Two Hiatuses in Human Bone Radiocarbon Dates in Britain (17 000 to 5000 Cal BP)
Blockley, Stella M., Antiquity
Examining radiocarbon estimates from c. 10 000-2000 [sup.14]C yr BP on human skeletal remains from cave sites in Britain, Chamberlain (1996) found an apparent gap at 7000-6000 [sup.14]C yr BP (with only one date in the 5999-5000 bracket). In the present study, the radiocarbon profile is re-examined, with the dataset extended to include older material down to 17 000 cal BP, and as recent as 5000 cal BE The dataset was expanded to include calibrated dates on all human bone (Table 1 at http://antiquity.ac.uk/projgall/blockley), and radiocarbon dates taken from anthropogenically modified non-human material (Table 2 at http://antiquity. ac.uk/projgall/blockley). With this new dataset it is clear that the hiatus initially identified by Chamberlain (1996) remains, but that another hiatus is apparent, from 13 850-11 100 cal BE Possible reasons for these remarkable gaps in the record of human bone dates from Britain are discussed.
In Chamberlain's 1996 study, dates that were within 400 radiocarbon years of each other on the mean uncalibrated date were grouped and assigned to 1000 year bins. The data were presented as a histogram, showing numbers of incidences of deposition of human bone against radiocarbon years. In the new database, the raw radiocarbon estimates were calibrated (to 2 [sigma]) using INTCAL 98 (Stuiver et al. 1998; Bronk Ramsay 1999) with the terrestrial calibration curve, except where evidence of marine protein necessitated the use of marine curves in order to offset the effects of old carbon in marine systems (the marine reservoir offset, see Haflidasen et al. 2000; Blockley et al. 2004). This applies to individuals from two sites on Oronsay (Cnoc Coig, Casteal nan Gillean, see online Table 1 at http://antiquity.ac.uk/projgall/blockley); the individuals from Cnoc Coig show 97-100 per cent marine dietary input (Schulting & Richards 2002a), and as such dates have been calibrated, using INTCAL 98, with the marine calibration curve (Stuiver et al. 1998; Bronk Ramsey 1999). The individual from Casteal nan Gillian, however, has only 58 per cent marine protein. The calibration of the date has thus been adjusted to reflect this lower marine input, using mixed terrestrial and marine calibration data (Stuiver et al. 1998; Bronk Ramsay 1999).
Rather than combining dates for sites as Chamberlain (1996) did, the data were individually plotted, allowing all the chronological data for each site to be shown and compared with other datasets, such as proxy climate data and radiocarbon dates on anthropogenically modified non-human material. Dates on humanly modified material as a proxy for human presence have been used before (see Jacobi & Pettitt 2000 for Aurignacian Somerset, Street & Terberger 1999 for occupation in Europe at the Last Glacial Maximum). The intention here is to compare the three datasets (the climatic sequence, dates taken directly from human bone, and dates from anthropogenically modified material) and to draw conclusions holistically from their relative places in time (Figure 1).
[FIGURE 1 OMITTED]
For the climatic sequence, the Greenland Ice Core [delta] [sup.18]O record GRIP SS08c is used, because it is a revised chronology based on layer-counting to 14 700 ICYBP (see Johnsen et al. 2001) and has now been adopted as the stratotype for the Last Glacial to Interglacial Transition (end Devensian) and early Holocene for Europe (Lowe et al. 2001). It must be noted that Ice Core Years and Calibrated Radiocarbon time are not the same, using as they do different methods to approach a chronology. However, as both methodologies constitute an attempt to reach 'true' years, or calendar time, they can be compared.
Upon calibration, the data shift slightly; the second hiatus remains, but is emphasised (Figure 1). The addition of other dates (as mentioned above) effectively reduces the period of the hiatus to some 500 years. …