Chemistry Reference
In-Depth Information
concentration of more than 40 elements in each shard, the data were sub-
jected to statistical analysis, disregarding any archaeological information
about the samples. This made it possible to divide the samples into “trial
compositional groups” and then to ascertain whether single samples fitted
into any of the groups (Bieber et al. 1976; Harbottle et al. 1969). Chemical
analyses by neutron activation made it possible also to differentiate among
pottery groups of different provenance. Obtaining reliable information on
the provenance of any specific shard would require, however, a very large
amount of data, which would have to be derived from a very large number
of analyses (estimations vary between 10,000 and probably more than
100,000 pieces of pottery), too large a number to be analyzed with the ana-
lytical equipment and statistical techniques available even today (Adan-
Bayewitz et al. 1999; Perlman 1984).
Dating Pottery
Most archaeological pottery has been and still is dated using typological cri-
teria. During the last half-century, however, scientific dating techniques such
as radiocarbon (see Textbox 52) and thermoluminescence (see Textbox 24)
have reduced the burden that typology once entailed. Even so, a major
problem in radiocarbon dating of pottery arises from the contamination of
samples during burial, as a result of the absorption of carbon-containing
substances from the burial environment. Although this problem may be
overcome if the carbon in the pottery samples is purified, the accurate radio-
carbon dating of archaeological pottery continues to be an area of ongoing
research. Some more recent work suggests that the time of use of pottery can
be dated indirectly by the radiocarbon methods: if for example, organic
matter such as oil, stored in ceramic vessels was absorbed within the walls
of the vessels and preserved there in sufficient amount to be dated by radio-
carbon, dating the absorbed matter would provide the time of the latest use
of the vessel (Stott et al. 2003).
