Order of construction
The radiocarbon evidence from the models for the stone settings provides a most likely order for the all of the dated constructional events of:Model 1
Sarsen Circle > Stonehole E > Sarsen Trilithons > Beaker burial > Bluestone Circle > Bluestone Horseshoe > Z Holes > Y Holes.The probability of this is though only 27%.
Model 2
Stonehole E > Sarsen Trilithons > Beaker burial > Bluestone Circle > Bluestone Horseshoe > Z Holes > Y Holes.The probability of this is only 31%.
Model 3
Sarsen Circle > Stonehole E > Sarsen Trilithons > Beaker burial > pit_WA_2448 > Bluestone Circle > Bluestone Horseshoe > Z Holes > Y Holes.The probability of this is though only 13%.
All three chronological models for the stone settings at Stonehenge presented above have produced stable model outputs with the prior beliefs they contain being compatible with the available radiocarbon dates. Thus, although the statistical models have allowed us to combine different types of information, we ultimately still need to use archaeological judgement to decide between them.
The models are all based on the belief that the major settings are the product of single (relatively quick) unitary episode of activity rather than the result of longer and more piecemeal episodes of construction (Bayliss et al 2007b, 46). Given the limited number of samples available at present such an assumption remains the only pragmatic way of modelling the chronology. Sensitivity analyses have highlighted the key component of these models that determines the differences in the monuments chronology is the relationship between the Sarsen Circle and Trilithons. The choice of a preferred model is therefore at present a simple matter of archaeological interpretation, and without further excavation to provide more samples associated with the major constructional events (for example the Sarsen Circle), reaching agreement is likely to be some way off.
Model 3 is our preference for the chronology of the monument because it incorporates what we believe to be the most reliable reading of the stratigraphy of the stone settings (Darvill and Wainwright 2009, Darvill et al 2012; Parker Pearson et al 2007, 2009).
Notes:
Stonehole E is by the Slaughter Stone.Pit WA 2448 is the large hole that was dug by the Great Trilithon that was identified by Atkinson as the erection ramp, an idea that Mike Parker Pearson et al dismissed: http://nora.nerc.ac.uk/9174/1/Age_of_Stonehenge_Antiquity-2.pdf
The probabilities only add up to 71% which indicates there is still a large amount of uncertainty in the results.