New Radiocarbon Dates from Bougon and the Chronology of French Passage-Graves

Article excerpt

New radiocarbon determinations from northwest France further contribute to the proposition -- 25 years old now -- that the megaliths of the region are astoundingly early in the west European Neolithic sequence.

European archaeologists were considerably surprised when the first radiocarbon determinations from monumental chambered tombs in Brittany were published some 25 years ago. These determinations suggested that the earliest of the so-called passage graves, notably Barnenez and Ile Guennoc, were almost 6000 years old (Radiocarbon 8 (1966): 76; Radiocarbon 13 (1971): 214-5). Subsequent calibration of radiocarbon dates to correct for variations in atmospheric 14C pushed the true ages of these monuments back to around 4500 bc. This made them some 2000 years older than the first Egyptian pyramids, and identified them as perhaps the oldest monumental stone architecture in the world (Renfrew 1973). The debate over the true antiquity of the passage graves has recently been reopened by scholars who have pointed to deficiencies in the radiocarbon determinations published in the 1960s. Some of the determinations have unacceptably large uncertainties; others are from charcoal which could conceivably precede the construction of the tomb. It has also been pointed out that where two determinations are available for the same burial chamber they are often in conflict. Supporters of this revisionist line of argument have suggested that the earliest reliable dates for passage grave construction are some four or five centuries later than the date of 5800 b.p. originally proposed (e.g. Boujot & Cassen 1992).

The importance of this question is shown by recent general interpretations of ideology and monument construction in Neolithic western Europe. One school of thought emphasizes the role of central European farming communities of Bandkeramik type in spreading new ideas and cultural attitudes to northern and western Europe (Hodder 1990; Sherratt 1990). This argument holds that in these peripheral regions of Europe, the linear ideology represented by the Bandkeramik long house was transformed into the long burial mound. Long mounds might therefore be expected to be the earliest kind of monument in these parts of Europe. By contrast, the early radiocarbon dates from Barnenez and Ile Guennoc suggest that in northwest France passage graves were at least as early as the first of the long mounds. It remains possible, therefore, that passage graves were an early and independent development of indigenous Atlantic communities, though the case relies heavily on the disputed evidence of the early passage-grave dates.

The solution to this divergence of views lies in new, more reliable radiocarbon determinations based on material which is clearly associated with the construction or early use of passage graves, and in taking multiple samples for dating which can be situated within a stratigraphic sequence. In terms of material, the preference must be for human remains which relate directly to the burial function and are less likely than charcoal to be residual from some other context. The Barnenez and Ile Guennoc passage graves are located in Brittany, and suffer from the fact that acid soils have destroyed the human bone. It is therefore necessary to look to adjacent limestone areas with good bone preservation to achieve satisfactory results.

Passage graves are found throughout the whole of northwestern France, from Normandy to Poitou. The only examples outside Brittany with radiocarbon determinations of around 5800 b.p. are at Bougon near Poitiers. This is a cemetery of five burial mounds containing eight stone-built chambers (Mohen 1977; 1986). Two samples of human bone from excavations in the 1970s gave determinations of around 5800 b.p., one from chamber F0 (Ly-1700, 5830 |+ or -~ 160 b.p.), the other from chamber E1 (Ly-966, 5800 |+ or -~ 230 b.p.). These agree with the early Breton determinations. The large uncertainties, however, make the Bougon determinations imprecise estimates of the age of the deposits to which they relate, and for this reason further samples have recently been taken from chamber F0 for verification of the chronology. …