12th October 2011, 07:47 PM
P Prentice Wrote:.....the majority of dates in older publications are pretty much useless because they were taken from dumps of burnt looking 'stuff' which may have been from a multiple burning episode and when calibrated now usually produce a date range often over half a millennium or more. some people used to be able to guess the real date from the uncalibrated BP date but nowadays it is better to use the most uptodate calibration programme
There are plenty of single-object old dates floating around though? [or are you suggesting starting again on all the dating and forget what's gone before?]- they just need running through modern calibration, the basic science of counting from the sample hasn't changed much, merely what's then done with the data