New dates from the north and a proposed chronology

New dates from the north and a proposed chronology for Irish court tombs RICK J. SCHULTING* School of Archaeology, University of Oxford

EILEEN MURPHY School of Geography, Archaeology and Palaeoecology, Queen’s University Belfast

CARLETON JONES Department of Archaeology, National University of Ireland, Galway

GRAEME WARREN School of Archaeology, University College Dublin [Accepted 15 September 2011. Published 12 January 2012.]

Abstract

We here present 33 new AMS measurements on human and animal bone from eight court tombs and two monuments with passage tomb affinities. While some determinations confirm earlier Neolithic use, others relate to Early Bronze Age burial activity. Unexpectedly, the animal bone dates all fall within the second millennium AD. The results are discussed in the context of the known corpus of court tomb dates, a critical assessment of which suggests that many early charcoal dates should be discounted. Bayesian modelling of 47 determinations from twelve sites places the initial use of court tombs*and by extension their construction*in the period 37003570 cal. BC, a time that is emerging as very dynamic, with strong peaks in activity in both the mortuary and settlement spheres across Ireland and Britain. Stable isotope data also obtained continue to support the pattern of little or no use of marine protein in coastal areas in the Irish and British Neolithic.

Introduction

Ireland is home to over 1,500 megalithic tombs, including some 230 passage tombs, 394 court tombs and 174 portal tombs, all of which are considered ´ Nualla´in 1989). to belong to the earlier Neolithic (Cody 2002; Jones 2007; O A fourth type, wedge tombs, are actually the most numerous, but fall within the

* Author’s e-mail: [email protected] doi: 10.3318/PRIAC.2011.112.06

Proceedings of the Royal Irish Academy Vol. 112C, 160 # 2011 Royal Irish Academy

R. J. Schulting et al. Beaker/Chalcolithic1 period and Early Bronze Age, and so are not dealt with in this paper. Traditionally, research on Irish megalithic tombs has focused on their origin and distribution, with each of the tomb types being regarded as a separate development. The degree of overlap evident in the chronology and distribution of the tombs has led Cooney and Grogan (1994, 578) to suggest they should all be considered in relation to each other and that the design of a tomb should be seen as a deliberate choice out of a range of possible options (see also Clarke 2006). At the same time, it is clear that the distribution of megalithic tombs shows genuine evidence for regionality with, for example, the highest density of court tombs ´ Nualla´ in 1989, Fig. 84; see also Mallory apparent in the north of the island (O and McNeill 1991, 56; Cooney and Grogan 1994, 58). The general chronology of Irish Neolithic mortuary monuments is well known, with passage tombs, portal tombs and court tombs all falling within the earlier Neolithic, c. 4000/38003200/3100 BC (Herity and Eogan 1977; O’Kelly 198l; Waddell 1998). Simple passage tombs may lie in the earlier part of this range, together with the first portal and court tombs, while elaborate passage tombs such as those of the Bru´ na Bo´ inne fall towards the end of the range (ApSimon 1986; Brindley et al. 2005; Sheridan 1986; 2003). There is of course continuing controversy over the dating of the simple passage tombs of Carrowmore (Bergh 1995; Burenhult 1984; 2001; 2005; Cooney et al. 2011, 6547; Sheridan 2003; 2010), but this debate lies beyond the scope of the present paper. Wedge tombs do not appear until c. 2500 BC at the earliest (Brindley and Lanting 1992a; O’Brien 1999; Schulting et al. 2008). The above is a general chronology, and the shortcomings in our knowledge are quickly revealed when a more detailed understanding is sought (Mallory and McNeill 1991, 55). Perhaps most crucially, and with very few but notable exceptions, there are far too few dates from any given monument, particularly in light of the often complex histories of construction, use, abandonment and re-use that characterise many sites; these very problems were noted four decades ago by Smith et al. (1971, 98), and yet remain as relevant today. In addition, many of the dates that are available are derived from charcoal from poor contexts bearing an uncertain relationship to the construction and use of the monuments; some determinations have large standard error terms that severely limit their usefulness; and, finally, geographical coverage is very uneven. Thus the statement that the three Neolithic monument types overlap chronologically at the level of Ireland as a whole does not address the possibility of more local sequences, let alone the histories of particular sites. These questions are crucial: history matters. Making use of large numbers of determinations for individual sites, in conjunction with a careful sampling strategy and the use of Bayesian modelling, recent work in southern Britain and Ireland has highlighted just how our perceptions of Neolithic practices can change, or at least be challenged, with improved 1

We follow O’Brien’s (in press) advocacy of the use of the term ‘Chalcolithic’ for the Beaker period, c. 25002000 cal. BC, in order to more clearly relate events in Ireland at this time to those on the Continent.

2

New dates from the north chronologies (Whittle et al. 2007a; 2011a). Many mortuary monuments and a number of causewayed enclosures thought to have been in use over centuries now appear to span only a few generations. Activity is being found to have been much more episodic than previously realised, and this has implications for the roles of monuments as well as for how society as a whole is viewed (Whittle et al. 2011b), and raises questions concerning burial practices other than in mortuary monuments (Dowd 2007; Schulting 2007). With a view to improving our understanding of the chronology of mortuary monuments in the north of Ireland a project was undertaken by two of us (RJS, EM) to date unburnt human and animal remains from eight court tombs: Aghanaglack, Co. Fermanagh (Davies 1939a); Annaghmare, Co. Armagh (Waterman 1965); Audleystown, Ballyalton, Ballyedmond and Mourne Park, Co. Down (Collins 1954; Davies 1938; Evans 1938a; Evans and Davies 1934); Legland, Co. Tyrone (Davies 1939b) and Tamnyrankin, Co. Derry (Herring 1941). In addition, samples were taken from the enigmatic megalithic monuments of Millin Bay (Collins and Waterman 1955) and Ballynahatty 1855 (after its year of discovery), Co. Down (MacAdam 1855). These two monuments share an affinity with the passage tomb tradition, but their relationship with court tombs provided a point of interest, as did, in the case of Millin Bay, its close proximity both to the Audleystown court tomb and to the sea (Figure 1). The original focus on unburnt bone was intended to provide concomitant information on palaeodiet, through the application of stable carbon and nitrogen isotope analysis (see Schulting 2011 for a recent overview for the European Neolithic). Due to the failure of many unburnt bone samples to yield sufficient collagen for dating, however, the project was subsequently modified to include unburnt human teeth and to take advantage of recent developments allowing the reliable dating of cremated bone (Lanting et al. 2001). Three new determinations from the Tully court tomb, Co. Fermanagh (Waterman 1978), obtained as part of another project (INSTAR2’s People of Prehistoric Ireland project, see: http://www.qub.ac.uk/sites/INSTAR peopleofprehistoricIreland/), are also presented, together with recent results from three other court tombs in western Ireland: Rathlackan and Behy, both in Co. Mayo, and Parknabinnia, Co. Clare (Byrne et al. 2009; C. Jones pers. comm. 2009; Warren et al. 2009). Because of the latter’s very narrow and straight-sided forecourt and its U-shaped cairn, it might be best described as an atypical court tomb (Jones 2003). These results, along with published determinations from other court tombs, provide the basis for a reconsideration of the beginnings and the use-span of this monument type in Ireland. The inclusion of Millin Bay and Ballynahatty 1855 allow us to present the first dates for these sites, with their passage-tomb affiliations, and also provide a point of comparison with the chronology of court tombs in the same area. 2

The Heritage Council’s Irish National Strategic Archaeological Research (INSTAR) Programme.

3

R. J. Schulting et al.

FIG. 1*Map of Ireland showing locations of key sites discussed in the text.

The dating programme A word on sample selection The majority of the recorded excavations (as opposed to earlier antiquarian explorations) of Irish mortuary monuments took place in the mid-twentieth century. In many cases this preceded the advent of radiocarbon dating, and certainly preceded the use of AMS for dating small samples. Thus it is not surprising that most of the available determinations have been derived from charcoal, with that material’s well-known attendant problems (Bowman 1990). For this project, dates on unburnt human bone were preferentially sought, first because these should refer to the use phase/s of the monument in a way that cannot be assumed for charcoal samples, and are (usually) free from the problems of a potentially significant in-built age; and second, to facilitate stable isotope analysis for dietary reconstruction. Unburnt animal bone was sought in 4

New dates from the north cases where there either was no human bone, or it was very limited; as will be apparent from the results this strategy proved less successful in understanding the early use of the monuments. A summary of the samples selected for dating from the ten sites (eight court tombs and two monuments with passage tomb affinity) included in the current project is provided in Table 1, with locational details and simplified site plans presented in Figures 1 and 2 respectively. Additional details concerning the excavated evidence and the selected samples from each site are presented in Appendix 1. Unlike the situation across much of southern Britain, where bone survives reasonably well, preservation conditions in the north of Ireland are generally very poor, due to the acidic soils that dominate the province. Exceptions are to be found along the coast of County Down, where calcareous sands provide more favourable conditions, and in the limestone regions of County Fermanagh (Wilson 1972). This immediately severely limited the number of potential samples available to the project. Exacerbating the problem, the majority of the (relatively) large human bone assemblages from Audleystown and Millin Bay, Co. Down, have been lost. Unburnt human bone samples were selected where possible to represent distinct individuals; in fact in some cases the number of samples attempted (not necessarily successfully*see below) constitutes the minimum number of individuals (MNI) for that site. But in the case of sites with very few human remains*literally two or three small fragments*samples were selected to reflect different contexts, with no guarantee that different individuals are represented (in the event, this was not an issue, since many of these samples failed to yield collagen). Similarly, the cremated bone samples in some cases could conceivably refer to the same individual. Conversely, one sample, from TABLE 1*Sites and samples selected for AMS dating (‘3/2’, etc. refers to number of successful/failed attempts).

Site

County

Tomb type

human

fauna

unburnt calcined unburnt total

Aghanaglack Annaghmare Audleystown Ballyalton

Fermanagh Armagh Down Down

court court court court

2/0 1/0 3/2 2/2

1/0  2/0 1/0

1/0  2/0 0/2

Ballyedmond Ballynahatty 1855 Legland Millin Bay

Down Down

indet. passage?

1/0 4/2

 3/2

 0/1

Tyrone Down

court cist/ circle court court

 5/2

 1/0

2/0 

0/1 

 

1/0 1/1

18/9

8/2

7/4

Mourne Park Down Tamnyrankin Derry Total

Site

Primary reference

4/0 1/0 7/2 3/4

Davies 1939a Waterman 1965 Collins 1954 Evans & Davies 1934 1/0 Evans 1938 7/5 MacAdam 1855 2/0 Davies 1939b 6/2 Collins & Waterman 1955 1/1 Davies 1938 1/1 Herring 1941

33/15

5

R. J. Schulting et al. Ballyalton, Co. Down, comprised three calcined adult bone fragments from at least two different elements, since no single fragment met the required sample weight. While there is nothing to indicate the presence of more than one individual in the cremated bone from the site, this possibility cannot be completely excluded. No sites were associated with large faunal assemblages (see McCormick 1986; 2007), and the dates reported here on animal bone do not claim to be in any way comprehensive. A total of 48 samples were taken, of which 33 yielded results (Table 1), four of which are repeat measurements on the same individuals (see below), leaving 29 unique results. The success rate of c. 69% (33/48) belies the fact that the initial failure rate was far higher (c. 43%, thirteen of 30 unburnt human and animal bone samples), and that alternative samples of either unburnt tooth

FIG 2*Simplified plans for sites yielding Libby Mulqueeny).

6

14

C dates discussed in the text. (Drawn by

New dates from the north dentine or calcined bone were subsequently sought from the same sites. With the exceptions of Millin Bay and Ballynahatty 1855, cremated bone did not form part of the original sampling strategy, since in the absence of collagen no stable isotope measurements can be made for dietary analysis. Cremated bone samples were consequently only sought from those sites for which some or all of the unburnt bone samples had failed. For the same reason, teeth were initially avoided where possible (excepting a single tooth sample from Ballyedmond, which was the only securely identified human element from the site), since their retention of a childhood isotopic signature can make comparison with adult diet problematic (e.g., Schurr 1998). However, after the failure of so many preferred samples, the selection criteria were modified in order to concentrate on the dating aspect of the project. In the event, the analyses of calcined bone

FIG 2*(continued)

7

R. J. Schulting et al.

FIG 2*(continued)

and tooth dentine proved highly successful, with only two failures out of ten attempts in the former case (both from Ballynahatty 1855), and no failures in eight attempts in the latter case. This last statistic is especially striking when compared to the figure of only eight successful results from sixteen attempts on unburnt human bone, in some cases from the same individuals that were subsequently successfully dated using dentine (see Bocherens et al. (1995), who report higher average collagen yields from dentine than from bone in Pleistocene fauna from Kent’s Cavern, Devon). Calibrations were undertaken in Calib 5.0 and OxCal v4.1.7, using IntCal09 (Reimer et al. 2009). All date ranges referred to in the text are 8

New dates from the north presented at 95% probability. Modelled date ranges are rounded to the nearest decade (since the results will vary slightly from run to run in any case), and are presented in italics, following accepted practice (Bayliss et al. 2007a). Stable carbon and nitrogen isotope results for the Belfast AMS samples were run at the Environmental Engineering Research Centre, Queen’s University Belfast, and those for the Oxford AMS samples were run at the Radiocarbon Laboratory for Archaeology and the History of Art, University of Oxford. All measurements were made in duplicate, with the average of the two results presented below.

Dating results Results were obtained on 33 samples (Table 2, Figures 3 and 4). Of these, two are repeat measurements on the same calcined bone sample from Ballynahatty 1855 (AX34.80), yielding almost identical results that can be readily combined to 4446924 BP (x2 test: df 1, T 0.1, 5% 3.8) (Ward and Wilson 1978). Conversely, two determinations on the same unburnt individual from Ballynahatty 1855 (AX34.2 and AX34.11) gave significantly different results (x2 test: df 1, T 65.9, 5% 3.8). Similarly, three determinations undertaken on the same unburnt human element from Millin Bay (99.8912bd, a right humerus) yielded significantly different dates that cannot be combined (x2 test: df 2, T 46.1, 5% 6.0). These problems are discussed further below. One immediately striking result is that the seven successful faunal samples all yield determinations falling in the second millennium AD, between c. 1300 and]1950 (Figure 4). The sites in question are Aghanaglack, Co. Fermanagh, Audleystown and Mourne Park, Co. Down, Legland, Co. Tyrone and Tamnyrankin, Co. Derry. In the cases of Legland, Mourne Park and Tamnyrankin, faunal samples were selected because no human bone was present. Discounting repeat and rejected measurements, and the intrusive historic period faunal remains, leaves a total of 22 determinations on unburnt and calcined human bone. While all of these relate to the prehistoric use of the monuments, it is apparent that many refer to a distinct period of secondary burial activity in the Chalcolithic/Early Bronze Age (Figure 3). Surprisingly, this applies to all five determinations*three on unburnt and two on calcined bone*from Audleystown, as well as to the three dates from Aghanaglack and to one of three dates from Ballyalton. The earliest dates obtained in the present project are on unburnt human teeth from Ballyalton and Ballyedmond. The results, centring on c. 36003500 cal. BC, are within the expected range for court tombs. A greater number of determinations fall within the second half of the fourth millennium BC, including the only result on human bone, an unburnt child mandible, from the Annaghmare court tomb. As was expected given their developed passage tomb affinities, and with two exceptions, the dates for Ballynahatty 1855 and Millin Bay fall within the second half of the fourth millennium BC. One of the Ballynahatty 1855 samples (AX34.2) is significantly later than the others (28822629 cal. BC, UB-6723: 4165936 BP); indeed, it falls within what is otherwise quite a marked gap 9

10

AX5.1 AX5.2 AX5.6 AX5.4 AX33.1 AX29.1 AX29.2 102.89-MBL 102.89-MBL-2 102.89-FVb

102.89-NEa 102.89-NEa-2 AX29.3 AX29.4 AX32.2 AX32.8 AX32.4 AX32.5 AX32.9 AX32.11 AX32.10 AX18.121 AX34.1 AX34.2 AX34.3 A.64 AX34.5 AX34.6 AX34.8 AX34.8 AX34.8

Aghanaglack Aghanaglack Aghanaglack Aghanaglack Annaghmare Audleystown Audleystown Audleystown Audleystown Audleystown

Audleystown Audleystown Audleystown Audleystown Ballyalton Ballyalton Ballyalton Ballyalton Ballyalton Ballyalton Ballyalton Ballyedmond Ballynahatty Ballynahatty Ballynahatty Ballynahatty Ballynahatty Ballynahatty Ballynahatty Ballynahatty Ballynahatty

H. sapiens H. sapiens H. sapiens? Bos taurus H. sapiens Bos taurus Bos taurus H. sapiens H. sapiens H. sapiens

Species

H. sapiens H. sapiens H. sapiens H. sapiens B47 H. sapiens H. sapiens B74 Bos taurus B73 Sus scrofa B76 H. sapiens B42 H. sapiens H. sapiens H. sapiens H. sapiens H. sapiens H. sapiens H. sapiens Bos taurus H. sapiens Grp. 1 H. sapiens Grp. 1 H. sapiens Grp. 1 H. sapiens

56A 34B 43 3

Id. 2

Id. 1

Site

L mandible tooth fibula cranium R mandible R mandible R humerus ulna mand. RM2 max. LM1 mixed mand. M1/2 mandible mandible mandible max. M1 pelvis mand. LM2 cranium cranium cranium

vertebra occipital longbone femur mandible L humerus L femur L mandible mand LM1 L mandible

Element

bone bone crem crem bone bone bone bone bone bone crem bone bone bone bone bone bone bone crem crem crem

bone bone crem bone bone bone bone bone bone bone

State

UB  6979 UB  7189 UB  7190 UB  6726 UB  6729 UB  6727 UB  6725 UB  7191 UB  7192 UB  7193 UB  6742 UB  6722 UB  6723 UB  6724 UB  7059 UB  6725 UB  7194 UB  7247x UB  7247a UB  7247b

4452 4440

4587

4465

4165

4796 4787 3692 4737

3713 3719 3774

3732 3549

UB  7593 UB  6978

C

3433 3446 3608 269 4556 602 189

14

UB  6730 UB  6731 UB  7188 UB  6732 UB  6741 UB  6735 UB  6736

Lab no. BC/AD

2277 2011

2029 1771

1880 1637 1884 1667 2128 1881 AD 1469-1953 3486 3103 AD 1296-1408 AD 1649-1955

cal Date

failed 28 2199 2030 33 2205 1985 36 2333 2041 failed failed failed failed 34 3650 3520 36 3650 3385 37 2199 1965 35 3636 3377 failed 36 2882 2629 failed 38 3343 3020 failed 34 3501 3116 failed at graphitisation 33 3337 2943 33 3331 2929

39 38 38 32 35 31 30 failed 35 35

9

TABLE 2*New AMS results on human and faunal remains from Northern Ireland.

d13C

21.0

19.0

23.0

20.0 22.0 26.0 20.0

24.0 21.0

21.5 23.4 24.0 20.7 21.3 20.9 23.2 24.4 21.6 25.1 21.1 -

2.9

3.3 3.1 3.5 2.8 3.0 3.0

10.4 3.5 12.8 5.3 5.2 3.3 11.5 2.9 10.0 2.9 9.5 3.1 12.0 3.5 13.2 5.2 10.5 3.4 9.7 3.7 11.0 2.8 -

10.6

11.1 14.6 5.4 11.3 6.1 5.1

d15N C:N

EERC/RLAHA

23.0 22.0 24.2 22.3 22.7 21.4 failed 19.0 21.3 25.0

23.6 22.6 28.0 24.5 22.0 21.8 20.6

d13C

AMS

Comment

cusps worn flat worn to enamel ring calcined longbone frags tooth root surface sheen removed surface sheen removed surface sheen removed ‘Ballynahatty 2’ chalky; also treated? fits w AX34.1? adult adult adult

proximal end, in fusion adult; main burial layer adult; main burial layer adult 27B; w Food Vessel child 23B; w FV? child 23B; w FV? longbone frag, fibula? orbit and cranial frags adult adult

child age 3-5 yrs.

unfused adult highly fragmented

R. J. Schulting et al.

AX34.10 AX34.9 AX34.11 AX30.1 AX30.2 101.89-1 98.89-1a 98.89-2 99.89-12a 99.89-12b 99.89-12c

99.89-12d

MBcrem-1 AX28.1 AX28.2 AX31.1

Ballynahatty Ballynahatty Ballynahatty Legland Legland Millin Bay Millin Bay Millin Bay Millin Bay Millin Bay Millin Bay

Millin Bay

Millin Bay Mourne Park Mourne Park Tamnyrankin Tamnyrankin no. 93 no. 63

no. 74 no. 50

Grp. 3 Grp. 2

Id. 2

H. sapiens Ovicaprid H. sapiens Bos taurus Ovicaprid

H. sapiens

H. sapiens H. sapiens H. sapiens Bos taurus Bos taurus H. sapiens H. sapiens H. sapiens H. sapiens H. sapiens H. sapiens

Species

longbone scapula cranium L femur tibia

R humerus

cranium longbone mand. RM1 tarsal/carpal L scapula mandible mandible mandible R humerus R humerus R humerus

Element

crem bone bone bone bone

bone

crem crem bone bone bone bone bone bone bone bone bone

State

OxA-16598 OxA-X2210-31 OxA-X2212-15 OxA-16467 UB  6739 UB  6740 UB  6737 UB  6738

UB  7248 UB  7520 UB  7521 UB  6733 UB  6734 OxA-16106 OxA-16107

Lab no.

663

4481 271

4740

4937 5048

4584 230 133 4597 4592

4507

C

cal Date

BC/AD

35 28 failed 33 failed

32

23.0 AD

1275-1394

21.1 20.6 22.0

3379

21.4 21.6

21.0 22.3 22.9 21.6 21.2

20.0

d C

3343 3029 AD 1517-1798

3636

36 3355 3095 failed 37 3501 3106 AD 1529-1955 31 AD 1667-1951 56 30 3500 3130 30 3500 3127 failed failed and resubmitted 32 3781 3651 33 3956 3766

9

13

AMS

21.9 27.1 23.8 23.5

21.1

21.5 23.3 22.6 21.6 21.2 21.4 21.6

d13C

3.4

3.4

7.0 2.9 6.1 6.1 3.1 4.6 3.8

9.7

10.4

11.2 2.9 7.5 3.2 6.8 3.1 10.9 3.3 11.5 3.3 -

d15N C:N

EERC/RLAHA

Calibrated with CALIB 5.0 (http://calib.qub.ac.uk/calib/calib.html) and OxCal v4.1, using IntCal09 (Reimer et al. 2009) EERC (UB dates) and RLAHA (OxA dates) isotope values are the averages of duplicate measurements. NB: Isotope values in italics are rejected for C:N ratios falling outside of range for in vivo collagen (DeNiro 1985). NB: d13C values obtained during the dating process (i.e., the ’AMS" column) are not suitable for palaeodietary interpretation.

Id. 1

14

AMS results on human and faunal remains from Northern Ireland.

Site

TABLE 2 (cont.)*New

adult E structure, Ch 1 E structure, Ch 1

adult

adolescent; repeat result

butchery marks-axe? adult, male? adult male; cut marks child; M1 cusps wearing adolescent adolescent; repeat result adolescent; repeat result

adult adult from mandible AX34.2

Comment

New dates from the north

11

R. J. Schulting et al.

FIG. 3*Calibrated dates (OxCal v4.1.7) on unburnt and calcined human remains from monuments in the north of Ireland (excludes repeat and rejected measurements from Ballynahatty 1855 and Millin Bay*see text). Note the gap between c. 3000 and 2300 cal. BC.

between c. 3000 and 2300 cal. BC (see Figure 3). The presence of a sheen to the mandible that provided the sample*incidentally the only one of the three sampled unburnt mandibles to yield a result*indicates the application of an unknown preservative at some point after the site’s discovery and excavation in 1855. While the surface was abraded before sampling, and the laboratory made aware of the likelihood of contamination, it is often difficult to be certain that all contaminants are removed. This obviously applies particularly strongly to contamination with more recent carbon, very small amounts of which can affect 14C measurements. The fact that the contaminant was in this case of unknown origin exacerbated the problem, but it is suspected to be collagenbased, which makes its removal by current pre-treatment methods practically impossible. In the light of this, and of the failure of the other two mandibles to 12

New dates from the north

FIG. 4*Calibrated dates on unburnt faunal remains from court tombs in the north of Ireland.

provide any result, three molar teeth were subsequently sampled, one of which was from the mandible yielding the late date. The tooth (AX34.11) provided a substantially earlier determination of 35013106 cal. BC (UB-7521: 4584937 BP), very much in line with results from the two other tooth samples, and with cremated bone from the site. The explanation of this is likely to be found in the facts that firstly, being in their sockets, teeth would be to some extent physically protected from the application of a preservative, and secondly, that dentine is a denser material than cortical bone, and so is less likely to absorb any preservative. UB-6723 on the mandible is therefore rejected and has been omitted from Figure 3. One of the Millin Bay samples (adolescent 99.8912) also requires discussion. The initial attempt to date a humerus from this individual marginally failed in the preparation stage, and so a second sample of the same bone was submitted. The result of 37813651 cal. BC (99.8912b, OxA16598: 4937 9 32 BP) was unexpectedly and significantly earlier than the other three dates from Millin Bay, though admittedly only two of these are on unburnt bone. One could say that the date is also unexpectedly early for the monument type, although such a statement needs to be qualified by noting the unique character of Millin Bay. Its supposed affinities with developed passage tombs rest on the presence of megalithic art (though of a different character than much of the classical passage tomb art: Shee Twohig 1981, 2334), and of a single sherd of Carrowkeel ware, which can no longer be thought of as specifically ‘passage tomb’ pottery (Sheridan 1995). Another possibility is that the bones of this particular individual were ancestral relics; the burial deposits in the central long cist were clearly secondary, being brought in as disarticulated, defleshed remains (Collins and Waterman 1955). A similar explanation was invoked to account for early dates on human bone from the ´ Donnabha´ in 1994). Poulnabrone portal tomb, Co. Clare (Lynch and O 13

R. J. Schulting et al. However, this is not thought likely in the present case. The more parsimonious explanation reflects the fact that the Millin Bay human bone assemblage was treated with a 20% dilution of the consolidant Bedacryl (Morton 1955, 60), the trademark name for a synthetic butyl-methacrylate resin (the laboratory was aware of the specific nature of the contamination). Being petroleum-based, contamination with this substance could be expected to make samples appear older than they really are (Long and Fritz 2001). A further sample from the same element was therefore submitted to check the first result, and yielded two determinations, one even earlier at 39563766 cal. BC (99.89.12c, OxA-X-2210 31: 5048933 BP) and the other significantly later at 36363379 cal. BC (99.89.12d, OxA-X-221215: 4740932 BP). Given the possibility of contamination with ‘dead’ carbon, the last determination would appear to be the most acceptable of the three. However, it too is best viewed with extreme caution at this point, and further analysis is planned to try to resolve the issue. Audleystown was excavated by one of the co-excavators at Millin Bay (Collins) and with the aid of the same bone specialist (Morton), raising the possibility that its assemblage was also treated with Bedacryl. However, this dual court tomb was excavated in 1952 (Collins 1954), whereas Millin Bay was excavated the following year. There is no specific mention of such treatment in the Audleystown report, nor is there evidence for any comparable discrepancy in the dates. Presumably, then, the decision to apply Bedacryl to the recovered osteological material was only made after the excavation of Audleystown. Two of the three unburnt mandible samples failed, leading to the substitution of teeth from two of the mandibles which, as noted above, are arguably less susceptible to contamination and can exhibit better collagen preservation. Both gave acceptable results. The single unburnt mandible that did yield a result actually returned the most recent of all the human bone dates from the site (20111771 cal. BC, UB-6978: 3549935 BP), and so is unlikely to be affected by Bedacryl contamination making it older than it really is. Furthermore, the sample in question (102.890FVb) appears to have been associated with Food Vessel sherds (taking the form of a highly decorated tripartite bowl: Collins 1954, 20, 25, Fig. 8), and so falls within the expected range of 21601920 cal. BC for this ceramic tradition as recently proposed by Brindley (2007).

Stable isotope methods and results A second aim of the present project was to further explore the possibility of regional variation in Neolithic diets through the use of stable carbon and nitrogen isotope analysis on human bone collagen. In the absence of C4 plants such as millet and maize, stable carbon isotopes (d13C) primarily reflect the relative contributions of terrestrial and marine sources of protein, with endpoints ranging from c. 2191˜ for a purely terrestrial diet to c. 1291˜ for a purely marine diet in north-west Atlantic Europe (Schulting and Richards 2002a; 2002b). Stable nitrogen isotope (d15N) measurements to a large extent reflect the trophic level of the protein component of the diet (Bocherens and Drucker 2003; Hedges and Reynard 2007). They are often significantly higher 14

New dates from the north in marine systems, because of the longer food chains that are possible; importantly, they are also typically higher in freshwater riverine and lacustrine systems for the same reason (e.g. Bonsall et al. 2004), and so can help identify a reliance on this resource that would not usually be apparent from the d13C values, which typically appear ‘terrestrial’. Measurements on adult bone collagen reflect an averaged protein intake over a period of approximately ten years. As discussed above, the failure of many of the initial samples necessitated a shift in sampling strategy to include cremated bone and unburnt dentine. Stable isotope values on cremated bone are at present not suitable for dietary reconstruction, and so are not considered further here. Measurements on dentine are suitable, but refer to the period during which the tooth root was forming. Thus, depending on the tooth, both d13C and d15N values may reflect a ‘nursing signal’, in which breastfeeding infants are raised one trophic level (approximately 1˜ for carbon and 3˜ for nitrogen) above their mothers (Schurr 1998; Richards et al. 2002). This obviously depends on the age of weaning. Permanent first molars are often preferentially sampled for dating because of their large size and good survival rate, and this practice has been followed here. Their roots form between approximately three and six years of age (Smith 1991; Ubelaker 1978); the early part of this range may just fall within what might be expected for weaning age in a farming society. This is confirmed by some studies that have documented elevated d15N values in permanent first molars compared to teeth forming in later childhood (e.g. Wright and Schwartz 1999), though of course this would be expected to be population-specific. Stable isotope results for human bone and tooth dentine and faunal bone are presented in Tables 3 and 4. The associated C:N ratios all fall within the acceptable range of 2.9 to 3.6 for in vivo collagen (DeNiro 1985). Both the Neolithic and Early Bronze Age averages, of d13C value of 21.590.6˜, and 21.990.7˜, respectively, are typical for terrestrial diets, and are broadly comparable to an average of 20.890.7˜ for mainly southern Neolithic Britain (Richards et al. 2003a). If anything, the results reported here are even more depleted than those in Britain*i.e. more ‘terrestrial’*but understanding the significance of this difference would require isotopic data for contemporary fauna that are simply not available at present. The modern faunal values obtained during this study*though they are indeed rather more depleted than British Neolithic fauna*are not relevant to the Neolithic situation, since changing vegetation patterns, soil fertility, anthropogenic inputs (e.g., the use of industrial fertilisers in the modern context) and climate all affect baseline plant values, which in turn affect herbivore consumers (Table 4). The very depleted d13C values for the inland sites of Annaghmare (Neolithic) and Aghanaglack (Early Bronze Age) are worth noting, though again in the absence of faunal values no satisfactory explanation can be offered here. The Neolithic average d15N of 10.790.7˜ is slightly high when compared to values from southern Britain, which tend to fall within the range 810˜ (Hedges et al. 2008; Richards 2000; Schulting and Richards 2002a; 15

16 inland inland coastal coastal

Ballynahatty AX34.6

Ballynahatty AX34.11

101.89-1 98.89-1a

99.89-12d

Millin Bay Millin Bay

Millin Bay 56A inland 34B inland coastal coastal

Aghanaglack AX5.1 Aghanaglack AX5.2 Audleystown 102.89-MBL-2

Audleystown 102.89-NEa-2

coastal

coastal inland

Ballyedmond AX18.121 Ballynahatty A.64

coastal

AX32.11

Ballyalton

B42

coastal

B76

AX32.9

Ballyalton

tooth

vertebra occipital mand LM1

R humerus

mandible mandible

mand. RM1

mand. LM2

mand. M1/2 max. M1

max. LM1

mand. RM2

mandible

Id. 2 Location Element inland

Id. 1

Annaghmare AX33.1

Site

UB-6979

OxA-X2212-15 UB-6730 UB-6731 UB-7593

OxA-16106 OxA-16107

UB-7521

UB-7194

UB-6742 UB-7059

UB-7192

UB-7191

UB-6741

Lab no.

C 9

3713 28

3433 39 3446 38 3732 35

4740 32

4597 30 4592 30

4584 37

4587 34

4737 35 4465 38

4787 36

4796 34

4556 35

14

TABLE 3*Stable C and N isotope values for unburnt human bone collagen. BC

2030

1637 1667 2029

3379

3130 3127

3106

3116

3377 3020

3385

3520

3103

(95%)

21.5

23.0 22.0 21.3

21.1

21.6 21.2

21.5

21.1

20.9 21.6

21.3

20.7

22.3

d13C

10.4

11.1 14.6 10.6

9.7

10.9 11.5

11.2

11.0

9.5 10.5

10.0

11.5

11.3

d15N

Neolithic average  21.3 10.7 SD  0.4 0.7 Early Bronze Age average  21.9 11.7 SD  0.7 2.0

2199

1880 1884 2277

3636

3500 3500

3501

3501

3636 3343

3650

3650

3486

cal

n 4

n 10

3.5

3.3 3.1 2.9

3.4

3.3 3.3

2.9

2.8

3.1 3.4

2.9

2.9

2.8

C:N

child age 3-5 yrs. cusps worn flat worn to enamel ring tooth root ‘Ballynahatty 2’ fits w AX34.1? from mandible AX34.2 adult, male? adult male; cut marks adolescent; repeat result unfused adult adult; main burial layer child 23B; w FV?

Comment

R. J. Schulting et al.

Ovicaprid Bos taurus

Mourne Park AX28.1 no. 93 Tamnyrankin AX31.1

scapula L femur

tarsal/carpal L scapula

femur L humerus L femur

Element

UB-6739 UB-6737

UB-6733 UB-6734

UB-6732 UB-6735 UB-6736

Lab no.

UB-6723 has been ommitted due to probable contamination as discussed in text Isotope values are the averages of duplicate measurements Failed and rejected samples are excluded (see Table 2)

Bos taurus Bos taurus

AX30.1 no. 74 AX30.2 no. 50

Species

Legland Legland

Id. 2 Bos taurus Bos taurus Bos taurus

Id. 1

Aghanaglack AX5.4 3 Audleystown AX29.1 Audleystown AX29.2

Site

TABLE 4*Stable C and N isotope values for unburnt animal bone collagen. C

271 663

230 133

269 602 189

14

28 33

31 56

32 31 30

9

(95%)

1517-1798 1275-1394

1529-1955 1667-1951

1469-1953 1296-1408 1649-1955

AD

21.9 7.0 23.8 6.1

23.3 7.5 22.6 6.8

3.5 3.0 3.0 proximal end, in fusion 3.2 3.1 butchery marksaxe? 2.9 3.1 E structure, Ch 1

d15N C:N Comment 24.2 5.4 22.7 6.1 21.4 5.1

d13C

Average 22.7 6.3 Standard deviation 1.0 0.9

AD

AD

AD

AD

AD

AD

AD

cal

New dates from the north

17

R. J. Schulting et al. 2002b), though values in northern Britain also tend to be more elevated (Schulting 2011; Schulting submitted; Schulting et al. 2010). The Early Bronze Age average is even more elevated at 11.792.0˜. In the absence of any positive correlation with d13C values this elevation cannot be easily explained by the consumption of even small amounts of marine protein: for the EBA, the two highest are actually from the inland site of Aghanaglack, with lower values from coastal Audleystown. The extremely high d15N value of 14.6˜ on an adult cranium from Aghanaglack is intriguing, but is so far removed from the other results that it warrants re-analysis for confirmation before an explanation is attempted. An initial suggestion might be that the measurements on teeth are responsible for the elevated d15N results, retaining a partial nursing signal, but in fact the opposite pattern is seen, with bone values being higher than tooth values*11.6 and 10.6˜, respectively*though not significantly so. Another possibility is that there is some contribution from freshwater fish, which can exhibit the observed combination of depleted d13C values and elevated d15N values (e.g., the EBA adult from Aghanaglack, mentioned above) (Dufour et al. 1999). This would be unexpected in the context of the Irish Neolithic, where there is little evidence for the use of such resources, although this has to be seen in the context of very limited faunal remains overall (McCormick 2007). Nor is it clear why freshwater fish*which would have been species-poor in prehistoric Ireland (Stuart and van Wijngaarden Bakker 1985)*would be utilised, but not marine fish, including salmon (since they spend most of their time feeding in the sea, salmon exhibit marine isotopic values). Another possibility is that the practice of manuring raised the d15N ‘baseline’ for both cereals and domestic animals (if fed grain and/or chaff), which in turn would elevate the values in human consumers (Bogaard et al. 2007). But, if so, farming practices would need to have differed significantly between Britain and Ireland to account for the observed difference in human values*certainly an intriguing scenario, but one requiring additional corroborating evidence. There is also the recent demonstration that coastal wetlands can yield high d15N values in plants, and hence in animals feeding on those plants (Britton et al. 2008). A final possibility that cannot be discounted is that the difference relates at least partly to interlaboratory variability; this is currently being investigated by one of the authors (RJS). Research undertaken as part of the INSTAR-funded Cultivating Societies project (see: http://www.chrono.qub.ac.uk/instar/) includes the dating and isotopic analysis of some Early Neolithic faunal remains, and may shed further light on the matter.

Discussion The early use of court tombs For some decades now, human and animal bones have been the preferred sample material for dating Neolithic mortuary monuments (e.g., Saville 1986). The problems with charcoal dates are well known, particularly when bulk 18

New dates from the north samples are used from the same or even different contexts (Ashmore 1999), and no species identification is undertaken. When carefully chosen, charcoal is of course perfectly acceptable as a dating material*much depends on the question being asked, and the availability or otherwise of alternative sample materials. The latter in particular is a factor in Irish Neolithic monuments (and settlements for that matter). The question facing us now is how to interpret those charcoal dates that are available in both the older and more recent literature. Of the 77 Neolithic determinations from court tombs in Ireland as a whole (Table 5) (excluding two dates with error terms over 200 years, two very recent charcoal dates from Annaghmare, and all the animal bone dates from the present project), 45 (58%) are on charcoal and 29 (38%) on unburnt or calcined bone, with the remaining three (4%) on charred hazelnut shell (Table 5). Taken at face value, the calibrated results suggest the beginnings of court tomb construction/use by c. 4000 BC (Figure 5),3 which would conform to traditional expectations concerning the appearance of the Neolithic in Ireland. An examination of the early dates in light of the above discussion, however, immediately raises difficulties with this scenario. The radiocarbon dating evidence may be explored using a series of simple Bayesian models. In the absence of secure stratigraphic sequences, and because we are dealing with multiple sites, the model makes only the basic assumption that all the available determinations from court tombs relate to a sample of all possible dates for a period of use with a beginning and end, and a uniform intervening distribution. The modelled ‘start’ and ‘end’ ranges are then estimated from the sample of dates themselves, and from the characteristics of the calibration curve over the relevant period (Bronk Ramsey 1998; 2000; Buck et al. 1994; 1996). From the distribution of the available calibrated dates the model calculates the likelihood of the existence of earlier samples that are not represented, a prospect that becomes increasingly improbable as sample size increases. To investigate the Neolithic use of court tombs, fifteen post-2500 cal. BC dates, referring to Beaker/Chalcolithic and Early Bronze Age re-use, are excluded and discussed further below. When the model is run with the remaining 62 determinations, one sample from Primrose Grange 1,4 Co. Sligo, of 43133811 cal. BC (Ua-16967: 5230975 BP) (Burenhult 2001), is rejected as an outlier (its ‘index of agreement’ of 33.9% falls well below the recommended 60%*for an explanation of this statistic see Bronk Ramsey (2001) and Bayliss 3

The modelled ‘summed probability distributions’ in this paper are used as simple visual aids to impart a general sense of the distribution of dates. They are not used for analytical purposes, nor should the details of the distribution be taken as indicative of the intensity of activity, as there are other contributing factors, most notably the error terms associated with the dates, and the shape of the calibration curve itself (cf. Bayliss et al. 2007a). 4 Although identified as a court tomb in the Megalithic Survey and by its excavator ´ Nualla´ in 1989), there is some question over whether Primrose (Burenhult 2001; 2003; O Grange 1, Co. Sligo, should be designated instead as unclassified (Clarke 2006). It is retained as a court tomb here, but with its early charcoal dates excluded, its inclusion or exclusion has little impact on the Bayesian model.

19

20 Fermanagh Fermanagh Fermanagh Tyrone Armagh Armagh Down Down Down Down Down Down Down Down Derry Derry Down Mayo Antrim Antrim Antrim

Ballybriest/Carnanbane

Ballyedmond Ballyglass 2 (Ma. 14)

Ballymacaldrack Ballymacaldrack Ballymacaldrack

County

Aghanaglack Aghanaglack Aghanaglack Altanagh Annaghmare Annaghmare Audleystown Audleystown Audleystown Audleystown Audleystown Ballyalton Ballyalton (47.11) Ballyalton (47.13) Ballybriest/Carnanbane

Site chamber? chamber chamber chamber 2 floor chamber beneath court blocking 102.89-MBL-2 AX29.3 102.89-NEa-2 102.89-FVb AX29.4 AX32.10 AX32.9 AX32.11 black layer beneath cairn black layer beneath cairn chamber dark layer directly under cairn mort. en. under cairn mort. en. under cairn forecourt blocking

Context

charcoal charcoal charcoal

human bone charcoal

charcoal

cremated bone human bone human bone charcoal human bone charcoal human tooth cremated bone human tooth human bone cremated bone cremated bone human tooth human tooth charcoal

Material

Lab No.

UB-2030 UB-2029 UB-2045

UB-6742 SI-1463

UB-534

UB-7188 UB-6731 UB-6730 GrN-11447 UB-6741 UB-241 UB-7593 UB-7189 UB-6979 UB-6978 UB-7190 UB-7193 UB-7191 UB-7192 UB-535

TABLE 5*Irish court tomb dates (excluding animal bone determinations).

5150 4940 4630

4737 4270

4930

3608 3446 3433 3810 4556 4395 3732 3719 3713 3549 3774 3692 4796 4787 5045

Date BP 2128 1884 1880 2573 3486 3330 2277 2205 2199 2011 2333 2199 3650 3650 4038

cal

90 4231 50 3912 130 3555

35 3636 90 3311

80 3947

38 38 39 120 35 55 35 33 28 35 36 37 34 36 95

9 BC

3714 3639 2943

3377 2578

3536

1881 1667 1637 1924 3103 2900 2029 1985 2029 1771 2041 1965 3520 3385 3650

(95%)

this paper ´ Nualla´ in ApSimon 1986; O 1998 Collins 1976; O’Kelly 1989 Collins 1976; O’Kelly 1989 Collins 1976; O’Kelly 1989

Evans 1939; ApSimon 1986

this paper this paper this paper Williams 1986 this paper O’Kelly 1989 this paper this paper this paper this paper this paper this paper this paper this paper Evans 1939; ApSimon 1986

Source

R. J. Schulting et al.

Armagh Armagh Armagh Mayo Mayo Mayo Mayo Mayo Mayo Mayo Tyrone Tyrone Antrim Clare Clare Clare

Behy

Behy

Behy

Behy

Behy Behy

Behy

Creggandevesky

Creggandevesky

Dunloy Parkabinnia

Parkabinnia

Parkabinnia

County

Ballymacdermot Ballymacdermot Ballymacdermot

Site

chamber 1

chamber 2

fac¸ ade chamber 2

AA43430

AA43428

AA43429

AA43416

UB-694 UB-698 UB-207

Lab No.

4685

4790

4805

5005

4830 4715 3660

Date BP

charcoal

human bone

human bone

charcoal human bone

GU-10577

GU-10580

UB-3533 GU-10578

UB-2539

UB-2540

4705

4725

4980 4785

4740

4825

charcoal, mixed UCD00118 4580

charcoal, mixed UCD00142 4680 charcoal, mixed UCD00141 4610

charcoal, SLS

charcoal, SLS

charcoal, SLS

charcoal, oak

charcoal charcoal charcoal

Material

context L11, chamber 3 charcoal

chamber 3 chamber 1 below forecourt blocking front chamber under paving court, same area as pottery court, same area as pottery court, among cobbles at E end spread S of gallery C5 spread in C1, under cobbles spread in C1, along N court context L25, chamber 1

Context

TABLE 5 (cont.)*Irish court tomb dates (excluding animal bone determinations).

cal

60 3635

60 3638

40 3938 60 3693

85 3695

80 3780

60 3525

70 3640 60 3619

55 3636

55 3660

55 3697

75 3965

95 3894 190 3940 60 2201

9 BC

3368

3372

3656 3376

3356

3376

3097

3345 3102

3355

3379

3381

3645

3370 2929 1889

(95%)

Bell and Foley 2005; Foley 2000 Bell and Foley 2005; Foley 2000 Conway and Williams 1994 this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm.

Warren et al. 2009

Warren et al. 2009 Warren et al. 2009

Warren et al. 2009

Warren et al. 2009

Warren et al. 2009

Warren et al. 2009

ApSimon 1986 ApSimon 1986 Smith et al. 1971

Source

New dates from the north

21

22 Clare Clare Clare Clare Clare Clare Clare Clare Sligo Sligo Sligo Sligo Sligo Sligo Mayo

1 1 1 1 1

Parkabinnia

Parkabinnia

Parkabinnia

Parkabinnia

Parkabinnia

Parkabinnia

Parkabinnia

Parkabinnia

Primrose Primrose Primrose Primrose Primrose

Primrose Grange 1 Rathlackan

Grange Grange Grange Grange Grange

Clare

County

Parkabinnia

Site

dark area in cist B F87, layer in Ch.3 (under F95)

cist A

dark area in cist A

chamber 1

chamber 1

chamber 1

chamber 1

chamber 2

chamber 1

chamber 1

chamber 1

chamber 1

Context

charcoal, SLS charcoal

charcoal, SLS charcoal, SLS charcoal, SLS charcoal, SLS human tooth

human bone

human bone

human bone

human bone

human bone

human bone

human bone

human bone

human bone

Material

Ua-12738 UBA-16466

Ua-16967* Ua-16968 Ua-11582 Ua-12739 Ua-16969

GU-10575

GU-10571

GU-10579

GU-10574

GU-10572

GU-10576

GU-10581

GU-10573

GU-10570

Lab No.

TABLE 5 (cont.)*Irish court tomb dates (excluding animal bone determinations).

4360 4685

5230 5145 5140 4645 4545

4195

4235

4315

4455

4455

4535

4550

4640

4645

Date BP cal

4313 4228 4223 3636 3518 80 3339 26 3625

75 75 65 70 80

55 2905

55 3006

55 3097

60 3346

60 3346

60 3496

60 3499

75 3635

55 3632

9 BC

2875 3371

3811 3715 3771 3117 2944

2621

2628

2764

2930

2930

3025

3030

3107

3168

(95%)

this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. this paper / C. Jones pers. comm. Burenhult 2001; 2003 Burenhult 2001; 2003 Burenhult 2001; 2003 Burenhult 2001; 2003 Burenhult 2001; 2003, pers. comm. 2011 Burenhult 2001; 2003 G. Byrne and G. Warren pers. comm.

Source

R. J. Schulting et al.

Mayo Mayo Mayo Mayo Mayo Mayo Mayo Mayo Mayo Mayo Mayo Mayo Mayo Mayo Tipperary

Rathlackan

Rathlackan Rathlackan

Rathlackan Rathlackan

Rathlackan

Rathlackan

Rathlackan

Rathlackan

Rathlackan

Rathlackan

Rathlackan

Rathlackan

Shanballyedmond

County

Rathlackan

Site

upper pitfill, rear Ch 3 F68, around hearth in court around hearth in court F21, spread on court surface F21, spread on court surface F107, under hearth in court top of socket SW end Ch3 F21, spread on court surface above basal stones in Ch3 F66, fill of pit in Chamber 3 w secondary pottery in C3 w secondary pottery in C3 post-hole at entrance

F78, stakehole in court

F95, layer in Ch.3

Context

charcoal

charcoal

charcoal

charcoal

charcoal

charcoal

charcoal

charcoal

charcoal

charcoal hazelnut shell

charcoal hazelnut shell

hazelnut shell

charcoal

Material

GrN-11431

Beta-76586

Beta-76584

UBA-16463

Beta-76585

Beta-76583

Beta-76590

UBA-16677

Beta-76587

Beta-76591 UBA-16462

Beta-76588 UBA-16464

UBA-16465

UBA-16467

Lab No.

TABLE 5 (cont.)*Irish court tomb dates (excluding animal bone determinations).

4930

3630

3640

3655

4090

4110

4130

4449

4520

4570 4559

4640 4600

4641

4674

Date BP cal

60 3938

80 2266

80 2276

28 2134

70 2874

90 2889

80 2892

26 3333

80 3498

90 3629 25 3483

80 3636 27 3498

25 3514

25 3617

9 BC

3542

1756

1771

1945

2488

2475

2491

3014

2930

3016 3110

3106 3141

3361

3370

(95%)

pers.

pers.

pers.

pers.

O’Kelly 1958, 1989; Grogan 1988

Byrne et al. 2009

G. Byrne and G. Warren pers. comm. Byrne et al. 2009

Byrne et al. 2009

Byrne et al. 2009

G. Byrne and G. Warren pers. comm. Byrne et al. 2009

G. Byrne and G. Warren comm. G. Byrne and G. Warren comm. Byrne et al. 2009 G. Byrne and G. Warren comm. Byrne et al. 2009 G. Byrne and G. Warren comm. Byrne et al. 2009

Source

New dates from the north

23

24 Fermanagh Fermanagh Fermanagh Fermanagh Fermanagh Fermanagh

Tully Tully

Tully

Tully Tully

Tully

cremated bone charcoal

cremated bone charcoal

charcoal

Material

chamber 3 chamber 2 court area chamber pitfill in Ch 3

cist below forecourt blocking chamber 2, mixed sample charcoal charcoal charcoal human bone charcoal

charcoal

cremated bone charcoal

chamber 1, under stone charcoal

Chamber 1 main forecourt surface under blocking Chamber 2 main chamber 1, under stone

base of cairn-spread

Context

UB-209 UB-240 D-48 UB6899 Beta-76589

UB-2116,18

UBA-13545 UB-2114

UB-2120

UBA-13547 UB-2119

UBA-13546 UB-2115

GrN-11432

Lab No.

1425 1600 4120 638 4390

4445

4698 4575

4785

4816 4890

5058 4960

3475

Date BP cal BC

2778

3374 3098

3366

3532 3523

3718 3545

1691

(95%)

50 AD 541-675 60 AD 264-596 300 3515 1890 30 AD 12841396 240 3655 2370

130 3520

24 3627 50 3506

85 3711

22 3650 65 3910

40 3961 85 3959

40 1893

9

Calibrated with CALIB 5.0 (http://calib.qub.ac.uk/calib/calib.html) and OxCal v4.1, using IntCal09 (Reimer et al. 2009) *NB: Ua-16967 is rejected as too early by the Bayesian model; ’SLS’  short-lived species

Annaghmare Annaghmare Ballyutoag Behy Rathlackan

Armagh Armagh Antrim Mayo Mayo

Fermanagh Fermanagh

Tully Tully

Rejected dates

Tipperary

County

Shanballyedmond

Site

TABLE 5 (cont.)*Irish court tomb dates (excluding animal bone determinations).

Smith et al. 1971 Smith et al. 1971 ApSimon 1986 Warren et al. 2009 Byrne et al. 2009

O’Kelly 1958, 1989; Grogan 1988 Murphy et al. 2011 Waterman 1978; ApSimon 1986 Murphy et al. 2011 Waterman 1978; ApSimon 1986 Waterman 1978; ApSimon 1986 Murphy et al. 2011 Waterman 1978; ApSimon 1986 Waterman 1978; ApSimon 1986

Source

R. J. Schulting et al.

New dates from the north

FIG. 5*Modelled summed probability distribution for the use-phase of court tombs (n 77) excluding four dates as detailed in text (Amodel 97.1; Aoverall 97.3). This is intended as a simple visual aid to impart a general sense of the distribution of dates, and does not take into account the small amount of added uncertainty estimated by the ‘start’ and ‘end’ boundaries. Note the resurgence of activity in the Early Bronze Age.

et al. (2007a): it is essentially the equivalent of a x2 test at the .05 level). Removing this sample provides a modelled ‘start’ range of 39803810 cal. BC (95% probability) for the onset of activity in court tombs (Figure 6), comparable to what was suggested initially based on a simple viewing of calibrated dates before modelling. But again there is a problem with this scenario. Rather strikingly, all but one of the fifteen earliest court tombs dates (including the rejected sample from Primrose Grange 1; see below for the exception) are derived from charcoal, from the sites of Dooey’s Cairn/ Ballymacaldrack, Co. Antrim (Evans 1938b; Collins 1976); Behy, Co. Mayo (Warren et al. 2009); Dunloy, Co. Antrim (Conway and Williams 1994); Ballymacdermot, Co. Armagh (Collins and Wilson 1964; ApSimon 1986); Carnanbane, Co. Derry (Evans 1939); Creggandevesky, Co. Tyrone (Foley 2000; Bell and Foley 2005); Primrose Grange 1, Co. Sligo (Burenhult 2001; 2005); Shanballyedmond, Co. Tipperary (O’Kelly 1958) and Tully, Co. Fermanagh (Waterman 1978) (Table 5).5 However, with five samples from three sites*Ballymacaldrack, Behy and Carnanbane*the charcoal is explicitly noted as deriving from pre-cairn contexts, in the case of Ballymacaldrack interpreted as the remains of a mortuary enclosure (Collins 1976), and thus they provide only termini post quos for the construction of these monuments. Ballymacaldrack, it might be noted, is one of the sites on which Sheridan (2006; 5

A single determination on oak charcoal (GrN-24991: 4990 9 110 BP) from Ballyglass (Ma 13), Co. Mayo, has not been included in the models as it is described as deriving from an ash spread ‘east of [the rectangular timber] house’ and so bears no clear relationship to the court tomb at the same site (Cooney et al. 2011, Table 12.5).

25

R. J. Schulting et al.

FIG. 6*Modelled ‘start’ date for the use-phase of court tombs (n 61) excluding Ua16967 for poor agreement (33.9%) and fifteen Early Bronze Age dates. The 95% probability range is 39803810 cal. BC (Amodel 100.8%; Aoverall 101.5%).

see also ApSimon 1997) bases her argument that timber-built mortuary structures precede the construction of megalithic monuments in Ireland. At another site, Dunloy, the excavators note that the charcoal providing the date was not from a securely sealed context, so that its relationship with the structure is uncertain (Conway and Williams 1994). Four of the six dates in a brief preliminary publication on Primrose Grange 1 are on charcoal and span the entirety of the fourth millennium BC (Burenhult 2001; see also Sheridan 2003, Fig. 4). In addition to the rejected determination noted above, two other samples recovered from Primrose Grange 1 (Ua-16968: 42283715 cal. BC; Ua-11582: 42233771 cal. BC) feature in the earliest fifteen dates, despite the fact that these are specifically identified as not being on oak charcoal (Burenhult 2001, 13). However, the two human samples from the chamber produced substantially later dates of 36363117 cal. BC (Ua-12739: 4645970 BP) and 35182944 cal. BC (Ua-16969: 4545980 BP)6, well within the range of human bone dates from a number of other court tombs. Thus, the likelihood that the earliest charcoal determinations are either statistical outliers, subject to the old wood effect, or residual (i.e., relating to pre-tomb activity) must be considered. The exception referred to above is a determination of 39613718 cal. BC (UBA-13546: 5058940 BP) on calcined human bone from Chamber 1 at Tully, Co. Fermanagh. This is over two hundred 14C years older than the next oldest date on cremated bone from this site (UBA-13545: 4826922 BP), or indeed 6

There has been some confusion in the literature concerning which dates from Primrose Grange 1 are on bone and which are on charcoal (Cooney et al. 2011, Table 12.5; Kytmannov 2008, Table 7.3). We have confirmed with the excavator and the Uppsala laboratory that both Ua-12739 and Ua-16969 are on human bone/tooth (G. Burenhult pers. comm. 2011; I. Sundstro¨ m pers. comm. 2011).

26

New dates from the north from any court tomb in Ireland (Table 5). The result can therefore be considered as an outlier, a possible explanation for which involves our current understanding of the source/s of carbon in calcined bone. The original presentation of the method of dating cremated bone proposed that the structural carbon in the bone’s mineral component was re-crystallised and stabilised upon reaching a certain temperature ( 600 8C), and hence became resistant to the diagenesis that otherwise generally precludes the use of this material for dating (Lanting et al. 2001). Recent experimental work, however, has raised the possibility that carbon deriving from the fuel used in the cremation pyre forms a significant part of the carbon that becomes locked into the calcined bone (Hu¨ ls et al. 2010). In most cases the difference is likely to be negligible, since the fuel used in cremations was probably often branchwood, and therefore of a similar age to the deceased. But if older wood with a substantial in-built age were used (e.g., structural oak timbers from an abandoned building), then this could affect the dating of the calcined bone. This might account for occasional dates on cremated bone that seem too old for their contexts (e.g., Schulting et al. forthcoming); alternatively, of course, they might be statistical outliers, or they may be perfectly valid, and refer to rare instances of earlier activity. Further research is required to better understand this issue, and its impact on the use of calcined bone for dating. In light of the potential problems with charcoal even when derived from reliable contexts, serious reservations must also be raised over the validity of the remaining charcoal dates from the other early sites. Of course, similar concerns could be raised over later charcoal dates, but these have far less impact on our understanding of the onset of court tomb activity (the same rationale applies to potential issues regarding the dating of calcined bone, raised above). Removing the fourteen early charcoal dates from further consideration at this point thus seems the most acceptable course of action. In the resulting Bayesian model, comprising 48 determinations from twelve sites, UBA-13546 on calcined bone from Tully, discussed above, can be formally rejected as an outlier (A 21.6%) (Figure 7). Excluding it leaves another result on calcined bone from Tully as the earliest directly dated human bone from a court tomb (UBA-13547: 36513531 cal. BC), followed closely*indeed, indistinguishably*by four results: two on human teeth from Ballyalton (UB-7191: 36503520 cal. BC; UB-7192: 3650 3385 cal. BC), another the earliest in a series of twelve determinations on unburnt human bone from Parknabinnia (GU-10578: 36933376 cal. BC), and finally a charcoal date from Tully (UB-2120: 37113366 cal BC). With the problematic exceptions of Primrose Grange 1 and Annaghmare, no sites have dates on both charcoal and bone so it is not possible to compare dates on the two materials at a site-specific level. However, there is no particular pattern in the remaining 47 determinations (i.e., excluding UBA-13546 and the fourteen earliest charcoal dates) as regards the charcoal and bone samples. Placing these into the same simple Bayesian model described above situates the onset of court tomb activity at 37003570 cal. BC (95% probability) (Figures 8 and 9). As no samples can be specifically related to a monument’s construction, this is better thought of as an estimation of their initial use-phase, though presumably it is 27

R. J. Schulting et al.

FIG. 7*Modelled dates for earlier Neolithic use of court tombs (n 47), excluding thirteen early charcoal dates and fifteen Early Bronze Age dates (Amodel 107.3%; Aoverall 106.8%). UBA-13546 is shown but is excluded from the model for its low agreement index (A 21.6%).

28

New dates from the north

FIG. 8*Modelled summed probability distribution for the Neolithic use-phase of court tombs (n  47) excluding UBA-13546, the fourteen earliest charcoal dates and fifteen Early Bronze Age dates (Amodel 107.3%; Aoverall 106.8%). This is intended as a simple visual aid to impart a general sense of the distribution of dates, and does not take into account the small amount of added uncertainty estimated by the ‘start’ and ‘end’ boundaries.

also a reasonable indicator of their construction date. It is worth noting that the twelve court tombs represented in the model include sites in the north, northeast and the west of the island (Figure 1), i.e., the main areas in which the monument type occurs. While this may be seen by some as surprisingly late, it does in fact conform to the picture emerging from southern Britain, where many CotswoldSevern chambered tombs, as well as some earthen long barrows, were built and

FIG. 9*Modelled ‘start’ date for the use-phase of court tombs (n  47) excluding the fourteen earliest charcoal dates and fifteen Early Bronze Age dates. The 95% probability range is 37003570 cal. BC (Amodel 107.3%; Aoverall 106.8%).

29

R. J. Schulting et al. used over a short period beginning in the late thirty-eighth/early thirty-seventh centuries cal. BC (Bayliss et al. 2007b; 2007c; Meadows et al. 2007; Whittle et al. 2007a; 2007b; Wysocki et al. 2007; see also Darvill 2004). General parallels have of course long been drawn between British Cotswold-Severn tombs and Irish court tombs and they, together with the ‘Clyde’ chambered tombs of south-west Scotland, do seem to present variations on a common theme, though the details are debated, especially as regards the perennial question of origins (Childe 1940; Corcoran 1969; 1973; Collins 1973; de Vale´ ra 1961; Evans 1938b; Piggott 1954; Scott 1962). This is not to deny the possibility of earlier construction and use for some court tombs: there are simply too few well-dated examples to make any such claim. Nor can the question of earlier timber-built mortuary structures be adequately addressed on the basis of the currently available radiocarbon dating evidence. Nevertheless, until additional information becomes available, the range of 37003570 cal. BC is suggested to be the most robust timeframe for the initial use of court tombs in Ireland. Their initial construction, then, might be suggested to lie in the slightly earlier range 3750 3700 cal. BC. A similar critical approach recently applied to the available 14C dates for Irish Early Neolithic timber-built rectangular houses restricts building and use of these structures to the tightly constrained period 37153625 cal. BC (McSparron 2008). This has since been confirmed with a much expanded dataset (Whitehouse et al. 2010). It is rather striking that the modelled ranges for the initiation of both court tombs and houses over Ireland as a whole are so similar. Taken together, this more critical and nuanced use of the available radiocarbon dating evidence highlights the fact that the earlier Neolithic is not a homogeneous period for which it can be assumed that the building and use of large rectangular houses and megalithic monuments*features often seen as defining the period*were necessarily familiar and recurrent practices (see Smyth 2006; 2010; Whittle et al. 2007a; 2011a). Rather, they may have been concentrated at particular times, and perhaps in particular places, though the evidence is less clear on this latter point. In fact, the available data suggest the existence of certain island-wide horizons of activity (e.g. court tomb- and house-building), such that temporal changes play a more important role*at least for those areas with houses and monuments*than regional variability. This in turn argues for a considerable degree of communication and interaction in the Irish Neolithic, which is of course hardly a new suggestion (Cooney 2000; 2004). Indeed, given the apparent synchronicity around 37603600 cal. BC in the onset of court tomb use in Ireland and the construction and initial use of a number of long barrows and chambered tombs in southern Britain, there may be even wider links and processes at work (Table 6). To this may now be added a flurry of intensive construction activity at causewayed and related enclosures in southern Britain, modelled as commencing around 37153670 cal. BC (Bayliss et al. 2011, 686), again showing a striking correspondence with the intensification of both mortuary and settlement practices in Ireland. What is novel is placing the onset of this activity within such a tightly constrained time-frame, one that is later by perhaps a century or two than is 30

New dates from the north TABLE 6*Modelled start dates for selected earlier Neolithic sites/phenomena.

Site/group Irish court tombs Irish rectangular houses initial Irish Neolithic Ascott-underWychood Hazleton North West Kennet Fussell’s Lodge Wayland’s Smithy southern Britain

Comment 47 dates (see text)

Modelled start range (95% probability)

Source

37003550 cal.

BC

this paper

37153625 cal. 37203660 cal.

BC

excludes Magheraboy, 37503680 cal. Donegore, domestic 38503740 cal. fauna Cotswold-Severn tomb 37603695 cal.

BC BC

McSparron 2008; Whitehouse et al. 2010 Cooney et al. 2011, models 2 and 3

BC

Bayliss et al. 2007b

Cotswold-Severn tomb 36953650 cal.

BC

Meadows et al. 2007

Cotswold-Severn tomb 36703545 cal. earthen long barrow 36453475 cal. earthen long barrow 36103525 cal.

BC

Bayliss et al. 2007c Wysocki et al. 2007 Whittle et al. 2007b

causewayed enclosures 37153670 cal

BC

BC BC BC

Bayliss et al. 2011

usually expected. Indeed, aside from the still-problematic Magheraboy causewayed enclosure, Co. Sligo (Danaher 2004; 2007), and a few individual early dates, such as that of 42333799 cal. BC (OxA-4269: 5190980 BP) on a domestic cow from Kilgreany, Co. Waterford (Molleson 1986), there is little 14C evidence for an Irish Neolithic earlier than c. 3800 cal. BC. The Ferriter’s Cove, Co. Kerry, cattle bones (Woodman et al. 1999) can be excluded from this discussion as they occur within a Late Mesolithic context and may represent, as Sheridan (2010) has suggested, a failed colonisation attempt from the Continent. Bayesian modelling of the appearance of the Neolithic in Ireland (again, Magheraboy aside) places it as either 38503740 cal. BC or 37503680 cal. BC (Cooney et al. 2011, 663). This is later by one to two centuries than the earliest Neolithic across much of Britain (Bayliss et al. 2011; Collard et al. 2009).

The later use of court tombs The dates from the Annaghmare court tomb represent a slightly later period of activity, in the second half of the fourth millennium. The single unburnt human bone date of 34853105 cal. BC (UB-6741: 4556935) from Chamber 2 can in fact be matched by a charcoal sample sealed by the primary blocking of the forecourt, which yielded a determination of 33302900 cal. BC (UB-241: 4395955 BP) (Smith et al. 1971). Both results would seem too late to relate to the monument’s construction, and it is possible that they refer to secondary use. Incidentally, additional charcoal samples from Chambers 2 and 3 yielded 31

R. J. Schulting et al. results of cal. AD 264596 (UB-240: 1600960 BP) and cal. AD 541675 (UB-209: 1425950 BP), respectively, though no evidence for disturbance was noted during the excavation (Smith et al. 1971). Falling within a similar time-frame as the probable secondary use of Annaghmare for burial are Ballynahatty 1855 and Millin Bay, both of which have been suggested to have greater affinity with the developed passage tomb tradition of the late fourth millennium BC. A Carrowkeel sherd was recovered from the pre-cairn surface at Millin Bay, but the main factor at this site is the presence of abundant passage tomb art, though not immediately comparable to the classic sites of Bru´ na Bo´ inne (Collins and Waterman 1955; Cooney 2000, 122; O’Kelly 1989, 143; O’Sullivan 1993; Shee Twohig 1981, 2334). Given its early discovery date, little is known concerning Ballynahatty 1855 (MacAdam 1855), but its context near the Ballynahatty henge and timber circle complex (Hartwell 1998; 2004), its circular chamber and division into cells suggest a passage tomb connection. The expected chronological attribution of both sites is here supported by their placement in the second half of the fourth millennium BC. A more detailed discussion of these sites is being prepared separately, but it is worth commenting here on their overlap with the continuing use of court tombs into the late fourth millennium BC, serving to emphasise the point that the use of court tombs did not necessarily cease when developed passage tombs began to be built. In a similar vein, it is worth commenting on the evidence for even later activity at Parknabinnia, Co. Clare, and at Ballyglass 2 (Ma 14; though see below) and Rathlackan, Co. Mayo. Unusually at these sites, activity extends into the first half of the third millennium (i.e. the Late Neolithic), a period which, in Britain, sees only occasional evidence for the continued use of earlier Neolithic mortuary monuments, whether for burial or other activity (see papers in Bayliss and Whittle 2007; Bradley 2007, 89). This is most striking in the case of Parknabinnia, since all of the twelve determinations here are on human bone, with two or three falling within the period in question, while charcoal provides the dates for the two County Mayo monuments, together with three new determinations on charred hazelnut shell at Rathlackan (ApSimon 1986; Byrne et al. 2009; G. Byrne, ´ Nualla´ in 1998; the long-suspect SI (Smithsonian pers. comm. 2011; O Institution) dates on charcoal from the both the Ballyglass house and tomb Ma 14 reported by ApSimon (1986) are too recent and are likely in error). What is particularly intriguing about Parknabinnia is its proximity to two other earlier Neolithic mortuary monuments, each of different type: the Poulnabrone portal tomb and the Poulawack Linkardstown-type cist and ´ Donnabha´ in 1994; Ryan mound (Hencken and Movius 1935; Lynch and O 1981). All show complex patterns of bone deposition, including successive interment, probable secondary burial and possible bone removal (Beckett 2005; Beckett and Robb 2006). Moreover, while the three monuments seem to follow different histories in their later lives, their use for burial clearly overlaps in the mid- to late fourth millennium (Brindley and Lanting 1992b; ´ Donnabha´ in 1994) (Figure 10). The apparently contemporary Lynch and O 32

New dates from the north use of what we would consider different monument types within an area spanning only a few kilometres is reminiscent of the above discussion concerning the overlapping use of court tombs and Millin Bay in Co. Down, as well as Cooney and Grogan’s (1994, 578) remarks on tomb types as representing a range of options available to people.

FIG. 10*AMS determinations on unburnt human bone from Parknabinnia, Poulnabrone and Poulawack, Co. Clare.

33

R. J. Schulting et al.

Chalcolithic and Early Bronze Age re-use The secondary re-use of court tombs for burial activities in the Chalcolithic/ Early Bronze Age is such a prominent feature of the present dating project that it warrants further discussion (see Table 2 and Figure 3). The first thing to note is that this is not a new phenomenon. Even before the advent of radiocarbon dating it was apparent from their finds that many court tombs, as well as passage and portal tombs, showed later re-use (Kilbride-Jones 1954; Herity ´ Rı´ordain 1957; Roche 2010). Radiocarbon dating is 1974; 1982; 1987; O increasingly confirming this re-use, in some cases where there is no artefactual evidence, and in other cases where it was expected, for example in the later insertions into the passage tomb of the Mound of the Hostages, Tara (Brindley et al. 2005). Portal tombs in both Ireland and Wales are also showing re-use for burial in the Bronze Age (Kytmannov 2008). Charcoal from beneath the stone blocking in the forecourt of Ballymacdermot provided a determination of 22001890 cal. BC (UB-207: 3660960 BP), in this case consistent with the Early Bronze Age pottery in the same layer (Smith et al. 1971). Similarly, late third-/ early second-millennium radiocarbon dates from the chambers of Rathlackan court tomb are in keeping with the Vase Urn and Cordoned Urn variant vessels found there (Byrne et al. 2009). Burials in pits surrounding the probable destroyed court tomb at Altanagh, Co. Tyrone, span the Early Neolithic to the Early Bronze Age, with appropriate associated pottery for these periods (Murphy et al. 2010; Williams 1986). Somewhat less expected were the three results from Aghanaglack obtained as part of the present project, two on unburnt human bone from two different individuals, and one on calcined bone. All three fall within the Early Bronze Age, with the calcined sample being significantly earlier. Case (1969, 21) had commented on the presence of a barbed and tanged arrowhead (Chalcolithic/EBA) and ‘Goodland bowl’ sherds (MN) in apparently primary contexts at Aghanaglack, indicating mixing, so perhaps the results are not so surprising. The single calcined sample from Ballyalton falls within the same period. The five determinations from Audleystown, three on unburnt human bone/teeth and two on calcined human bone, all fall within the Early Bronze Age. The unburnt material derives from Chamber 6, which yielded the greater part of the assemblage (Collins 1954, 20). The surviving documentation specifies that one element derives from the main burial layer (‘MBL’), another from the north-east corner of the chamber, and another apparently associated with Food Vessel bowl sherds. While the late date for the latter was thus not unexpected, the other samples were thought to be in primary Neolithic contexts, although others had expressed concerns: ‘There is some evidence that most of these [unburnt inhumations in court tombs such as Audleystown] were not the primary burials in the tombs and may be secondary interments, in some cases coming towards the later part of the Neolithic’ (Mallory and McNeill 1991, 59). While there is no surviving documentation to confirm this, the calcined bone almost certainly derives from Chamber 5, as this is the only chamber noted as having completely cremated bone, as opposed to the far more 34

New dates from the north ubiquitous, partially burnt material (Collins 1954, 19; Morton 1954), none of which survives in the small extant collection. Also noteworthy is the gap between the earlier Neolithic construction and use of megalithic mortuary monuments, and their re-use in the late third millennium and into the early second millennium BC. With the exceptions of Ballyglass 2, Rathlackan, and particularly Parknabinnia, there is very little evidence for activity at court tombs in the intervening five or so centuries of the Late Neolithic, c. 29002400 cal. BC. Rather, they appear to have been ‘rediscovered’ in the Chalcolithic/Early Bronze Age. This is of course not to suggest that they were somehow ‘lost’ to awareness in the Late Neolithic, but rather that they were not seen by people at that time as an appropriate place for the deposition of either material culture or human remains. Interestingly, the above three exceptions to this are all located on the west coast. This has to be seen in the light of the small number of radiocarbon determinations from any individual court tomb in other parts of the country. Nevertheless, it is possible that the west of Ireland saw the continuation of a tradition of use of these monuments that was not present across the island as a whole (while the available 14C determinations from Behy court tomb, Co. Mayo, provide no direct evidence for later use, there is a stratigraphically later but otherwise undated wall around the tomb that may be Neolithic, and if so may parallel the Rathlackan enclosure (Byrne et al. 2009)). The same seems to be true of passage tombs, with the well-dated Mound of the Hostages, Tara, showing a distinct gap in the distribution of dates on human bone between c. 3000 and 2400 cal. BC (Bayliss and O’Sullivan in press; Brindley et al. 2005). This gap can be paralleled in Britain (see papers in Bayliss and Whittle 2007), and may be part of the same general phenomenon. The main difference in Ireland is that the reuse of these Neolithic monuments*constructed in many cases a millennium or more earlier*is accompanied by the building of new communal megalithic monuments in the form of wedge tombs (Brindley and Lanting 1992a; O’Brien 1999; Schulting et al. 2008), the only British parallels for which are the entrance graves of Cornwall and the Scilly Isles (Hencken 1932; M. Davies 1945), and, as shown more recently, the Clava cairns of north-east Scotland (Bradley 2000). Parknabinnia, however, remains an important exception to this pattern as, though perhaps to a somewhat lesser extent, does Rathlackan. Ballyglass 2 also looks late, but, as noted above, its dating cannot be said to be secure. The case of Parknabinnia is particularly interesting as it is located in the middle of the densest concentration of wedge tombs in Ireland (de Vale´ ra and ´ Nualla´ in 1961), consisting of at least fourteen monuments within a 1.7km O radius, the closest of which is less than 270 metres distant from Parknabinnia (Jones 1998; Jones and Walsh 1996). Furthermore, the immediate surroundings of the Parknabinnia tomb are dominated by the remains of a Chalcolithic/ Early Bronze Age landscape of farms and field divisions (Jones 1998). With all this late third/early second millennium activity in close proximity to the Parknabinnia tomb, it is surprising that no evidence for re-use of the tomb in this period was uncovered, especially as the nearby Neolithic monuments of Poulnabrone and Poulawack show re-use in the second millennium (Brindley 35

R. J. Schulting et al. ´ Donnabha´ in 1994). It may be that these and Lanting 1992b; Lynch and O monuments were perceived differently from Parknabinnia, emphasising the importance of considering individual monument histories.

To burn or not to burn The ability to directly date both unburnt and cremated bone samples permits a far more detailed consideration of the chronology of these two mortuary treatments than has previously been possible. In the southern British Neolithic, collective burial of unburnt remains is the dominant funerary tradition in earlier Neolithic mortuary monuments, with cremation only appearing as a normative mortuary rite in the Bronze Age (Parker Pearson 2005; Bradley 2007). This can be contrasted with the situation across much of Ireland, and parts of western and northern Britain, where cremation appears to be more common than inhumation in the earlier Neolithic (Bradley 2007; Bryce 1902; 1904; Coles and Simpson 1965; Herity 1987; Vatcher 1961). But this observation needs to be qualified by taking into consideration preservation conditions, which over much of this area are not conducive to the survival of unburnt bone. Thus, even if inhumation was actually more common, the archaeological impression would be that cremation dominated. Indeed, at sites where bone preservation conditions are good, such as Parknabinnia, Audleystown and Millin Bay, there is relatively little cremated bone. Looked at another way, cremated bone is present at Ascott-under-Wychwood, Oxfordshire, and West Kennet, Wiltshire*albeit in what may be secondary contexts*and in what should be an early context at Fussell’s Lodge, Wiltshire (Whittle et al. 2007a). All three sites have yielded large assemblages of unburnt bone, but had the soils been acidic, the impression might well have been that cremation was the dominant rite (see also Smith and Brickley 2009). It is worth noting here the absence of any prehistoric bone at Behy, Ballyglass 1 and 2 and Rathlackan (with the possible exception of a single as yet unidentified bone fragment from Rathlackan*G. Byrne pers. comm. 2011). While the acidic conditions of the peat would have destroyed unburnt bone, calcined bone might be expected to survive, as indeed it did under similar conditions at Creggandevesky (Bell and Foley 2005). The implication of this is that inhumation was the dominant funerary treatment at these sites. In the present study, acceptable results were obtained on both unburnt and cremated human bone from three court tombs*Aghanaglack, Audleystown, Ballyalton*as well as Ballynahatty 1855 and Millin Bay (Figure 11 and Plate I). At Aghanaglack, the cremated bone is actually earlier than the two unburnt bone samples, but all fall within the Early Bronze Age. Similarly, the identical early dates on burnt and unburnt bone from Audleystown fall within the Chalcolithic/Early Bronze Age, with one later unburnt inhumation associated with a Food Vessel. Only at Ballyalton do inhumations belong to the initial use-phase of the monument, with cremated remains being a secondary Bronze Age addition. Thus, the question of the contemporaneity of the two funerary rites, while expected, remains open until additional results 36

New dates from the north

FIG. 11*Comparison of AMS determinations on unburnt and calcined human bone from the north of Ireland (see Table 2 for details).

are obtained, particularly on cremated bone from court tombs. Determinations on cremated bone apparently in primary contexts*though here again no unburnt bone survives*from the Creggandevesky court tomb, Co. Tyrone, are forthcoming and will contribute to the discussion (Foley in prep.). The relatively abundant cremated bone from Dooey’s Cairn (Ballymacaldrack court tomb, Co. Antrim) (Collins 1976; Evans 1938b) would also benefit from a comprehensive dating programme, particularly as it may be associated with pre-court tomb funerary use of the site (ApSimon 1997; Sheridan 2006). At Ballynahatty 1855 and Millin Bay the cremated and unburnt bone dates are statistically identical, but fall within a very flat part of the calibration curve between c. 3380 and 3050 BC (cf. Brindley 1999; Schulting et al. forthcoming), so that a more subtle chronological distinction cannot be ruled out, though this has not been found at either the Knowth passage tomb complex, or at the Mound of the Hostages, Tara, where both inhumation and 37

R. J. Schulting et al.

Pl. I*Unburnt and calcined human remains from Ballynahatty 1855 (Photo courtesy of Barrie Hartwell).

cremation are of comparable age (Bayliss and O’Sullivan in press; Brindley et al. 2005; Schulting et al. forthcoming). In the case of Ballynahatty 1855, there is no archaeological evidence to indicate any separation between the two practices, with both the cremated and unburnt human remains found together in the slab-defined compartments of a circular megalithic chamber (MacAdam 1855). At Millin Bay the remains of sixteen individuals (seven children, four adolescents and four adults) were recovered from within the central long cist (Figure 2). With the exception of the scattered cremated remains of an adult male all of the remaining individuals in the central cist were represented by unburnt bones. The unburnt bones largely comprised long bones and skulls, while the cremated remains were thought to represent a skull. Eight small cists were located to the exterior of the shingle bank surrounding the long cist*four of these contained deposits of cremated bone, with a further two uncisted cremations also being recovered (Collins and Waterman 1955; Morton 1955). While there is no clear stratigraphic relationship between the large central cist and the surrounding small cists, the excavators advanced the opinion that all constituent parts of the site may have been in use at the same time (Collins and Waterman 1955, 26). Cooney (2000, 1224) has more recently noted that the presence of megalithic art in all phases of activity supports this view. He also suggests that contemporary individuals in society may have been afforded different funerary practices*i.e. children and adolescents were deposited in an unburnt condition within a communal grave, whereas adults were cremated and largely buried in individual deposits. His alternative explanation is that over time the use of the communal ossuary was gradually replaced by the burial of the cremated remains of adults in cists. Unfortunately, no documentation survives regarding the provenance of the small amount of extant cremated bone from Millin Bay, beyond its attribution to the site. Thus it could derive from 38

New dates from the north either the central cist or one of the surrounding small cists, though we suspect the former is more likely.

The problem with animals The failure of any of the faunal remains to provide a prehistoric date, let alone a Neolithic date, was unexpected. Using strict stratigraphic criteria, some of the samples would not have been selected, but in these cases they provided the only available options for the site. Large elements were chosen to minimise the possibility of contamination by small bones filtering through loose deposits. Similarly, neonatal animals were avoided as the chambers of court tombs might be expected to serve as shelters for very young animals that subsequently died, as the lairs of predators, or as convenient locations for the disposal of their bodies by farmers. Two samples, one each from Legland and Tamnyrankin, exhibit cut marks. In the case of Legland these did appear as though they were made with a metal-edged implement; the date was in part sought to test this. McCormick (1986; 2007) has summarised the limited faunal material known from prehistoric Irish monuments, and drawn attention to the recurrent practice of placing only token deposits of the three main domestic species in Neolithic mortuary monuments. However, the faunal dating results reported here argue for considerable caution in attributing animal remains from these sites to the Neolithic, even in cases where they were thought by the excavator to be in a primary context (as at Aghanaglack). On the other hand, it should be noted that four of the eleven (c. 36%) faunal samples*two from Ballyalton, and one each from Ballynahatty and Tamnyrankin*failed to yield a date. This is slightly better than the 50% (nine of eighteen) failure rate for unburnt human bone samples. But more to the point, it raises the strong possibility that the failed faunal samples were in fact Neolithic, or at least Early Bronze Age, and that this is the very reason they failed to yield collagen (i.e., collagen, particularly in less than ideal conditions, will degrade over time). No animal tooth samples were available as replacements for the failed bone samples to test this hypothesis. Incidentally, the only human from a court tomb to fall in this late phase is the unexpectedly late result of cal. AD 12841396 (UB-6899, 638930 BP) on a cranium from the eastern chamber at Behy, Co. Mayo (Warren et al. 2009).

Stable isotopes and diet The stable isotope results, while limited in number, provide no evidence for the use of marine resources, contributing to the ongoing debate over the extent and speed of economic change at the Mesolithic-Neolithic transition in north-west Europe (Milner et al. 2004; Richards and Schulting 2006). Most importantly, there is no indication in the Neolithic human results that individuals living (or at least dying) near the coast (Audleystown, Ballyalton, Ballyedmond and Millin Bay) consumed any more marine protein than those living inland 39

R. J. Schulting et al. (Annaghmare, Ballynahatty). The slightly higher d13C average of 21.2˜ (n 6) for the coastal group does not differ statistically from the inland group at 21.6˜ (n 4),7 nor is it supported by a similar trend in d15N values (Table 3), which would be expected if fish and/or marine mammals were being consumed (Richards and Hedges 1999) (Figure 12). The even smaller number of Chalcolithic/Early Bronze Age results*two each from coastal and inland*are comparable. This is consistent with the pattern noted for Britain (Richards et al. 2003a), as well as with that previously noted for Ireland on the basis of (less accurate) isotope measurements associated with radiocarbon determinations (Woodman 2004). Eight of the twelve dated human samples from Parknabinnia have associated d13C values suitable for palaeodietary interpretation, and provide an average of 21.390.4˜ (Beckett 2005, table 5.36), fully in keeping with the results presented above (as the site is situated some 20km inland from the County Clare coast, it does not contribute to the debate over marine resource utilisation). This is not to say that marine foods were never used by coastal Neolithic communities*the technique is not sufficiently sensitive to detect the consumption of small amounts of marine protein, on the order of 510%, by an individual, though a larger sample size should pick up such low-level

FIG. 12*Plot of of stable carbon and nitogen isotope values for unburnt human bone from coastal and inland Neolithic sites included in the present project (EBA results from Audleystown and Aghanaglack are excluded). Late Mesolithic humans from coastal sites in Ireland (Ferriter’s Cove, Rockmarshall, and a dog from Dalkey Island) and Scotland (Cnoc Coig and Caisteal nan Gillean II, Oronsay) and an inland site in Ireland (Kiluragh) are shown for comparison. Rockmarshall is unusual for the coastal Mesolithic is presenting a purely terrestrial isotopic signal. Sources for Mesolithic values: Richards and Mellars 1998; Woodman 2009. 7

This difference is not significant at the .05 level (Student’s t-test with unequal variance: t  1.75, p  0.08), nor is it supported by corresponding d15N averages of 10.5 9 0.9˜ and 11.0 9 0.4˜ for coast and inland, respectively.

40

New dates from the north contributions at a population level. Unfortunately, there are simply not enough skeletal remains from the sites being considered here to investigate this properly (making the loss of the greater part of the large assemblages from Audleystown and Millin Bay particularly regrettable), but isotope values from earlier Neolithic near-coastal sites across Ireland as a whole suggest that any exploitation of marine resources was at best minor (Woodman 2004; Schulting 2005). In this regard the results reported here continue to support a strong shift away from the use of marine resources from the onset of the Neolithic as seen in Britain, and across much of north-west Europe (Tauber 1986; Fischer et al. 2007; Richards et al. 2003a; 2003b; Richards and Schulting 2006; 1998; 2005; 2011; Schulting and Richards 2002a; 2002b), though exceptions can be found in the eastern Baltic (Lide´ n et al. 2004) and the Netherlands (Smits et al. 2010). The clear distinction between the Neolithic humans analysed here, and coastal Mesolithic samples from Ireland (comprising two humans from Ferriter’s Cove, one from Rockmarshall, Co. Louth, and a dog from Dalkey Island, Co. Dublin) (Woodman 2009) and western Scotland (Richards and Mellars 1998) emerges clearly in Figure 11. Rockmarshall stands out as a coastal Mesolithic individual exhibiting an unusual ‘terrestrial’ signal. Similar values for two Late Mesolithic individuals are seen at Killuragh, Co. Limerick, but as this is some distance from the coast it is classed as an inland site and so the systematic exploitation of marine foods would not necessarily be expected (Schulting 2011). The situation in Neolithic Ireland cannot be said to be entirely resolved, however, as there is still the possibility of regional variation*the shell middens ¨ sterholm and of Donegal, Sligo and Mayo might be mentioned in this respect (O ¨ Osterholm 1984; Kimball 2000; Woodman 2001; Milner and Woodman 2007; Murray 2007; Warren 2009). Equally importantly, the first few centuries (c. 4000/39003700 cal. BC, though the early part of this range may perhaps itself now be questioned, as discussed above) following the initial appearance of Neolithic monuments (if the early dating of Magheraboy is accepted: Danaher 2004; 2007), material culture and domestic plant and animals, are poorly represented in the human isotopic data. On the other hand, if the larger British dataset is anything to go by, there is no compelling reason to expect that the situation in Ireland should be substantially different. If anything, the impoverished Irish faunal remains suggest that the alternatives to a mixed farming economy would be fewer and less attractive here, speeding the process of neolithisation (Woodman 2000).

Conclusions

It has not been possible to resolve one of the key problems raised at the beginning of this paper, that is, the small number of high-quality dates available from individual sites. In the present project, this was a limitation largely imposed by the paucity of available material and, to a lesser extent, by poor collagen preservation. Attempts to increase the available sample size by including animal bone were singularly unsuccessful, with all results post-dating 1200 AD. Nevertheless, the project has addressed another major concern regarding the extension of the geographical spread of dated court tombs, and has emphasised the importance of dating human bone/dentine rather than 41

R. J. Schulting et al. charcoal. For the purposes of modelling the early use of court tombs, we have proposed the rejection of a number of early determinations on charcoal, leaving the earliest acceptable results on burnt and unburnt human bone at c. 3650 3500/3400 cal. BC. We model the initial use of court tombs as lying in the range 37003570 cal. BC, later than previously thought for this monument class, but comparable to the ranges recently proposed for a number of mortuary monuments in southern Britain, including examples of the Cotswold-Severn type that show a degree of affinity with court tombs. While it remains possible that the earliest use of court tombs is not represented in the twelve sites included in the model, this is becoming increasingly improbable as more highquality dates on short-lived materials accumulate (though the early date on calcined bone from Tully comes to mind here). Their actual construction is likely to lie mainly in the earlier part of the proposed range for their use. One interesting observation to emerge from the dating programme is the demonstration that at least some court tombs remained in use towards the end of the fourth millennium BC, at a time when developed passage tombs began to be built. Following this, there is an apparent absence of activity lasting some centuries in the Late Neolithic, with the important exception of a number of sites in the west, suggesting the possibility of regional differences in how monuments were perceived and used. Significant episodes of later re-use of court tombs for burial in the Chalcolithic and Early Bronze Age are clearly seen in the results discussed here, as has also been documented at other sites across Ireland, most notably the Mound of the Hostages, Tara. The most striking case in the present study is Audleystown where much, if not all, of the unburnt bone assemblage in this classic dual court tomb may be the result of late third millennium activity. Aghanaglack and Ballyalton also show evidence for burial at this time or slightly later. While there are currently no other excavated court tombs in the north of Ireland with unburnt human bone apart from those featuring in the present paper, there are a number with cremated bone, in some case in substantial quantities (e.g. Creggandeveksy, Co. Tyrone, and Dooey’s Cairn, Co. Antrim). Similarly, cremated bone, and possibly small amounts of unburnt bone, is available from court tombs in other parts of Ireland. In the light of the success of calcined bone as a dating material, there will be further opportunities to improve the chronological picture presented here. The importance of recent developments in dating cremated bone cannot be over-emphasised, but its potential is only beginning to be realised in Ireland (Brindley et al. 2005; Kytmannov 2008; Schulting et al. 2008; Schulting et al. forthcoming). Given the often poor survival of unburnt bone, cremated bone will undoubtedly come to play a greater role in the future, though there remain some unresolved issues regarding the possibility of contamination with the fuel used for the pyre (Hu¨ ls et al. 2010). Unfortunately, calcined bone currently also lacks the ability to provide palaeodietary information through stable isotope analysis, as well as providing relatively little in the way of basic osteological data. The small number of unburnt individuals analysed here for stable carbon and nitrogen 42

New dates from the north isotopes continue to confirm the absence of any clear contribution from marine resources in the earlier Neolithic, even from sites near the coast. The apparent synchroneity in the use of court tombs and the currency of timber-built houses in the thirty-eighth century BC marks this as a dynamic period in Irish, and indeed British prehistory (Bayliss et al. 2011), following on not long after the initial introduction and adoption of a farming way of life on the western periphery of Europe. Unlike the houses, however, court tombs remained visible and important places on the Irish landscape, witnessing periodic episodes of renewed deposition into the Chalcolithic and the Early Bronze Age, although the gaps in this record may be as informative as the presences. While the construction dates for court tombs and developed passage tombs may indeed conform to the traditional expectation of Early and Middle Neolithic attributions, respectively, it is clear that the histories of both individual monuments and overall monument classes was more complex, with overlapping use within restricted geographical regions. It falls to a muchimproved chronology, combined with material culture studies (barely touched upon here), to tease out these relationships. We hope to have made a start in this direction.

Acknowledgements The authors would like to thank the Ulster Museum and the Department of Anatomy, Queen’s University Belfast, for permitting sampling of material in their care, and Sine´ ad McCartan of the Ulster Museum for assisting with queries and facilitating access to the collections. Thanks also go to Claire Foley and Janet Bell (Creggandevesky) and to Gretta Byrne (Rathlackan) for permission to quote unpublished radiocarbon dates from their respective sites, to Go¨ ran Burenhult and Ingela Sundstro¨ m for a point of clarification concerning Primrose Grange, and to an anonymous reviewer for his/her constructive comments. Funding for the AMS determinations was supplied through the 14CHRONO Laboratory of Queen’s University Belfast and NERC’s ORADS programme via the Radiocarbon Laboratory for Archaeology and the History of Art, University of Oxford. We are very grateful to Paula Reimer and Stephen Hoper of the 14CHRONO Laboratory and to Tom Higham at the Oxford facility for their help with the dating process. Stable isotope measurements for UB samples were run separately at the Environmental Engineering Research Unit, Queen’s University Belfast.

References

ApSimon, A. 1986 Chronological contexts for Irish megalithic tombs. Journal of Irish Archaeology 3, 515. ApSimon, A. 1997 Wood into stone: origins for the Irish megalithic tombs. In A.A. Rodrı´guez Casal (ed.), O Neolı´tico Atla´ntico e as Orixes Megalitism, 12940. Santiago de Compostela: University de Santiago de Compostela. Ashmore, P.J. 1999 Radiocarbon dating: avoiding errors by avoiding mixed samples. Antiquity 73, 12430. Bayliss, A. and Whittle, A. 2007 Histories of the dead: building chronologies for five southern British long barrows. Cambridge Archaeological Journal 17. 43

R. J. Schulting et al. Bayliss, A., Bronk Ramsey, C., van der Plicht, J. and Whittle, A. 2007a Bradshaw and Bayes: towards a timetable for the Neolithic. Cambridge Archaeological Journal 17, 128. Bayliss, A., Benson, D., Galer, D., Humphrey, L., McFayden, L. and Whittle, A. 2007b One thing after another: the date of the Ascott-underWychwood long barrow. Cambridge Archaeological Journal 17, 2944. Bayliss, A., Healy, F., Whittle, A. and Cooney, G. 2011 Neolithic narratives: British and Irish enclosures in their timescapes. In A. Whittle, F. Healy and A. Bayliss (eds), Gathering time: dating the early Neolithic enclosures of southern Britain and Ireland, 682847. Oxford. Oxbow. Bayliss, A., Whittle, A. and Wysocki, M. 2007c Talking about my generation: the date of the West Kennet long barrow. Cambridge Archaeological Journal 17, 85101. Bayliss, A. and O’Sullivan, M. forthcoming Interpreting chronologies for the Mound of the Hostages, Tara and its contemporary contexts in Neolithic and Bronge Age Ireland. In M. O’Sullivan (ed.), The Mound of the Hostages: Tara symposium. Dublin. Beckett, J. 2005 Death and burial on the Burren: a taphonomic study of three Megalithic monuments in County Clare, Ireland. Unpublished PhD thesis, Department of Archaeology, University of Cambridge. Beckett, J., and Robb, J. 2006 Neolithic burial taphonomy, ritual, and interpretation in Britain and Ireland: a review. In R. Gowland and C. Knu¨ sel (eds), The social archaeology of funerary remains, 5780. Oxford. Oxbow. Bell, J. and Foley, C. 2005 Excavations at Creggandevesky, Co. Tyrone (CAF Data Structure Report 45). Unpublished report prepared for the Environment and Heritage Service, DOE NI, by the Centre for Archaeological Fieldwork, Queen’s University Belfast. Bergh, S. 1995 Landscape of the monuments. Stockholm. Arkeologiska Underso ¨ kningar Skrifter 6. Bocherens, H. and Drucker, D. 2003 Trophic level isotopic enrichments for carbon and nitrogen in collagen: case studies from recent and ancient terrestrial ecosystems. International Journal of Osteoarchaeology 13, 4653. Bocherens, H., Fogel, M.L., Tuross, N. and Zeder, M. 1995 Trophic structure and climatic information from isotopic signatures in Pleistocene cave fauna of southern England. Journal of Archaeological Science 22, 32740. Bogaard, A., Heaton, T.H.E., Poulton, P. and Merbach, I. 2007 The impact of manuring on nitrogen isotope ratios in cereals: archaeological implications for reconstruction of diet and crop management practices. Journal of Archaeological Science 34, 33543. Bonsall, C., Cook, G.T., Hedges, R.E.M., Higham, T.F.G., Pickard, C. and Radovanovı´c, I. 2004 Radiocarbon and stable isotope evidence of dietary change from the Mesolithic to the Middles Ages in the Iron Gates: new results from Lepenski Vir. Radiocarbon 46, 293300. Bowman, S. 1990 Radiocarbon Dating. Berkeley. University of California Press.

44

New dates from the north Bradley, R. 2000 The good stones: a new investigation of the Clava Cairns. Edinburgh. Society of Antiquaries of Scotland. Bradley, R. 2007 The prehistory of Britain and Ireland. Cambridge. Cambridge University Press. Brindley, A.L. 1999 Sequence and dating in the grooved ware tradition. In R. Cleal and A. MacSween (eds), Grooved ware in Britain and Ireland, 13344. Oxford: Neolithic Studies Group Seminar Papers 3. Brindley, A.L. 2007 The dating of food vessels and urns in Ireland. Galway. Department of Archaeology, National University of Ireland. Brindley, A.L. and Lanting, J.N. 1992a Radiocarbon dates from wedge tombs. Journal of Irish Archaeology 6, 1926. Brindley, A.L. and Lanting, J.N. 1992b Radiocarbon dates from the cemetery at Poulawack, County Clare. Journal of Irish Archaeology 6, 137. Brindley, A.L., Lanting, J.N. and van der Plicht, J. 2005 Radiocarbon-dated samples from the Mound of the Hostages. In M. O’Sullivan (ed.), Duma na nGiall*the Mound of the Hostages, Tara, 28196. Dublin. Wordwell and University College Dublin. Britton, K., Mu¨ ldner, G. and Bell, M. 2008 Stable isotope evidence for saltmarsh grazing in the Bronze Age Severn Estuary, UK: implications for palaeodietary analysis at coastal sites. Journal of Archaeological Science 35, 211118. Bronk Ramsey, C. 1998 Probability and dating. Radiocarbon 40, 46174. Bronk Ramsey, C. 2000 Comment on ‘The use of Bayesian statistics for 14C dates of chronologically ordered samples: a critical analysis’. Radiocarbon 42, 199202. Bronk Ramsey, C. 2001 Development of the radiocarbon calibration program OxCal. Radiocarbon 43, 35563. Bryce, T.H. 1902 On the cairns of Arran. Proceedings of the Society of Antiquaries of Scotland 36, 74181. Bryce, T.H. 1904 On the cairns and tumuli of the Island of Bute. Proceedings of the Society of Antiquaries of Scotland 38, 1781. Buck, C.E., Cavanagh, W.G. and Litton, C.D. 1996 Bayesian approach to interpreting archaeological data. Chichester. John Wiley and Son. Buck, C.E., Christen, J.A., Kenworthy, J.B. and Litton, C.D. 1994 Estimating the duration of archaeological activity using 14C determinations. Oxford Journal of Archaeology 13, 22940. Burenhult, G. 1984 The Archaeology of Carrowmore. Stockholm. Theses and Papers in North-European Archaeology 14. Burenhult, G. 2001 The megalithic cemetery of Carrowmore, Co. Sligo. Tjo ¨ rnarp, Sweden. Go ¨ ran Burenhult. Burenhult, G. 2005 Carrowmore-tombs for hunters. British Archaeology 82, 227. Byrne, G. 1993 Rathlackan. In I. Bennett (ed.), Excavations 1992: summary accounts of archaeological excavations in Ireland, 4950. Bray. Wordwell. Byrne, G., Warren, G., Rathbone, S., McIlreavy, D. and Walsh, P. 2009 Archaeological Excavations at Rathlackan (E580). Stratigraphic Report. 45

R. J. Schulting et al. Dublin, University College Dublin School of Archaeology. Report prepared for the Heritage Council. Case, H. 1961 Irish Neolithic pottery: distribution and sequence. Proceedings of the Prehistoric Society 27, 174233. Case, H. 1969 Settlement patterns in the North Irish Neolithic. Ulster Journal of Archaeology 32, 327. Childe, V.G. 1940 The prehistoric communities of the British Isles. London. W. and R. Chambers. Cody, E. 2002 Survey of the megalithic tombs of Ireland. Vol. VI. County Donegal. Dublin. Stationary Office. Clarke, S.R. 2006 Irish court tombs: structure, morphology and landscape setting. Unpublished PhD thesis, School of Geography, Archaeology and Palaeoecology, Queen’s University Belfast. Coles, J.M. and Simpson, D.D.A. 1965 The excavation of a Neolithic round barrow at Pitnacree, Perthshire, Scotland. Proceedings of the Prehistoric Society 31, 3457. Collard, M., Edinborough, K., Shennan, S. and Thomas, M.G. 2009 Radiocarbon evidence indicates that migrants introduced farming to Britain. Journal of Archaeological Science 37, 86670. Collins, A.E.P. 1954 The excavation of a double horned cairn at Audleystown, Co. Down. Ulster Journal of Archaeology 17, 756. Collins, A.E.P. 1973 A re-examination of the Clyde-Carlingford tombs. In G. Daniel and P. Kjærum (eds), Megalithic graves and ritual, 93103. Copenhagen. Jutland Archaeological Society. Collins, A.E.P. 1976 Dooey’s Cairn, Ballymacaldrick, Co. Antrim. Ulster Journal of Archaeology 39, 17. Collins, A.E.P. and Waterman, D.M. 1955 Millin Bay, a late Neolithic cairn in Co. Down. Belfast. HMSO, Archaeological Research Publication (N. Ireland), No. 4. Collins, A.E.P. and Wilson, B.C.S. 1964 The excavation of a court cairn at Ballymacdermot, Co. Armagh. Ulster Journal of Archaeology 27, 322. Conway, M.B. and Williams, B.B. 1994 Excavation of a court tomb, Dunloy, Co. Antrim. Ulster Journal of Archaeology 57, 320. Cooney, G. 2000 Landscapes of Neolithic Ireland. London. Routledge. Cooney, G. 2004 Neolithic stone axe quarries and production sites in Scotland and Ireland. In I.A.G. Shepherd and G.J. Barclay (eds), Scotland in ancient Europe: the Neolithic and early Bronze Age of Scotland in their European context, 191203. Edinburgh. Society of Antiquaries of Scotland. Cooney, G., Bayliss, A., Healy, F., Whittle, A., Danaher, E., Cagney, L., Mallory, J.P., Smyth, J., Kador, T. and O’Sullivan, M. 2011 Ireland. In A. Whittle, F. Healy and A. Bayliss (eds), Gathering time: dating the early Neolithic enclosures of southern Britain and Ireland, 562669. Oxford. Oxbow. Cooney, G. and Grogan, E. 1994 Irish prehistory: a social perspective. Dublin. Wordwell.

46

New dates from the north Corcoran, J.X.W.P. 1969 The Cotswold-Severn group. In T.G.E. Powell (ed.), Megalithic enquiries in the west of Britain, 13104. Liverpool. Liverpool University Press. Corcoran, J.X.W.P. 1973 The chambered cairns of the Carlingford culture: an enquiry into origins. In G. Daniel and P. Kjærum (eds), Megalithic graves and ritual, 10516. Copenhagen. Jutland Archaeological Society. Danaher, E. 2004 Symbolic enclosure: a tantalising Neolithic space in the north-west. Archaeology Ireland 18, 1821. Danaher, E. 2007 Monumental beginnings: the archaeology of the N4 Sligo inner relief road. NRA Scheme Monographs 1. Dublin. National Roads Authority. Darvill, T. 2004 Long barrows of the Cotswolds and surrounding areas. Stroud. Tempus. Davies, M. 1945 Types of megalithic monuments of the Irish Sea and North Channel coastlands: a study in distributions. Antiquaries Journal 25, 12544. Davies, O. 1938 Excavations at Mourne Park. Proceedings of the Belfast Natural History and Philosophical Society 1937 38, 1826. Davies, O. 1939a Excavation of a horned cairn at Aghanaglack, Co. Fermanagh. Journal of the Royal Society of Antiquaries of Ireland 69, 2138. Davies, O. 1939b Excavations at Legland horned cairn. Proceedings of the Belfast Natural History and Philosophical Society 1939 40, 1624. De Vale´ ra, R. 1959 The court cairns of Ireland. Proceedings of the Royal Irish Academy 59C, 9140. De Vale´ ra, R. 1961 The ‘Carlingford Culture’, the long barrow and the Neolithic of Great Britain and Ireland. Proceedings of the Prehistoric Society 27, 23452. ´ Nualla´ in, S. 1961 Survey of the Megalithic Tombs of De Vale´ ra, R. and O Ireland, Volume I, Co. Clare. Dublin. Stationary Office. ´ Nualla´ in, S. 1964 Survey of the Megalithic Tombs of De Vale´ ra, R. and O Ireland, Volume II, Co. Mayo. Dublin. Stationary Office. DeNiro, M.J. 1985 Post-mortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317, 8069. Dowd, M. 2007 The use of caves for funerary and ritual practices in Neolithic Ireland. Antiquity 82, 30517. Dufour, E., Bocherens, H. and Mariotti, A. 1999 Palaeodietary implications of isotopic variability in Eurasian lacustrine fish. Journal of Archaeological Science 26, 61727. Evans, E.E. 1938a A chambered cairn in Ballyedmond Park, County Down. Ulster Journal of Archaeology 1, 4958. Evans, E.E. 1938b Doey’s Cairn, Dunloy, County Antrim. Ulster Journal of Archaeology 1, 5978. Evans, E.E. 1939 Excavations at Carnanbane, County Londonderry: a double horned cairn. Proceedings of the Royal Irish Academy 45C, 112.





47

R. J. Schulting et al. Evans, E.E. and Davies, O. 1934 Excavation of a chambered horned cairn at Ballyalton, Co. Down. Proceedings of the Belfast Natural History and Philosophical Society 1933 4, 79103. ´ .E. Fischer, A., Olsen, J., Richards, M.P., Heinemeier, J., Sveinbjo¨ rnsdo´ ttir, A and Bennike, P. 2007 Coastinland mobility and diet in the Danish Mesolithic and Neolithic: evidence from stable isotope values of humans and dogs. Journal of Archaeological Science 34, 212550. Foley, C. 2000 Prehistoric settlement in Tyrone. In C. Dillon and H.A. Jefferies (eds), Tyrone: history and society, 138. Dublin. Geography Publications. Hartwell, B. 1998 The Ballynahatty complex. In A.M. Gibson and D.D.A. Simpson (eds), Prehistoric ritual and religion, 3244. Stroud. Sutton. Hartwell, B. 2004 A Neolithic ceremonial timber complex at Ballynahatty, Co. Down. Antiquity 76, 52632. Hedges, R.E.M. and Reynard, L.M. 2007 Nitrogen isotopes and the trophic level of humans in archaeology. Journal of Archaeological Science 34, 1240 51. Hedges, R.E.M., Saville, A. and O’Connell, T.C. 2008 Characterizing the diet of individuals at the Neolithic chambered tomb of Hazleton North, Gloucestershire, England, using stable isotope analysis. Archaeometry 50, 11428. Hencken, H.O’N. 1932 The archaeology of Cornwall and Scilly. London. Methuen. Hencken, H.O’N. and Movius, H.L. 1935 A cairn at Poulawack, County Clare. Journal of the Royal Society of Antiquaries of Ireland 65, 191222. Herity, M. 1974 Irish passage graves. Dublin. Irish University Press. Herity, M. 1982 Irish decorated Neolithic pottery. Proceedings of the Royal Irish Academy 82C, 247404. Herity, M. 1987 The finds from Irish court tombs. Proceedings of the Royal Irish Academy 87C, 103281. Herity, M. and Eogan, G. 1977 Ireland in prehistory. London. Routledge and Kegan Paul. Herring, I.J. 1941 The Tamnyrankin cairn: west structure. Journal of the Royal Society of Antiquaries of Ireland 71, 3152. Hu¨ ls, C.M., Erlenkeuser, H., Nadeau, M.-J., Grootes, P.M. and Andersen, N. 2010 Experimental study on the origin of cremated bone apatite carbon. Radiocarbon 52, 58799. Jones, C. 1998 The discovery and dating of the prehistoric landscape of Roughan Hill in Co. Clare. The Journal of Irish Archaeology 9, 2743. Jones, C. 2003 Neolithic beginnings on Roughan Hill and the Burren. In I. Armit, E. Murphy, E. Nelis and D. Simpson (eds), Neolithic settlement in Ireland and western Britain, 18894. Oxford. Oxbow. Jones, C. 2007 Temples of stone: exploring the megalithic tombs of Ireland. Cork. The Collins Press. Jones, C. and Walsh, P. 1996 Recent discoveries on Roughan Hill, County Clare. Journal of the Royal Society of Antiquaries of Ireland 126, 86107.



48

New dates from the north Kilbride-Jones, H.E. 1954 The excavation of an unrecorded megalithic tomb on Kilmashogue Mountain, Co. Dublin. Proceedings of the Royal Irish Academy 56C, 46179. Kimball, M.J. 2000 Human ecology and Neolithic transition in eastern County Donegal, Ireland. British Archaeological Reports (British Series) 300. Oxford. Archaeopress. Kytmannov, T. 2008 Portal tombs in the landscape. The chronology, morphology and landscape setting of the portal tombs of Ireland, Wales and Cornwall. Oxford. Archaeopress, British Archaeological Reports (British Series) 455. Lanting, J.N., Aerts-Bijma, A.T. and Van Der Plicht, J. 2001 Dating of cremated bones. Radiocarbon 43, 249454. Lide´ n, K., Eriksson, G., Nordqvist, B., Go¨ therstro¨ m, A. and Bendixen, E. 2004 The wet and the wild followed by the dry and the tame: or did they occur at the same time? Diet in Mesolithic-Neolithic southern Sweden. Antiquity 78, 2333. Long, A. and Fritz, G.J. 2001 Validity of AMS dates on maize from the Tehuaca´ n Valley: a comment on MacNeish and Eubanks. American Antiquity 12, 8790. ´ Donnabha´ in, B. 1994 Poulnabrone, Co. Clare. The Other Lynch, A. and O Clare 18, 57. MacAdam, J. 1855 Discovery of an ancient sepulchral chamber. Ulster Journal of Archaeology 3, 35865. Mallory, J.P. and McNeill, T.E. 1991 The archaeology of Ulster from colonisation to plantation. Belfast. Institute of Irish Studies, Queen’s University Belfast. McCormick, F. 1986 Animal bones from prehistoric Irish burials. Journal of Irish Archaeology 3, 3748. McCormick, F. 2007 Mammal bone studies from prehistoric Irish sites. In E. Murphy and N. Whitehouse (eds.), Environmental archaeology in Ireland, 77101. Oxford. Oxbow. McSparron, C. 2008 Have you no homes to go to: calling time on the early Irish Neolithic. Archaeology Ireland 22, 1821. Meadows, J., Barclay, A. and Bayliss, A. 2007 A short passage of time: the dating of the Hazleton long cairn revisited. Cambridge Archaeological Journal 17, 4564. Milner, N., Craig, O.E., Bailey, G.N., Pedersen, K. and Andersen, S.H. 2004 Something fishy in the Neolithic? A re-evaluation of stable isotope analysis of Mesolithic and Neolithic coastal populations. Antiquity 78, 922. Milner, N. and Woodman, P.C. 2007 Deconstructing the myths of Irish shell middens. In N. Milner and G. Bailey (eds), Shell middens and coastal resources along the Atlantic fac¸ ade, 10110. Oxford. Oxbow. Molleson, T. 1986 New radiocarbon dates for the occupation of Kilgreany Cave, County Waterford. Journal of Irish Archaeology 3, 13. Morton, W.R.M. 1954 Report on the human skeletal remains found at Audleystown, 1952. In A.E.P. Collins, The excavation of a double horned cairn at Audleystown, Co. Down. Ulster Journal of Archaeology 17, 3753. 49

R. J. Schulting et al. Morton, W.R.M. 1955 The human skeletal material. In A.E.P. Collins and D.M. Waterman (eds), Millin Bay, a Late Neolithic cairn in Co. Down, 5981. Archaeological Research Publication (N. Ireland), No. 4. Belfast. HMSO. Murphy, E. 2003 Funerary processing of the dead in prehistoric Ireland. Archaeology Ireland 17, 135. ´ Donnabha´ in, B., Welsh, H., Tesorieri, M. and Mcsparron, C. Murphy, E., O 2010 INSTAR: The People of Prehistoric Ireland. Progress report on Phase 2*Synthesis of osteological, contextual and chronological data relating to Irish prehistoric human remains. Report prepared for the Heritage Council. Murray, E.V. 2007 Molluscs and middens: the archaeology of ‘Ireland’s early savage race’? In E.M. Murphy and N.W. Whitehouse (eds), Environmental archaeology in Ireland, 11935. Oxford. Oxbow. ´ Nualla´ in, S. 1989 Survey of the megalithic tombs of Ireland, volume V, Co. O Sligo. Dublin. Stationary Office. ´ Nualla´ in, S. 1998 Excavation of the smaller court-tomb and associated hut O sites at Ballyglass, County Mayo. Proceedings of the Royal Irish Academy 98C, 12575. ´ Rı´ordain, S.P. 1957 Tara*the monuments on the hill. Dundalk. Dundalgan O Press. O’Brien, W. 1999 Sacred ground: megalithic tombs in coastal south-west Ireland. Bronze Age Studies 4. Galway. Galway University Press. O’Brien, W. in press The Chalcolithic in Ireland: a chronological and cultural framework. In M.J. Allen, J. Gardiner, A. Sheridan and D. Mcomish (eds), The British Chalcoliithic: place and polity in the later third millennium BC. Oxford. Oxbow. O’Kelly, M.J. 1958 A horned cairn at Shanballyedmond, Co. Tipperary. Journal of the Cork Historical and Archaeological Society 63, 3772. O’Kelly, M.J. 1981 The megalithic tombs of Ireland. In C. Renfrew (ed.), The megalithic monuments of western Europe, 11326. London. Thames and Hudson. O’Kelly, M.J. 1989 Early Ireland. Cambridge. Cambridge University Press. O’Sullivan, M. 1993 Megalithic art in Ireland. Dublin. Country House. ¨ sterholm, I. and O ¨ sterholm, S. 1984 The kitchen middens along the coast of O Ballysadare Bay. In G. Burenhult (ed.), The archaeology of Carrowmore, 326 45. Stockholm. Theses and Papers in North-European Archaeology 14. Parker Pearson, M. 2005 Bronze Age Britain. London. B.T. Batsford and English Heritage. Piggott, S. 1954 The Neolithic cultures of the British Isles. Cambridge University Press. Cambridge. Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C., Buck, C.E., Burr, G.S., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P., Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., McCormac, F.G., Manning, S.W., Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S., Turney, C.S.M., van der Plicht, J. and Weyhenmeyer, C.E. 2009 IntCal09 and Marine09

50

New dates from the north radiocarbon age calibration curves, 050,000 years cal bp. Radiocarbon 51, 111150. Richards, M.P. 2000 Human consumption of plant foods in the British Neolithic: direct evidence from bone stable isotopes. In A.S. Fairbairn (ed.), Plants in Neolithic Britain and beyond, 12335. Oxford. Oxbow. Richards, M.P. and Mellars, P. 1998 Stable isotopes and the seasonality of the Oronsay middens. Antiquity 72, 17884. Richards, M.P. and Hedges, R.E.M. 1999 Stable isotope evidence for similarities in the types of marine foods used by Late Mesolithic humans on the Atlantic coast of Europe. Journal of Archaeological Science 26, 71722. Richards, M.P. and Schulting, R.J. 2006 Against the grain? A response to Milner et al. (2004). Antiquity 80, 44458. Richards, M.P., Mays, S.A. and Fuller, B.T. 2002 Stable carbon and nitrogen isotope values of bone and teeth reflect weaning age at the Medieval Wharram Percy site, Yorkshire, UK. American Journal of Physical Anthropology 119, 20510. Richards, M.P., Schulting, R.J. and Hedges, R.E.M. 2003a Sharp shift in diet at onset of Neolithic. Nature 425, 366. Richards, M.P., Price, T.D. and Koch, E. 2003b Mesolithic and Neolithic subsistence in Denmark: new stable isotope data. Current Anthropology 44, 28895. Roche, H. 2010 Ceramics. In S. Caulfield, G. Byrne, N. Dunne and G. Warren (eds), Neolithic and Bronze Age Landscapes of North Mayo: Report 2010, 11042. Dublin. Heritage Council and University College Dubin. Ryan, M.F. 1981 Poulawack, Co. Clare: the affinities of the central burial structure. In D. O’Corra´ in (ed.), Irish Antiquity, 13446. Cork. Tower Books. Saville, A. 1986 Radiocarbon dating of Hazleton North chambered tomb: a preliminary statement. In J.A.J. Gowlett and H. R.E.M (eds), Archaeology results from accelerator dating, 13337. Oxford. Oxford University Committee for Archaeology, Monograph 11. Schulting, R.J. 1998 Slighting the sea: the transition to farming in north-west Europe. Documenta Praehistorica 25, 20318. Schulting, R.J. 2005 Comme la mer qui se retire: les changements dans l’exploitation des ressources marines du Me´ solithique au Ne´ olithique en Bretagne. In G. Marchand and A. Tresset (eds), Unite´ et diversite´ des processus de ne´ olithisation sur la fac¸ ade atlantique de l’Europe (74e`me mille´ naires avant J.-C.), 16371. Paris. Me´ moire de la Socie´ te´ Pre´ historique Franc¸ aise 36. Schulting, R.J. 2007 Non-monumental burial in Neolithic Britain: a (largely) cavernous view. In L. Larsson, F. Lu¨ th and T. Terberger (eds), Nonmegalithic mortuary practices in the Baltic*new methods and research into the development of Stone Age society, 581603. Schwerin. Bericht der Ro¨ misch-Germanischen Kommission 88.

51

R. J. Schulting et al. Schulting, R.J. 2011 Mesolithic-Neolithic transitions: an isotopic tour through Europe. In R. Pinhasi and J. Stock (eds), The bioarchaeology of the transition to agriculture, 1741. New York. Wiley-Liss. Schulting, R.J. submitted On the northwestern fringes: earlier Neolithic subsistence in Britain and Ireland as seen through faunal remains and stable isotopes. In S. Shennan, K. Dobney, S. Colledge and J. Connolly (eds), The origins and spread of stock-keeping in the Near East and Europe. Schulting, R.J., Bronk Ramsey, C., Reimer, P.J., Eogan, G., Cleary, K., Cooney, G. and Sheridan, A. forthcoming Dating the human remains from Knowth. In G. Eogan and K. Cleary (eds), Excavations at Knowth 6: the archaeology of the large passage tomb at Knowth, Co. Meath. Dublin. Royal Irish Academy. Schulting, R.J., Sheridan, A., Clarke, S.R. and Bronk Ramsey, C. 2008 Largantea and the dating of Irish wedge tombs. Journal of Irish Archaeology 17, 117. Schulting, R.J., Sheridan, A., Crozier, R. and Murphy, E. 2010 Revisiting Quanterness: new AMS dates and stable isotope data from an Orcadian chamber tomb. Proceedings of the Society of Antiquaries of Scotland 140, 150. Schulting, R.J. and Richards, M.P. 2002a Finding the coastal Mesolithic in southwest Britain: AMS dates and stable isotope results on human remains from Caldey Island, Pembrokeshire, South Wales. Antiquity 76, 101125. Schulting, R.J. and Richards, M.P. 2002b The wet, the wild and the domesticated: the Mesolithic-Neolithic transition on the west coast of Scotland. European Journal of Archaeology 5, 14789. Schurr, M.R. 1998 Using stable nitrogen isotope ratios to study weaning behavior in past populations. World Archaeology 30, 32742. Scott, J.G. 1962 Clyde, Carlingford and Connaught cairns*a review. Antiquity 36, 97101. Shee Twohig, E. 1981 The Megalithic art of western Europe. Oxford. Clarendon Press. Sheridan, A. 1986 Megaliths and megalomania: an account, and interpretation, of the development of passage tombs in Ireland. Journal of Irish Archaeology 3, 1730. Sheridan, A. 1995 Irish Neolithic pottery: the story in 1995. In I.A. Kinnes and G. Varndell (eds), ‘Unbaked urns of rudely shape’, 321. Oxbow Monograph 55. Oxford. Oxbow. Sheridan, A. 2003 Ireland’s earliest ‘passage’ tombs: a French connection? In G. Burenhult and S. Westergaard (eds), Stone and bones: formal disposal of the dead in Atlantic Europe during the Mesolithic-Neolithic interface 6000 3000 BC, 926. Oxford. BAR International Series 1201. Sheridan, A. 2006 A non-megalithic funerary tradition in early Neolithic Ireland. In M. Meek (ed.), The modern traveller to our past. Festschrift in honour of Ann Hamlin, 2431. Newtownards. DPK Publishing.

52

New dates from the north Sheridan, A. 2010 The Neolithization of Britain and Ireland: the ‘big picture. In B. Finlayson and G. Warren (eds), Landscapes in transition, 89105. Oxford. Oxbow. Smith, A.G., Pilcher, J.R. and Pearson, G.W. 1971 New radiocarbon dates from Ireland. Antiquity 45, 97102. Smith, B.H. 1991 Standards of human tooth formation and dental age assessment. In A. Kelley and C.S. Larsen (eds), Advances in dental anthropology, 14368. New York. Wiley-Liss. Smith, M. and Brickley, M. 2009 People of the long barrows: life, death and burial in earlier Neolithic Britain. Stroud. The History Press. Smits, E., Millard, A.R., Nowell, G. and Pearson, D.G. 2010 Isotopic investigation of diet and residential mobility in the Neolithic of the Lower Rhine Basin. European Journal of Archaeology 13, 531. Stuart, A.J. and van Wijngaarden-Bakker, L.H. 1985 Quaternary vertebrates. In K.J. Edwards and W.P. Warren (eds), The Quaternary history of Ireland, 22149. London. Academic Press. Tauber, H. 1986 Analysis of stable isotopes in prehistoric populations. Mitteilungen der Berliner Gesellschaft fu ¨ r Anthropologie, Ethnologie und Urgeschichte 7, 318. Ubelaker, D.H. 1978 Human skeletal remains: excavation, analysis, interpretation. Chicago. Aldine. Vatcher, F.M. 1961 Seamer Moor, Yorkshire. Proceedings of the Prehistoric Society 27, 345. Waddell, J. 1998 The prehistoric archaeology of Ireland. Bray. Wordwell. Ward, G.K. and Wilson, S.R. 1978 Procedures for comparing and combining radiocarbon age determinations: a critique. Archaeometry 20, 1931. Warren, G. 2009 Belderrig: a ‘new’ later Mesolithic and Neolithic landscape in northwest Ireland. In N. Finlay, S. Mccartan, N. Milner and C.R. WickhamJones (eds), From Bann Flakes to Bushmills: Papers in Honour of Professor Peter Woodman, 14352. Oxford. Prehistoric Society Research Paper 1. Warren, G., McIlreavy, D., Rathbone, S. and Walsh, P. 2009 Archaeological Excavations at Behy (E747). Stratigraphic Report. Dublin. University College Dublin School of Archaeology. Report prepared for the Heritage Council. Waterman, D.M. 1965 The court cairn at Annaghmare, Co. Armagh. Ulster Journal of Archaeology 28, 346. Waterman, D.M. 1978 The excavation of a court cairn at Tully, Co. Fermanagh. Ulster Journal of Archaeology 41, 314. Whitehouse, N.J., Schulting, R.J., McClatchie, M., Barratt, P., McLaughlin, T.R. and Bogaard, A. 2010 Cultivating Societies. Final Report*Phase 3. Belfast, School of Geography, Archaeology and Palaeoecology. Report prepared for INSTAR programme. Whittle, A., Barclay, A., Bayliss, A., Schulting, R. and Wysocki, M. 2007a Building for the dead: events, processes and changing worldviews from the thirty-eighth to the thirty-fourth centuries cal. BC in southern Britain. Cambridge Archaeological Journal 17, 12347. 53

R. J. Schulting et al. Whittle, A., Bayliss, A. and Wysocki, M. 2007b Once in a lifetime: the date of the Wayland’s Smithy long barrow. Cambridge Archaeological Journal 17, 10321. Whittle, A., Healy, F. and Bayliss, A. (eds.). 2011a Gathering time: dating the early Neolithic enclosures of southern Britain and Ireland. Oxford. Oxbow. Whittle, A., Healy, F. and Bayliss, A. 2011b Gathering time: the social dynamics of change. In A. Whittle, F. Healy and A. Bayliss (eds), Gathering time: dating the early Neolithic enclosures of southern Britain and Ireland, 848919. Oxford. Oxbow. Williams, B.B. 1986 Excavations at Altanagh, County Tyrone. Ulster Journal of Archaeology 49, 3388. Wilson, H.E. 1972 Regional Geology of Northern Ireland. Belfast. HMSO. Woodman, P.C. 2000 Getting back to basics: transitions to farming in Ireland and Britain. In T.D. Price (ed.), Europe’s first farmers, 21959. Cambridge. Cambridge University Press. Woodman, P.C. 2001 Mesolithic middens*from famine to feasting. Archaeology Ireland 15, 325. Woodman, P.C. 2004 The exploitation of Ireland’s coastal resources*a marginal resource through time? In M.R. Gonza´ lez Morales and G.A. Clarke (eds), The Mesolithic of the Atlantic fac¸ ade, 755. Arizona State University Anthropological Research Paper No. 55. Tucson. Arizona State University. Woodman, P.C. 2009 Ireland’s place in the European Mesolithic: why it’s ok to be different. In S.B. McCartan, R.J. Schulting, G. Warren and P.C. Woodman (eds), Mesolithic horizons, xxxvixlvi. Oxford. Oxbow. Woodman, P.C., Andersen, E. and Finlay, N. 1999 Excavations at Ferriter’s Cove, 198395: last foragers, first farmers in the Dingle Peninsula. Bray. Wordwell. Wright, L.E. and Schwarcz, H.P. 1999 Correspondence between stable carbon, oxygen and nitrogen isotopes in human tooth enamel and dentine: infant diets at Kaminaljuyu´ . Journal of Archaeological Science 26, 115970. Wysocki, M., Bayliss, A. and Whittle, A. 2007 Serious mortality: the date of the Fussell’s Lodge long barrow. Cambridge Archaeological Journal 17, 6584.

54

New dates from the north

Appendix Sample contexts Aghanaglack court tomb The small amount of human remains from Aghanaglack, Co. Fermanagh cannot be confidently attributed to more than a single individual. However, different contexts are represented, and so two samples were selected. A cattle femur provided the single faunal sample. All appear to belong to the primary burial deposits of the two outer chambers (Chamber I East and Chamber I West). Though as Case (1969, 21) notes, the presence of a barbed and tanged arrowhead clearly raises a question over this. Some of the faunal remains can be attributed to Ch I West (Davies 1939a).

Annaghmare court tomb The single human element (a child mandible) available from Annaghmare, Co. Armagh can be attributed to the west side of Chamber 2. Two fragments of an adult femur were also recorded as present on the east side of this chamber. Goodland bowl sherds were found in the upper levels of the fill of this chamber. A small amount of cremated human bone was found in Chamber 1, with much more substantial amounts in Chambers 2 and 3. This material was not targeted in the present project. Pottery from Chamber 3 included carinated bowl and Ballyalton sherds (Waterman 1965).

Audleystown dual court tomb Relatively good provenance information is available for Audleystown, Co. Down. The three partial mandibles initially selected for dating represent three distinct individuals, and can all be definitely attributed to Chamber 6, which yielded the richest assemblage of human remains at the site, representing some thirteen individuals (Collins 1954, 38). The child mandible in particular has a secure context, deriving from the lowest deposits of the chamber (Collins 1954, 20, plates Va, VIb) (Plate II). The pottery in this deposit is noted as being exclusively ‘Neolithic A’, and so it was expected that the associated date would refer to the primary use of the monument. An adult mandible can be attributed to the ‘main burial layer’ of Chamber 6. Finally, a second adult mandible is part of a group of human remains found near a Food Vessel (Collins 1954). Two cremated bone samples selected, a fibula and a cranium, cannot certainly be attributed to two distinct individuals. Plain Neolithic bowls and Carrowkeel ware are both present. Chamber 1 is also represented by a pig ulna attributed to the lowest layer. The context of the cattle bone that represents the other faunal element is uncertain. Unfortunately, the selected mandibles, together with a talus and a few other small fragments (also from Chamber 6), are all that survive from the 55

R. J. Schulting et al.

Pl. II*Child mandible in situ in the lowest deposits of chamber 6 at Audleystown, Co. Down (Collins 1954, plate Va). Crown copyright, image supplied by Tony Corey, NIEA.

original large excavated assemblage of the partial and fragmentary remains of some 3033 individuals. The remaining material seems to have been lost some decades ago during a fire and flood in the Department of Anatomy at Queen’s University Belfast.

Ballyalton court tomb Ballyalton, Co. Down, held a small amount of fragmented human remains, and a relatively large assemblage of faunal remains. Two human right mandible fragments were initially selected for sampling. Two molar roots were subsequently sampled when both of these failed. Two dates on faunal material were selected: a cattle humerus and a pig ulna. Both the human and faunal remains derive from the lower deposits of both the inner and outer chambers (Evans and Davies 1934). This is the type-site for decorated Ballyalton bowls (Case 1961; Sheridan 1995).

56

New dates from the north

Ballyedmond court tomb The recovered material from Ballyedmond, Co. Down, consists only of an eroded human molar and two long bone fragments not definitely identifiable as human. This being the case, the molar root was selected for sampling. Both the tooth and the long bone fragments were found lying on an apparently undisturbed forecourt paving floor, together with sherds of Early Neolithic pottery in the form of a simple rounded bowl (Evans 1938b, 53). Some 30cm above the paving were coarse ware sherds typical of Early Bronze Age Food Vessels (Evans 1938b, 54).

Ballynahatty 1855 megalithic circular chamber Excavations at the Ballynahatty megalithic chamber, Co. Down, in the midnineteenth century produced human and animal bone that has been identified and can be securely provenanced to compartments E/F within the main circular chamber (MacAdam 1855). The unburnt bone from this site is problematic in that it has been treated with an unknown preservative resulting in a slight sheen to the bone surfaces. The calcined bone shows no indication of having been treated in this way.

Behy court tomb The Behy court tomb, Co. Mayo lies within the Ce´ ide Fields complex, and is not far from Ballyglass and Rachlackan court tombs. Excavations in the early ´ Nulla´ in 1960s confirmed its identification as a court tomb (de Vale´ ra and O 1964). A total of eight radiocarbon dates have now been obtained from Behy, most of which were obtained by UCD or SUERCC in 2001. Seven of the eight are on charcoal, with an additional determination on a human cranium from the chamber obtained through 14CHRONO, Queen’s University Belfast, in 2006. The dates are difficult to relate to the stratigraphy, but the mid-fourth millennium results obtained are in keeping with the lithic and pottery assemblage, which consists entirely of carinated bowl sherds (Roche 2010)

Legland court tomb No unburnt human bone was found at the Legland court tomb, Co. Tyrone. The fauna derives from the upper fill of Chamber I (Davies 1939), and so could refer either to disturbed primary deposits, a final blocking, or to post-Neolithic deposition. A cattle scapula chosen for dating exhibits a probable cut mark.

57

R. J. Schulting et al.

Pl. III*(a) Mandible showing cutmarks relating to dismemberment, from the central cist at Millin Bay, Co. Down (b) Detail of the cutmarks (Murphy 2003).

Millin Bay cairn, central cist and stone circle All the unburnt human remains from Millin Bay, Co. Down were found as a disarticulated jumble of bone from some fifteen individuals in an elongated central cist (Collins and Waterman 1955). Unfortunately, as with Audleystown,

58

New dates from the north the great majority of this material has been lost. Four samples were selected, representing at least three individuals (the surviving MNI). No more specific provenance information is available, but the deposit seems to show little internal differentiation, and so the four samples should provide a good indication of the period of use for the main burial deposit at the site. A cremated bone sample was also selected. One of the three mandibles selected for dating shows clear cut marks relating to dismemberment, presumably as part of a post-mortem funerary rite (Plate III) (Murphy 2003).

Mourne Park court tomb At Mourne Park court tomb, Co. Down a group of unburnt human bones was found among and under the stones sealing the north side of the inner portal of Chamber II, which appeared undisturbed (Davies 1938, 22). A cranial fragment, attributed to this group with a fair degree of confidence, was selected, as was a sheep scapula.

Parknabinnia (atypical) court tomb The skeletal remains from Parknabinnia were recovered from both chambers of the tomb but it was clear from the stratigraphy that the rear chamber (Chamber 2) was blocked at some point during the use of the tomb while deposits continued to be made in the front chamber (Chamber 1) after this blocking episode. Both chambers were partially filled with a matrix of loose stones interspersed with human bone (both inhumed and cremated), animal bone, lithic artefacts, pot sherds, and bone artefacts. The skeletal remains from Parknabinnia were highly fragmented, but out of a total 20,533 bone fragments recovered, 6,084 were identified to skeletal element as human. The majority of the remainder were most likely human as well but were not identifiable to element due to their fragmentary condition. At least twenty individuals are represented in this collection and these can be divided into at least fifteen adults and five sub-adults. The stratigraphy of the site, the radiocarbon dates, and a taphonomic analysis of the skeletal collection indicate that the burial rite at Parknabinnia consisted of successive primary inhumations that were subsequently rearranged and disturbed by later inhumations (Beckett 2005; Beckett and Robb 2006). The samples selected for radiocarbon dating from Parknabinnia were all inhumed human bone. These samples were, for the most part, from different contexts, but given the loose stony matrix within which the bones were contained and the evidence for subsequent rearrangement and disturbance, it is not certain that the dates are all from separate individuals.

59

R. J. Schulting et al.

Rathlackan court tomb Only a single small bone fragment, as yet unidentified, was recovered from the Rathlackan court tomb, Co. Mayo. A moderate-sized pottery assemblage was comprised mainly of Middle Neolithic globular bowl sherds (‘Goodland Bowl’), but also included sherds of a single Early Neolithic Carinated Bowl, a small number of probable Late Neolithic Grooved Ware sherds, an Early Bronze Age Vase Urn and Cordoned Urn sherds (Roche 2010). A total of eighteen radiocarbon determinations on charcoal and hazelnut shells are now available from the site, of which two relate to a small square structure with an internal hearth within an enclosure to the north of the court tomb, post-dating its primary use, but overlapping with its later use (as seen in the 14C dates and the pottery). The earliest dated samples relate to the rearmost Chamber 3, which may be a later addition to the monument (Byrne et al. 2009). However, there is some discrepancy between the earlier Beta dates and the more recent UBA dates from a pit in the floor of this chamber, and an additional date is pending to try to resolve this. A series of spreads and deposits in the rear chamber were sealed beneath a layer of stone forming a rough paved surface. Charcoal above this paving has been dated to the Late Neolithic, with further deposits above this providing Early Bronze Age dates and ceramics.

Tamnyrankin court tomb No unburnt human bone was recovered from Tamnyrankin, Co. Derry. Two faunal samples were selected, a cattle humerus and an ovicaprid long bone (Herring 1941). Both elements can be attributed to Chamber 1, and the humerus shows a definite cutmark. ‘Kilhoyle’ sherds were found in a side chamber (Case 1961).

Tully court tomb Adding to the five previously published dates on charcoal from Tully, Co. Fermanagh (Waterman 1978), are three new determinations on calcined human bone, deriving from each of the monument’s two chambers and from a cist. The cremated remains in Chamber 1 are identified as belonging to two children, while the majority of those in Chamber 2 are identified as belonging to a single adult, with a single fragment indicating a second individual. A few small sherds of Early Neolithic carinated bowl were found in Chamber 1, including three everted rim sherds from shouldered bowls. No later pottery forms were found. A small stone cist had been built against the outer east side of the gallery, and contained cremated bone belonging to a child (Wells in Waterman 1978).

60