g 38.17.0017 38.17.062 32 20 35 E 38.17.0019 38.17.063 32 105 60 230 D 38.17.0020 38.17.070 28 25 25 C 38.17.0023 38.17.073 30 300 26 C 38.17.0037 38.17.102 28 80 12 Pit 60 38.17.0015 38.17.057 15 10 10 C+0.3g 38.17.0018 38.17.061 5 4 2 D 38.17.0024 38.17.057 32 5 13 C Pit 56 38.17.0035 38.17.090 21 16 3 D 38.17.0036 38.17.091 6 1 1 E Pit 64 38.17.0029 38.17.065 12 5 3 C 38.17.0030 38.17.066 38 36 11 38.17.0031 38.17.066 40 50 13 Coal ash D Pit 26 38.17.0007 38.17.027 7 280 15 300 A+0.4g 38.17.0010 38.17.047 6 260 32 A A+0.6g E 38.17.0011 38.17.047 7 600 nd Res. all charcoal 38.17.0012 38.17.027 2 105 14 0.1g Pit 23 38.17.0004 38.17.024 2 30 17 D 38.17.0009 38.17.042 7 42 7 D Coal ash? E Pit 58 38.17.0027 38.17.083 11 85 22 A/15g A C Clinker E 38.17.0028 38.17.089 4~ 24 4 D Coal ash/clinker D 38.17.0032 38.17.098 5 10 8 B 38.17.0033 38.17.099 5 10 5 A Clinker C Pit 80 38.17.0025 38.17.081 12 200 nd B 38.17.0026 38.17.082 6 10 4 D Hearth 28 38.17.0005 38.17.029 26 900 30 3400 E A+0.4g 38.17.0006 38.17.030 8 105 68 E E E D+1.6g Pit 52 38.17.0013 38.17.055 24 30 12 E A vitrified clay D 38.17.0014 38.17.054 8 5 6 38.17.0021 38.17.051 28 2.5 nd D Vitrified clay E 38.17.0022 38.17.071 ? 2.5 1 Pit 100 38.17.0034 38.17.101 32 50 2 C Pit 108 38.17.0038 38.17.109 6 20 6 Charcoal rich res. Palaeosol?
35 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
1st 2nd char- vol. flot flot coal crem. proc vol. Burnt Fired Worked Magnetic sample context human slag coal sample comments bone clay flint material bone
38.17.0039 38.17.106 28 15 6 Pit group east of the barrow 38.17.0001 38.17.004 15 1300 200 2000 D E D+0.6g E Charcoal rich res. 38.17.0002 38.17.006 8 800 22 38.17.0003 38.17.008 4 200 3 D * abundance rating – E= 1-10 items; D=11-50, C=51=100, B=101-200, A=>200; # weight in grammes; nd – no data; ~ volume recorded on site. (weights in the magnetic column reflect additional material recovered from the <2mm residue after refloating)
Charred Plant remains (John Giorgi) Twelve samples from ten features, largely pits, and dated to the early and middle Bronze Age, produced identifiable charred plant remains (Table 8 - in addition to charcoal) mainly represented by only small amounts of material with occasional Corylus avellana (hazel) nutshell fragments in six features including early Bronze Age pit [100] and Hearth [28], and middle Bronze Age pit [5]. There were relatively rich botanical assemblages in the fills of two pits [3] and [5] (undated and middle Bronze Age) circa 100m east of the ring-ditch. Both contain large amounts of charcoal and herbaceous stem (grass-sized culm nodes, bases and internode) fragments with occasional bud and moss fragments in Pit [5], and grassland/disturbed ground plants, Carduus/Cirsium (thistles), Danthonia decumbens (heath grass) and Cyperaceae (sedges), possibly reflecting a damp disturbed grassland environment in the immediate vicinity of these features. Another fairly rich plant assemblage was found in early Bronze Age charcoal rich fills of Hearth [28] (cut into the ring-ditch) which produced evidence indicative of a hedgerow/scrubland habitat, with poorly preserved and fragmentary Prunus fruit stones including Prunus spinosa (blackthorn) and Rubus (brambles) seeds including Rubus idaeus (raspberry); the fruit stones showed evidence of having been gnawed. The charred plant remains provide only limited information on the economy and environment of the site although there is tentative evidence for both woodland/hedgerow/scrub and possibly damp disturbed grassland habitats; the hazel nutshell, fruit stones and brambles may have been gathered for food, wild food resources continued to play an important role in the early prehistoric diet, while the herbaceous stem fragments may point to the collection of grassland vegetation by uprooting for various uses on site, for example, as fodder, flooring, roofing or fuel.
36 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Table 8. Site 38.17. The charred plant remains
Flot 1st 2nd 1st 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 1st 2nd 1st 2nd 1st 2nd 1st 1st 2nd Period EBA? EBA? EBA EBA? EBA? EBA? EBA? EBA EBA EBA? EBA? EBA EBA EBA EBA EBA EBA EBA Und. MBA MBA Feature type PAL PAL PIT PIT PIT PIT PIT PIT PIT PIT PIT PIT Hearth Hearth Hearth Hearth PIT PIT PIT PIT PIT Feature no. - - 60 56 56 56 56 64 64 23 23 80 28 28 28 28 100 100 3 5 5 Context no. 106 106 57 90 90 91 91 66 66 42 42 81 29 29 30 30 101 101 4 6 6 Sample no. 39 39 24 35 35 36 36 30 30 9 9 25 5 5 6 6 34 34 1 2 2 Proc. Vol. (l) 28 28 32 21 21 6 6 38 38 7 7 12 26 26 8 8 32 32 15 8 8 Vol. flot (ml) 15 8 7 16 2 1 2 38 11 44 5 210 900 34 107 156 16 3 1300 800 22 hazel nut Corylus avellana L. 3 1 1 4 1 2 5 shell frags Rubus idaeus L. raspberry 5 23 Rubus spp brambles etc. 8 27 Prunus spinosa L. blackthorn 1 indet cherries Prunus spp 1 etc. Indet Prunus spp 2 fragments Indet cf. Prunus spp 17 fragments small round cf Fabaceae indet. 1 1 1 cotyledon cf. Euphorbia ?sun spurge 1 helioscopia
Carduus/Cirsium sp. thistles 1 cf.Carduus/Cirsium ?thistles 3 spp.
Asteraceae indet. daisy family 4 Cyperaceae indet. sedges etc 1 Danthonia heath grass 1 2 1 decumbens (L) DC
Bryophyta indet. moss + Herbaceous stem culm 28 42 7 fragments nodes/bases
Indet bud fragments 1 1 Indet seeds 2 3 6 1 1 Charcoal >2mm/<2mm +++++ +++++ +++++ +++++ +++ +++ ++ +++++ +++++ +++++ +++++ +++++ +++++ +++++ +++++ +++++ +++ +++ +++++ +++++ +++++ Total nos. of items - 3 1 2 - - 1 1 - 1 - 2 31 4 57 - 1 2 48 47 9
37 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Charcoal (Dana Challinor) Thirty-two samples from features associated with the ring ditch were examined. Standard methodological procedures were followed, with 20-30 fragments identified (depending upon diversity). In general, diversity was low and of the 930 fragments studied, only four taxa were positively identified; Quercus sp. (oak), Betula sp. (birch), Corylus avellana (hazel) and Cytisus/Ulex (broom/gorse) (Tables 9 and 10). Preservation was variable, with some abundant assemblages and large fragments with a clean and clear structure, while others were sparse, with small-sized fragments, covered with sediment. Tyloses were frequently observed in oak, and the absence of significant ring curvature indicated that most fragments derived from trunk or large branchwood. Many of the oak fragments were comminuted slivers, and determining maturity was impossible. There was also some evidence for slow growth, with only the large earlywood pores visible in some charcoal. Roundwood fragments in birch were more common, but most were incomplete (without pith and/or bark). Insect tunnels, with oval shaped holes, were observed in birch in sample <38.17.011> and in oak from <38.17.020>. It is clear that oak was the dominant fuel used in the activities within the ring ditch and associated features, representing 75.5% of the dataset. Birch is also a significant component of the assemblage, at 22.6%, with notably little from other taxa. The dominance of a single taxon is a common feature of Bronze Age cremation- related assemblages and represents a deliberate selection of firewood for fuel, potentially a ritual choice (Thompson 1999, 253). The use of mature oak is well-suited to cremation as it provides the high heat necessary for efficient burning of bone. The quantity of birch at Site 38.17 is interesting, and potentially unusual as the majority of the sites along the pipeline have large quantities of oak and hazel, reflecting the dominant woodland of the region. Birch is a light- demanding tree, and an effective coloniser, so its presence may indicate that the exploitation of fuelwood resources from a more open landscape than indicated by, for example, burnt mound sites. The other possibility is that the use of birch relates to specific funerary practice. Although only two features, [38.17.26] and [38.17.58], contained positively identified human bone, there were several other contexts with indeterminate bone and it is probable that many of the assemblages related to funerary activities. Birch was present in quantity in the confirmed cremation assemblages, but also in many of the pits within the ring ditch. Birch wood is hard and suitable for making various artefacts, but its presence in the charcoal assemblage from site 38.17 seems too widespread to be related to pyre goods. Birch makes a good firewood, but it produces a short-lived heat (Warren 2006), which would not be suitable for cremation; at least not on its own, though it could have been used for kindling. According to the radiocarbon dating, there are three phase of activity relating to the charcoal samples, in which the confirmed cremation pits fall into the Early-Middle Bronze Age period (c.1900-1700BC). Of particular interest is the contrast between the Earliest Bronze Age (including pit 38.17.60 with the halberd) and the Middle Bronze Age samples. The latter exhibit a paucity of oak, with a marked increase in birch, hazel and the only identifications of broom/gorse.Exactly why this change occurred may relate to the specific nature of the fires from which the charcoal originated, the birch rich assemblages deriving from the group of pits 100m east of the barrow, but it suggests, nonetheless, a potential change in the immediate landscape of the site. Birch favours open areas, rather than dense woodland but, if birch were on the ascendancy, it suggests that the site was no longer being cleared of colonising trees. The final analysis on the whole of the charcoal assemblage from the pipeline sites, particularly those in the Brecon Beacons area, may elucidate the issue.
38 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Table 9: Charcoal from Site 38.17 cremation pits and ring ditch
Feature type cremation pit cremation pit ring ditch Feature no. 26 58 11 Context no. 27 27 47 47 83 89 98 99 59 62 63 70 73 102 Sample no. 7 12 10 11 27 28 32 33 16 17 19 20 23 37 Quercus sp. oak 17 (h) 28 (h) 16 (h) 21 (h) 19 (h) 29 (h) 21 12 30 (h) 30 (h) 30 (h) 30 (h) 30 (h) 29 (h) Betula sp. birch 13 (r) 2 14 (r) 29 (r) 11 9 8 1 Alnus/Corylus alder/hazel 1 h=heartwood; r=roundwood; (brackets denotes presence in some fragments only)
Table 10: Charcoal from Site 38.17 pits, postholes and hearth
Feature type posthole pit pit pit pit pit pit pit pit pit pit hearth Feature no. 52 60 80 23 56 64 100 108 3 5 7 28 Context no. 54 55 57 81 82 24 42 90 65 66 68 101 109 4 6 8 29 30 Sample no. 14 13 15 25 26 4 9 35 29 30 31 34 38 1 2 3 5 6 29 Quercus sp. oak 20 (h) 29 (h) 20 29 (h) 30 (h) 3 20 (h) 26 (h) 24 (h) 27 (h) 20 20 (h) 7 (h) 22 (h) 27 (h)
Betula sp. birch 1 1 1 27 6 (r) 3 (r) 10 18 22 (r) 27 (r) 1 1 cf. Betula sp. birch 2 (r) 1 Corylus avellana L. hazel 4 (r) 4 6 (r) Alnus/Corylus alder/hazel 1 Cytisus/Ulex broom/gorse 1r 2r Indeterminate bark 2 3 1 2 h=heartwood; r=roundwood; (brackets denotes presence in some fragments only)
39 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Taxonomic composition of charcoal by phase (based upon fragment count, N=922).
Taxonomic composition of charcoal assemblage from Site 38.17 (based upon fragment count, N=922).
Discussion Despite a relatively poor collection of archaeological finds and environmental remains the data shows some patterns. The possible palaeosol layer beneath 38.17.105 on the inside edge of the barrow ditch produced three fragments of charred hazel nutshell. As a potential soil horizon this presumably reflects activity that might predate the barrow and might be an indication of Neolithic or earlier activity on the site, possibly contemporary with the Mesolithic/early Neolithic flint flake from the barrow ditch.
There is very little charred plant material indeed in the samples from the initial phase of barrow construction, internal features and ditch sedimentation. Just five items, seeds of small legume, sun spurge(?), two unidentifiable seeds and a single fragment of charred hazel nutshell, is the sum of material from 362 litres of processed deposit from the barrow ditch and pits 60, 56, and 64. This is not untypical of a funerary monument, but clearly indicates that little or no food consumption or processing occurred at the site. The little burnt bone in
40 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
ring ditch context 38.17.062 might be residual from earlier activity on the site, or more reasonably be seen as debris from cremations at the site, although the bone was completely unidentifiable (see Table 3). The charcoal from two of these pits, 60 and 56, is exclusively oak, with only heartwood definitely identified, while that from pit 64 is mainly oak heartwood, but with some birch roundwood present (Table 10). Preferential selection of oak heartwood would be typical for fuelling cremation pyres, and these assemblages are what might be expected from pyre debris. The charcoal from the barrow ditch fills is even more selective for oak heartwood (Table 9), with just one fragment of birch against 179 fragments of oak. The absence of hazel from so many samples is unusual when compared with other prehistoric sites along the pipeline and suggests that these assemblages have been selective.
The next phase of activity is represented by pits 23 and 26, both near the inner edge of the barrow ditch, possibly indicating new burials into the periphery of the mound. Pit 26 which is interpreted as a cremation pit on the basis of the identified human cremated bone in its lower fill (Table 7) produced no identifiable charred plant remains, with both the upper and lower fills containing only charcoal. The charcoal assemblages from this feature, although dominated by oak heartwood, include a significant proportion (32%) of birch, including roundwood. The two samples (9 litres processed) from Pit 23 produced just one identifiable charred plant item, a charred seed of heath grass, Danthonia decumbens, a species often found on damp acid heaths and moors, and quite likely growing locally. The charcoal from the two samples differs dramatically despite the fact that the two contexts, 38.17.024 and 38.17.042, are considered to be the same deposit (Hart 2013), with 024 dominated by oak and 042 by birch (Table 10).
Broadly contemporary with pit 23 was the excavation of two pits in the fills of the barrow ditch, 80 and 58, and a hearth, 28. The samples from pit 80 produced no archaeological finds and just a single fragment of charred hazel nutshell and a charred seed of a small legume, and the charcoal assemblage is dominated by oak with several fragments of heartwood recognised. Pit 58 is clearly a cremation pit, all four samples producing cremated human bone, probably from a single adult individual. No identifiable charred plant remains were present, while the charcoal assemblages were composed of oak and birch, as with cremation pit 26, with birch comprising 25% of the total assemblage. A single fragment of alder/hazel is present, the only fragment of alder or hazel identified from the barrow site outside of hearth 28, contrasting with this hearth and the middle Bronze Age assemblages from the small group of pits 100m east of the barrow. On the basis of the two cremation pits it would appear that oak and birch were used to fuel the pyres, although the latter, which burns hot and fast, may have been used to get the oak burning strongly rather than as a main fuel source, or possibly for some ‘ritual’ reason. As Challinor has indicated above birch is often used for wood turning and the making of small objects, but the frequency of fragments in these two cremation pits implies a fuel rather than objects placed on the pyre. The hearth samples have proved the richest of the samples associated with the barrow. Not only have they yielded burnt bone, worked flint and a very rich charcoal assemblage, but the two samples have produced more identified charred plant remains than all the other samples from the barrow put together (Table 8). Furthermore all of the identified remains derive from gathered ‘foods’, hazelnuts, blackberries, raspberries, sloes, and other members of the Prunus family. It is conceivable that these could derive from plants cleared off the barrow or nearby in the autumn and burnt, but an absence of charcoal of all but hazel in the samples indicates that the food is a more probable interpretation. As such we might suggest that the hearth represents activity in the autumn, while not excluding activity at other times, assuming the fruits were utilised fresh. It also indicates food consumption at the site, but perhaps specifically gathered foods, a ‘meal’ perhaps of some special significance. Despite the processing of some six hundred litres of samples no charred cereal remains are recorded from the barrow and its associated features which perhaps lends support to the notion that the gathered foods were ‘special’. The evidence for
41 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
gnawing on the Prunus stones indicates rodent damage prior to burning indicating that some stones, already free of their flesh, were scavenged by rodents before they were burnt. Incomplete, or no, carbonisation after discard into the fire might make available the kernel to rodents, with the shell being finally carbonised when the hearth was next fired.
Birch charcoal from the post-pipe upper fill in pit 52 is dated to 1880-1690 cal BC (Griffiths 2014), broadly contemporary with the features cut into the ditch, but this need not date the pit and its original post. The fill of the post-pipe may date from when the post was removed or finally decomposed allowing material to accumulate within the post-pipe void. It is therefore possible that the material from samples 21 and 22, the fill and packing in the posthole, could predate significantly the material from samples 13 and 14 from the upper and lower fills of the post-pipe. In the event apart from a little burnt bone in the upper fill of the pipe, and a little vitrified clay from the pipe and the post packing, the pit produced only charcoal. The charcoal was studied from the two post-pipe fills and was dominated by oak, with only heartwood recognised, but with just one fragment of birch from the upper fill identified in the sample of 50 fragments studied (Table 10). Pit 100, the latest radiocarbon dated feature within the barrow, produced very little from the 32 litres processed – no finds, three fragments of charred hazel nutshell and a studied charcoal sample exclusively of oak. Pit 108, an undated pit within the barrow (not located on Fig. 1) produced no finds and no identifiable charred plant macrofossils, but a charcoal assemblage of oak, including heartwood, and birch, the latter comprising 33% of the sample.
The youngest features, radiocarbon dated to the middle Bronze Age, are two pits in a group of three (pits 3, 5 and 7) about 100m east of the barrow (see Fig. 1). These features are not only younger than those associated with the barrow, but they also differ in the environmental data collected from them. All three are relatively rich in charcoal, particularly the fill of pit 3, and apart from this pit the deposits lack oak charcoal (Table 10), being dominated in all three pits by birch, with a little hazel charcoal in pits 3 and 5, and broom/gorse charcoal in pits 5 and 7. This contrasts with all but one of the studied samples from the barrow site. The charred plant remains also contrast with those from the barrow. They are composed of weed taxa and herbaceous stem fragments (Table 8) including small legumes, thistles, daisy family, sedges, heath grass and moss, with pit 3 additionally producing five fragments of charred hazelnut. Two grammes of burnt bone were also recovered from pit 5, but could not be identified to species, and pit 7 produced pottery. When compared with the other samples from the site these three pits appear distinct but their assemblages do not allow any clear interpretation. The weed taxa could derive from local vegetation, possibly used as kindling along with the gorse/broom, with the birch and hazel fuelwood suggesting that these features are probably not associated with cremation, despite the presence of burnt bone and oak charcoal in pit 3 (see below).
The barrow is located at about 355m OD in front of a ridge of land leading to an elevated promontory, later followed by the Roman road, capped by a cairn and overlooking the watershed between the Usk and Gwydderig valleys. The location of the barrow at this point implies that the landscape around was probably cleared of woodland although there is no corroborative evidence from the excavations. Pollen analyses on the sequence from Test Pit 78 on Mynydd Myddfai (Rackham et al in prep) some 2.8km west of the barrow at a similar altitude shows the contemporary landscape to be dominated by oak and hazel, with significant local birch woodland. There is evidence for some opening up of the landscape with local pastures, some heath and arable cultivation indicated by low levels of grasses, ribwort plantain, heather, cinquefoils and occasional cereal type pollen. This picture is supported by the contemporary pollen data from Test Pit 229a a further 1.5km to the west, although the birch woodland is less prominent.
42 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
The landscape around the barrow and to the west, a relatively flat upland zone, may well have been one of these cleared areas. The use of predominantly oak and birch wood at the barrow site suggests some selection of fuel for the cremations, since hazel, although common in the local woodland, was not utilised. But the dominance of birch, with hazel, in the eastern middle Bronze Age pit group appears more likely to be a functional choice rather than a change in available woodland resources, no change is apparent in the pollen diagrams, suggesting the lack of oak implies these features had nothing to do with cremation pyres, although unfortunately we can offer no alternative contextual/functional interpretation for these pits.
Browne, D. and Hughes, S. 2003. The Archaeology of the Welsh Uplands, Royal Commission on the Ancient and Historical Monuments of Wales Brück, J. 1995. ‘A place for the dead: the role of human remains in Late Bronze Age Britain’, Proceedings of Prehistoric Society 61: 245-277 Griffiths, S. 2014 Radiocarbon results. South Wales Pipeline Project. Site 38.17 , Land North-east of Llwyn- Meurig (“The Halberd Site”), Llywell, Powys. Margary, I.D. 1973. Roman Roads in Britain 3rd edition (London) McKinley, J.I. 1994. ‘Bone fragment size in British cremation burials and its implications for pyre technology and ritual’, Journal of Archaeological Science 21: 339-342 McKinley, J.I. 1997. ‘Bronze Age “barrows” and funerary rites and rituals of cremation’, Proceedings of the Prehistoric Society 63: 129-145 Rackham, J., Challinor, D., Langdon, C. and Scaife, R. in prep Palaeoenvironmental studies along the Milford Haven to Aberdulais and Felindre to Tirley Natural Gas Pipeline. Southampton University Archaeology Monograph Series Thompson, G B, 1999. The analysis of wood charcoals from selected pits and funerary contexts, in A Barclay and C Halpin, Excavations at Barrow Hills, Radley, Oxfordshire, volume 1: the Neolithic and Bronze Age monument complex, Thames Valley Landscapes, 11, 247-253, Oxford, Oxford Archaeological Unit Warren, P., 2006. British Native Trees; Their Past and Present Uses, including a guide to burning wood in the home. Wildeye.
43 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
APPENDIX E: RADIOCARBON-DATING EVIDENCE BY SEREN GRIFFITHS
Introduction For the analysis, radiocarbon measurements were produced on short-life, single entity (Ashmore 1999), charred plant remains and cremated human bone from site 38.17 (Table 1). Three existing results are also detailed below. The dated samples were recovered from two phases of activity comprising a ringditch with associated pits and a small cremation cemetery. Two models are presented below, one of which makes use of the stratigraphic relationships between the parent contexts from which samples for radiocarbon dating were recovered, and one which applies a much more speculative interpretation of the development of the site (see discussion below). The results indicate that the diachronic activity occurred at the site over a considerable period of time, and that this could have represented relatively discrete pulses of activity.
Radiocarbon Dating Samples with the ‘Beta-‘ laboratory code were pretreated as detailed here http://www.radiocarbon.com/ . Samples with the ‘SUERC-‘ laboratory code were pretreated using an acid-base-acid process (cf. Mook and Waterbolk 1985). Samples are combusted as described by Vandeputte et al. (1996) and (Freeman et al. 2010). Following combustion, the samples are graphitized using methods described by (Slota et al. 1987), and dated by Accelerator Mass Spectrometry (AMS; Xu et al. 2004; Freeman et al. 2010). The results are conventional radiocarbon ages (Stuiver and Polach 1977), quoted according to the international standard set at the Trondheim Convention (Stuiver and Kra 1986). The results have been calibrated using IntCal13 (Reimer et al. 2013), and OxCal v4.2 (Bronk Ramsey 1995; 1998; 2001; 2009). The date ranges in Table 1 have been calculated using the maximum intercept method (Stuiver and Reimer 1986), and have the endpoints rounded outward to 10 years. The probability distributions shown in the figures were obtained by the probability method (Stuiver and Reimer 1993).
Bayesian Modelling Bayesian modelling provides a means of revising understandings of radiocarbon, or other scientific, dates using archaeological “prior information”, or understandings of the relationships between measurements. Different types of archaeological understandings of the relationships between scientific dating measurements can be more or less informative (Bayliss et al. 2007), but in any case it is vital to critically examine the nature of the model applied to the data. The model applied here represents the radiocarbon measurements as related to the phases of activity associated with the construction and use of the radiocarbon dates from the pits and postholes and the sequence of ringditch fill deposits from which samples were dated (Fig. 13). The output of the Bayesian model shown here (Fig. 13) is quoted in the text by convention in italics.
Results and Discussion Three results existed on samples from the site going into the analysis (samples with the ‘Beta-‘ laboratory code), these included a measurement on the mineralised wood within the halberd handle (Beta-240338) one from pit 38.17.100 (Beta-253578) and one from hearth 38.17.028 (Beta-253577). The results from 38.17.100 and 38.17.028, were much later than the result on the halberd handle, so one of the key questions of the analysis was whether the halberd was already old (perhaps an heirloom piece) when it was deposited. Samples were selected to provide radiocarbon ages for material from a range of features including the large central features within the ring ditch (pit 38.17.064 and post pipe 38.17.55 within pit 38.17.052). Further dateable material from the halberd pit itself was not identified. Equally, attempts to identify material from the ring-ditch fill itself were not successful. More material suitable for radiocarbon dating was identified from hearth 38.17.028 and pits 38.17.080 and
44 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
38.17.058, all of which post-dated the ring ditch fills. Samples from internal pits 38.17.023 and 38.17.026 were also submitted. A repeat sample on the mineralised material from the halberd handle failed due to insufficient carbon (GU33681). Duplicate measurements on samples from pits 38.17.005, 38.17.007, 38.17.026, 38.17.058, 38.17.064 and hearth 38.17.28 were produced (Table 1); only the duplicate measurements on material from pit 38.15.007 were statistically inconsistent (SUERC-52574; SUERC-52573; T’=5.35; T’5%=3.8; df=1; Ward and Wilson 1978). Both of these results have been included as active likelihoods in the models presented below.
A model reflecting the stratigraphic relationships is presented in Fig. 13; this model has good agreement (Amodel =91.5%). In this the features that post-date the ring-ditch are presented as stratigraphically later than the single result on material from the fill of the ring-ditch (SUERC-52593). As discussed below, it will be noted that the features from within the ring-ditch show considerable age variation. As no other samples suitable for dating could be identified from the feature containing the halberd (pit 38.17.060), and the second result on the halberd failed (GU33681), the question of whether the halberd was already old when it came to be deposited in the parent feature is to some degree open. Three radiocarbon samples on Betula sp. roundwood charcoal were produced on material from pit 38.17.064, and while it cannot be demonstrated that these are independent measurements (i.e. measurements on charcoal from different organisms), the consistency of these results (T’=0.1; T’5%=6.0; df=2; SUERC-52590-3) might suggest they are robust estimates for the burning activity which generated the charred assemblage deposited in the pit. These results are also consistent with the original result produced on the halberd handle (Beta-240338; T’=0.2; T’5%=7.8; df=3) which might suggest that the halberd was associated with the currency of this activity, and was not an heirloom piece.
Excluding the measurement on unidentified material (Beta-253577), the results on features which stratigraphically post-date the ring-ditch statistically consistent and could be of the same actual age (SUERC- 52594, -52583, -52580, -52581, -52582, -52588, -52598, -52589; T’=12.6; T’5%=14.1; df=7).
Pits 100m east of the ring ditch (38.17.05 and 38.17.07) contained pottery that could only be tentatively dated to the Early Neolithic and/or Early Bronze Age (Gibson 2013). While the results are not statistically consistent (SUERC-52573-4; 52578-9; T’15.8; T’5%=3.8; df=3), they indicate that the pottery, and associated activity is later than the tentative dates for the pottery.
From the model shown in Fig. 13 the estimate for the start of activity on the site is 2590–2320 cal. BC (95% probable), or 2500–2370 ca.l BC (68% probable; Start 38.17 ring-ditch features). The estimate for the last activity at the site is 1760–1510 cal BC (95% probable), or 1730–1610 cal BC (68% probable; End 38.17 ring-ditch features).
The posterior density estimate from the halberd haft suggests it was in use in 2450–2200 cal BC (95% probable) or 2400–2280 cal BC (58% probable) or 2250–2210 cal BC (10% probable; Beta-240338; Fig. 13). An estimate for the first dated event associated with the large features from within the ring-ditch places the start of activity in 2470–2320 cal BC (95% probable) or 2460–2370 cal BC (68% probable; FirstLargeFeatures; Fig. 13). The last dated event estimated from the results available suggests activity associated with the large features ended in 1780–1620 cal BC (94% probable; or 1750–1670 cal BC 68% probable; LastLargeFeatures; Fig. 13).
The estimate for the first infilling of the ring-ditch is estimated to have occurred in 2440–2200 cal BC (95% probable; or 2350–2210 cal BC 68% probable; FirstInfillingRingDitch; Fig. 13. The last activity from features
45 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
superimposed on the ring-ditch is estimated to have occurred in 1810–1680 cal BC (95% probable; or 1770– 1690 cal BC 68% probable; LastInfillingRingDitch; Fig. 13).
The first activity associated with the small internal features is estimated to have occurred in 1970–1810 cal BC (95% probable; or 1940–1880 cal BC 68% probable; FirstSmallFeatures; Fig. 14). The last activity estimated to have been associated with the small internal features occurred in 1870–1690 cal BC (95% probable; or 1820– 1740 cal BC 68% probable; LastSmallFeatures; Fig. 13).
The start of activity associated with external features 38.17.05 and 38.17.07 is estimated to have occurred in 1760–1310 cal BC (95% probable; or 1550–1390 cal BC 68% probable; Start external features; Fig. 14). The end of this activity is estimated to have occurred in 1400–930 cal BC (95% probable; or 1360–1180 cal BC 68% probable; End external features; Fig. 13).
The model presented above uses the stratigraphic relationships between dated parent contexts. The available results suggest that there might have been more episodic activity at the site, which because of complex taphonomy or processes on going at the site, is not fully substantiated by the archaeological evidence. A number of early dates from the site cluster into the last half of the third millennium cal BC, while subsequent results appear to focus on the first quarter of the second millennium cal BC, with the external features suggesting a focus on the last half of the second millennium cal BC. As discussed above, the results of the large features from within the ring ditch are not statistically consistent, and do not support the interpretation of these features as part of the earliest phase of activity on the site. The result on postpipe 38.17.055 (SUERC-52584) and from pit 38.17.100 (Beta-25378) are both later than the other results from these features. However, it could be suggested that neither of these results represent the earliest activity from these large features (feature 38.17.100 was disturbed after initial excavation, while postpipe 38.17.055 cannot be demonstrated to be associated with the earliest excavation of posthole 38.17.052). In both cases, the age of the radiocarbon result from each feature, and the nature of these features, might lead us to suggest that they do not represent the earliest activity on the site, but later activity associated with the small internal features and the features post-dating the ring ditch.
If we use this interpretation, and associate the result from large features 38.17.060, and 38.17.064, and the early result from the ditch fill (SUERC-52593), with the first phase of activity on the site (followed by all the other results from the internal features and post-dating the ditch fills), and then followed by the later external feature, it is possible to build a highly interpretive model of the site chronology. While there are limited archaeological grounds for presenting the data as described, it may be that this interpretation reflects the actual course of activity at the site more accurately. Output from this model is shown in Fig. 14 and Table 2. In this model, activity associated with the use of the large features within the ring ditch is estimated to have begun in 2530–2290 cal BC (95% probable; or 2470–2320 cal BC 68% probable; start early internal; Fig. 14), and ended in 2440–2170 cal BC (95% probable; or 2390–2260 cal BC 68% probable; end early internal; 68% probable; Fig. 14). The start of the subsequent activity associated with the small internal features, and the features post-dating the ring ditch is estimated to have begun in 1940–1760 cal BC (95% probable; or 1900–1800 66% probable; Start later internal and post-dating ring ditch features; Fig. 14). This activity is estimated to end in 1870–1660 cal BC (95% probable; or 1780–1690 cal BC 60% probable; End later internal and post-dating ring ditch features; Fig. 14). In this model, the external features are estimated to have started in 1670–1320 cal BC (95% probable; or 1550–1400 cal BC 68% probable; Start external features; Fig. 14), and ended in 1390–910 cal BC (95% probable; or 1360–1180 cal BC 68% probable; End external features; Fig. 14).
46 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Table 1 Radiocarbon results from site 38.17 Context Sampled Laboratory Measured δ 13C Calibrated age Posterior density material reference age (95%) estimate (95%) 38.17.006 Cytisus/Ulex SUERC-52573 3168 +/-29 -24.4 1510–1400 1500–1380 cal BC roundwood (85%) charcoal 38.17.006 Betula sp. SUERC-52574 3078 +/-26 -26.1 1420–1260 1420–1270 cal BC roundwood charcoal 38.17.008 Cytisus/Ulex SUERC-52578 3070 +/-25 -25. 1420–1260 1420–1270 cal BC roundwood c 6 rings 38.17.008 Betula sp. SUERC-52579 3007 +/-29 -25.7 1380–1120 1400–1190 cal BC charcoal
38.17.024 Betula sp. SUERC-52580 3485 +/-29 -25.2 1900–1690 cal BC 1890–1740 cal BC charcoal
38.17.029 ? Beta-253577 3340 +/-40 1740–1520 cal BC
38.17.029 Betula sp. SUERC-52581 3442 +/-29 -27 1880–1670 cal BC 1880–1680 cal BC charcoal
38.17.030 Betula sp. SUERC-52582 3477 +/-29 -26.1 1890–1690 cal BC 1890–1690 cal BC charcoal
38.17.038 Betula sp. SUERC52593 3844 +/-27 -27.1 2460–2200 cal BC 2440–2200 cal BC charcoal
38.17.047 Cremated SUERC-52594 3554 +/-25 -21.4 1960–1780 cal BC 1980–1870 cal BC human bone, (78%) long bone and 1850–1770 cal BC tibia (17%) 38.17.047 Betula sp. SUERC-52583 3494 +/-29 -26.2 1900–1700 cal BC 1900–1740 cal BC charcoal
38.17.055 Betula sp. SUERC-52584 3451 +/-29 -26.4 1880–1680 cal BC 1880–1690 cal BC charcoal
38.17.057 Halberd handle Beta-240338 3860 +/-40 2470–2200 cal BC 2450–2200 cal BC (mineralised Pomoidiae wood) 38.17.057 Mineralised GU33681 failed: and insufficient consolidated carbon wood; Pomoideae 38.17.065 Betula sp. SUERC-52590 3870 +/-24 -25.5 2470–2210 cal BC 2460–2280 cal BC charcoal (92%)
38.17.066 Betula sp. SUERC-52591 3881 +/-29 -26.7 2470–2210 cal BC 2470–2280 cal BC charcoal (93%) roundwood whole with bark 38.17.066 Betula sp. SUERC-52592 3879 +/-27 -25.7 2470–2230 cal BC 2460–2280 cal BC charcoal (94%) roundwood 38.17.081 Betula sp. SUERC-52588 3453 +/-25 -25.8 1880–1690 cal BC 1880–1690 cal BC charcoal
38.17.083 Cremated SUERC-52598 3476 +/- 24 -20.6 1890–1690 cal BC 1890–1740 cal BC human bone, (94%) tibia fragments 38.17.101 Corylus Beta-253578 3390 +/-40 1860–1610 cal BC 1780–1620 cal BC avellana (89%) nutshell
47 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Table 2 Key posterior density estimates from site 38.17 Parameter name Parameter interpretation Posterior density Figure estimate (95% showing probable) posterior Start 38.17 ring Estimate for the start of activity in the stratigraphic 2590–2320 cal BC Fig. 14 ditch features model of activity End 38.17 ring Estimate for the end of activity in the stratigraphic 1760–1510 cal BC Fig. 14 ditch features model of activity Beta-240338 Use of the halberd in the stratigraphic model of 2450–2200 cal BC Fig. 14 activity FirstLargeFeature The first dated event associated with the radiocarbon 2470–2320 cal BC Fig. 14 results from the large features in the stratigraphic model of activity LastLargeFeatures The last dated event associated with the radiocarbon 1780–1620 cal BC Fig. 14 results from the large features in the stratigraphic model of activity FirstSmallFeatures The first dated event associated with the radiocarbon 1970–1810 cal BC Fig. 14 results from the small features in the stratigraphic model of activity LastSmallFeatures The last dated event associated with the radiocarbon 1870–1690 cal BC Fig. 14 results from the small features in the stratigraphic model of activity Start external Estimate for the start of activity associated with the 1760–1310 cal BC Fig. 14 features two small external pits in the stratigraphic model of activity End external Estimate for the end of activity associated with the two 1400–930 cal BC Fig. 14 features small external pits in the stratigraphic model of activity start early internal Estimate for the start of activity associated with the 2530–2290 cal BC Fig. 15 first phase in the interpretive model of activity end early internal Estimate for the end of activity associated with the 2440–2170 cal BC Fig. 15 first phase in the interpretive model of activity Start later internal Estimate for the start of activity associated with the 1940–1760 cal BC Fig. 15 and post-dating later phase in the interpretive model of activity ring ditch features End later internal Estimate for the end of activity associated with the 1870–1660 cal BC Fig. 15 and post-dating later phase in the interpretive model of activity ring ditch features Start external Estimate for the start of activity associated with the 1670–1320 cal BC Fig. 15 features two small external pits in the interpretive model of activity End external Estimate for the end of activity associated with the two 1390–910 cal BC Fig. 15 features small external pits in the interpretive model of activity
48 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Fig 13 A Bayesian statistical chronological model for site 38.17. For each radiocarbon result two distributions are plotted. In outline are the calibrated radiocarbon results, in black are the posterior density estimates — the model output. The OxCal keywords and the brackets define the overall model exactly.
49 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Fig. 14 A much more interpretative Bayesian model for the possible chronology of site 38.17. The format is the same as Fig. 7, but this model uses the radiocarbon dates themselves, and an assessment of the potential evidence for reuse of two of the features to provide revised estimates for key events (outlined in Table 2).
References Ashmore, P. 1999 ‘Radiocarbon dating: avoiding errors by avoiding mixed samples’, in Antiquity 73, 124–30 Bayliss, A., C. Bronk Ramsey, J. van der Plicht and A. Whittle 2007 ‘Bradshaw and Bayes: towards a timetable for the Neolithic’, in Cambridge Archaeological Journal 17.1, supplement, 1–28 Bronk Ramsey, C. 1995 ‘Radiocarbon calibration and analysis of stratigraphy: the OxCal program’, in Radiocarbon 37(2), 425–30 Bronk Ramsey, C. 1998 ‘Probability and dating’, in Radiocarbon 40(1), 461–74 Bronk Ramsey, C. 2001 ‘Development of the radiocarbon calibration program OxCal.’, in Radiocarbon 43(2A), 355–63 Bronk Ramsey, C. 2009 ‘Bayesian analysis of radiocarbon dates’, Radiocarbon 51(1), 337–60 Freeman, S., G. Cook, A. Dougans, P. Naysmith, K. Wicken and S. Xu 2010 ‘Improved SSAMS performance’, in Nuclear Instruments and Methods Physics Research B 268, 715–17 Mook, W. G. and H. T. Waterbolk 1985 Handbook for archaeologists. No 3. Radiocarbon dating. Strasbourg; European Science Foundation Reimer, P., E. Bard, A. Bayliss, J. Beck, P. Blackwell, C. Bronk Ramsey, P. Grootes, T. Guilderson, H. Haflidason, I. Hajdas, C. Hatté, T. Heaton, D. Hoffmann, A. Hogg, K. Hughen, K. Kaiser, B. Kromer, S. Manning, M. Niu, R. Reimer, D. Richards, E. Scott, J. Southon, R. Staff, C. Turney and J. van der
50 © Cotswold Archaeology South Wales Gas Pipeline Project Site 38.17: Archaeological Excavation
Plicht 2013 ‘IntCal13 and Marine13 Radiocarbon Age Calibration Curves 0–50,000 Years cal BP’, in Radiocarbon 55(4), 1869–87 Slota Jr, P. J., A. J. T. Jull, T. W. Linick and L. J. Toolin 1987 ‘Preparation of small samples for radiocarbon accelerator targets by catalytic reduction of CO.’, in Radiocarbon 29, 303–6 Stuiver, M. and R. S. Kra 1986 ‘Editorial comment’, in Radiocarbon 28, 2B, ii Stuiver, M. and H. A. Polach 1977 ‘Discussion, reporting of 14C data’, in Radiocarbon 19(3), 355–63 Stuiver, M. and P. J. Reimer 1986 ‘A computer program for radiocarbon age calculation’, in Radiocarbon 28, 1022–30 Stuiver, M. and P. J. Reimer 1993 ‘Extended 14C data base and revised CALIB 3.0 14C age calibration program’, in Radiocarbon 35, 215–30 Vandeputte, K., L. Moens and R. Dams 1996 ‘Improved sealed-tube combustion of organic samples to CO2 for stable isotope analysis, radiocarbon dating and percent carbon determinations’ Analytical Letters 29, 2761–73 Ward, G.K. and, Wilson, S.R. 1978 ‘Procedures for Comparing and Combining Radiocarbon Age Determinations: a critique’, in Archaeometry 20, 19–31 Xu, S., R. Anderson, C. Bryant, G. T. Cook, A. Dougans, S. Freeman, P. Naysmith, C. Schnabel and E. M. Scott 2004 ‘Capabilities of the new SUERC 5MV AMS facility for 14C dating’, in Radiocarbon 46, 59–64
51 A NNyferyfer
AAC Cothi Site 38.17 PEMBROKESHIREPEMBRP MBROROKESESHIRSSHIREH RE CARMARTHENSHIRECACCARMARARRRMARMARA THENH NSHISH RE n Lly AfonA LlynfiLly Brecon w Tywi
A CyninCyni on frfron Ty AfronAAf Tywi ddau BRECONRECONECO BEACONSBBEAEAACONSACACONS NATIONALNNATNATIONATIONAATIONAO ALA PARKPARPARKA WessternWes Cleddau CywynCy A Cywyn Cyw FawrFawFaawr endraet Milford Gwendraeth Fach draeth GwendraethGwendr Fach Haven aweaw Taw el A T
T h / A Nedd R Tawel AAberdulais TaweTaw Rhy R Neath / A Nedd
Felindre
0 25km SSWSWANSSWANSEAWWAANANSEEAA
Pant Madog cairn
Site 38.17 The Mawr Site 38.14 cairn
Figs 2-6 Inset Site 38.20
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Milford Haven to Aberdulais pipeline Andover 01264 326549 Archaeology w www.cotswoldarchaeology.co.uk Felindre to Brecon pipeline e [email protected] PROJECT TITLE 0-75m contour South Wales Pipeline. Site 38.17, Land North 75m contour of Llwyn-Meurig, Trecastle, Powys 200m contour FIGURE TITLE Site location plan 400m contour 600m contour 0 1km FIGURE NO. Reproduced from the 2005 Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 3-09-2013 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright Cotswold Archaeology Ltd 100002109 APPROVED BY LG SCALE@A4 1:25,000 1 N
SN 85880 stone-capped pit 29170 38.17.080 1880-1690 cal. BC
pit 38.17.026 1900-1740 cal. BC hearth 1980-1770 cal. BC 38.17.028 G 1740-1520 cal. BC 1890-1680 cal. BC 1890-1690 cal. BC C B G E postpipe C B 38.17.055 1880-1690 cal. BC posthole pit H 38.17.052 A 38.17.064 ring ditch 2460-2280 cal. BC 38.17.011 2470-2280 cal. BC stone-capped cremation pit 38.17.058 H A E 1890-1740 cal. BC halberd N 2450-2200 cal. BC F pit 38.17.056 F D
292 see pit pit Fig 2 38.17.096 38.17.023 see 1890-1740 cal. BC Fig 12A pit 38.17.060 see D Fig 12B Site 38.17
I SN 859 01:5000 100m
pit pipeline centreline 38.17.100 1780-1620 cal. BC Phase A - Early Bronze Age I (2590-2200 cal. BC) Phase B - Early Bronze Age ring ditch (1970-1680 cal. BC) 38.17.011 2440-2200 cal. BC cremation unidentified burnt bone
Reproduced from the digital Ordnance Survey Explorer map with the permission of Ordnance Survey on behalf of The Controller of Her Majesty's Stationery Office c Crown copyright Cotswold Archaeology Ltd 100002109
Cirencester 01285 771022 Milton Keynes 01908 218320 Cotswold Andover 01264 326549 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE South Wales Pipeline. Site 38.17, Land North 0 2.5m of Llwyn-Meurig, Trecastle, Powys
FIGURE TITLE Plan of archaeological features
PROJECT NO. 9150 DATE 10-09-2013 FIGURE NO. DRAWN BY DJB REVISION 00 APPROVED BY LG SCALE@A3 1:50 2 3 4
3 Pre-excavation view of ring- ditch, looking south-west 4 Working shot showing ring-ditch, looking south-west
5 Post-excavation view of ring-ditch, looking south-west
5
N
Fig. 3-5
Cirencester 01285 771022 Milton Keynes 01908 218320 Cotswold Andover 01264 347630 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE South Wales Pipeline. Site 38.17, Land North of Llwyn-Meurig, Trecastle, Powys
FIGURE TITLE Photographs
PROJECT NO. 9150 DATE 16-04-2015 FIGURE NO. DRAWN BY DJB REVISION 00 APPROVED BY JB SCALE@A3 N/A 3-5 Section AA Section BB Section CC
SE NW NW SE SW NE
38.17.015 38.17.015 38.17.001
ring-ditch 38.17.104 ring-ditch 38.17.011 38.17.105 38.17.011 ring-ditch 38.17.011
Section DD Section EE Section FF
SE NW NW SE NE SW
38.17.090 38.17.057 pit 38.17.060
38.17.068 pit 38.17.056 38.17.083 pit 38.17.058
38.17.066 38.17.091 pit 38.17.064
38.17.065
N
Section GG 292 see NE SW Fig 2 see Section HH Fig 12A see NW SE Fig 12B 38.17.030 Site 38.17
38.17.029 hearth 38.17.051 38.17.028 postpipe 38.17.055 38.17.051 postpipe SN 859 01:5000 100m 38.17.054 posthole 38.17.052
38.17.053 pipeline centreline
Reproduced from the digital Ordnance Survey Explorer map with the permission of Ordnance Survey on behalf of The Controller of Her Majesty's Stationery Office c Crown copyright Cotswold Archaeology Ltd 100002109
Section II Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 SE NW Andover 01264 326549 Archaeology w www.cotswoldarchaeology.co.uk e [email protected] 01m PROJECT TITLE
38.17.101 pit South Wales Pipeline. Site 38.17, Land North 38.17.100 of Llwyn-Meurig, Trecastle, Powys
FIGURE TITLE Sections through ring-ditch and associated features
PROJECT NO. 9150 DATE 16-04-2015 FIGURE NO. DRAWN BY DJB REVISION 00 APPROVED BY JB SCALE@A3 1:20 6 7 8
7 Working shot of halberd being excavated 8 Detail shot of halberd in situ (scale 0.1m) 9 detail shot of halberd following conservation 10 detail shot of halberd following conservation
9* 10*
* Photographs for figures 9 & 10 taken by Anthony Daly of the National Museum of Wales
N
Fig. 7 & 8
Cirencester 01285 771022 Milton Keynes 01908 218320 Cotswold Andover 01264 347630 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE South Wales Pipeline. Site 38.17, Land North of Llwyn-Meurig, Trecastle, Powys
FIGURE TITLE Photographs
PROJECT NO. 9150 DATE 16-04-2015 FIGURE NO. DRAWN BY DJB REVISION 00 APPROVED BY JB SCALE@A3 N/A 7-10 Illustration undertaken by Anthony Daly of the National Museum of Wales
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
292 PROJECT TITLE pipeline centreline South Wales Pipeline. Site 38.17, Land North Site 38.17 of Llwyn-Meurig, Trecastle, Powys
FIGURE TITLE Drawing of halberd following conservation
FIGURE NO. Reproduced from the digital Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 16-04-2015 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright 01:5000 100m SN 859 Cotswold Archaeology Ltd 100002109 APPROVED BY JB SCALE@A4 1:2 11 A SN 85958
29145 K
N
38.17.008 J pit 38.17.007 1420-1270 cal. BC K 1400-1190 cal. BC
38.17.006
pit 38.17.005 01m J 1500-1380 cal. BC 1420-1270 cal. BC
Section KK
Section JJ SN
SN
38.17.006 38.17.008 pit pit 38.17.005 38.17.007
0 0.5m
B SN N 85953 29144
L Section LL
38.17.004 SW NE
38.17.004
pit L pit 38.17.003 38.17.003
01m0 0.5m
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Archaeology w www.cotswoldarchaeology.co.uk pipeline centreline e [email protected] 292 see Phase C - Middle Bronze PROJECT TITLE Fig 2 South Wales Pipeline. Site 38.17, Land North see Age (1500-1190 cal. BC) Fig 12A of Llwyn-Meurig, Trecastle, Powys
see FIGURE TITLE Fig 12B Plans and sections of easternmost Site 38.17 features
FIGURE NO. Reproduced from the digital Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 12-09-2013 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright 01:5000 100m SN 859 Cotswold Archaeology Ltd 100002109 APPROVED BY LG SCALE@A4 1:20 1:10 12