Published online by Cambridge University Press: 12 April 2016
It may be laid down as a general rule in regard to the dating of pottery that shape, technique, and fabric must all be taken into account, but the first is much the most important. Fabric or ware is dependent on local materials and consequently remains constant over long periods, and helps therefore very little for dating purposes. Nor is technique or surface-finish always a sure chronological guide, for it is controlled usually by the potters' whim, or by tribal custom, and not by the purchasing public, and is therefore liable to irregular or sudden changes. Shape, on the other hand, though dependent in part on the potters' skill and on the function of the pot, is far more dependent on the public taste, and is consequently more likely to evolve gradually and regularly. Nor do its changes take long in their evolution. It often needs but a few centuries, even in a conservative country like Mesopotamia, for pot-shapes to develop so markedly that the connexion between the first and last of a series would be unrecognizable without the guidance of the intermediate variants. Conversely, if in an extensive array of pottery a close connexion can be shown to exist between specimens from the earliest and latest levels, it is as likely as not that no long period of time need be postulated for the development from one to the other.