The chemical composition, or “fingerprint,” of an object can be determined by trace- and major-element analysis. The method, neutron activation analysis (NAA), measures the quantities of a large number of constituent elements from only a small sample of an object (typically 50–100 mg). NAA's widest application in archaeology has been in determining the chemical composition of ceramics in order to establish place of origin. NAA has also been applied with great success to obsidian, flint, glass, jade, marble, and basalt. By applying statistical tests to the composition of a collection of samples, it is possible to determine if the samples come from the same or a different source. It is also possible to determine where an artifact originated by matching its chemical composition to that characterizing a particular site.
However, because the methods employed in ceramic production can change the composition of the clay (e.g., if temper is added or clays are mixed) and because the chemical reference compositions characterizing different pottery-production sites may not be available, tracking the sources of ancient pottery is a complicated task. Reference compositions are usually not based on clay analysis but on the chemical analysis of ceramics that can be unequivocally assigned to a specific site or region. The composition of the ceramic reflects the composition of the clay plus whatever else the potter added to the clay. Wasters are ideal for establishing local compositions. Kiln and tabun ceramic linings, mud bricks, footbaths, and loom weights can serve as reference materials to establish a local chemical profile. Ceramic chemical compositions that persist over long time periods at a given site and that represent different types of pottery can also serve as reference groups. The use of ceramic rather than clay compositions in NAA is also preferred because the composition of a given clay bed may not be uniform over a wide area and because it usually is not known where ancient potters got their clay. The availability of chemical reference compositions is a technical problem that is gradually being overcome by judicially selected analysis and interlaboratory calibrations. One of the advantages of NAA over other analytical techniques is that NAA laboratories can be intercalibrated very precisely.
A fairly simple way to uncover similarity between chemical compositions, and hence commonality of origin, is to do a regression analysis, in which a constant relationship between two chemical compositions is sought. If a constant factor is found that connects every element of one composition to every element of the other, the two represent the same clay source, diluted to different degrees. Considering dilution is a common but critical factor in chemical analyses of pottery.
In order to determine the place of manufacture of a collection of ceramic objects sampled, the chemical fingerprint of the clay sources or ceramic compositions at various sites across a wide area is needed. Without such reference groups, NAA, coupled with statistical analysis, can only tell whether the objects measured came from one or more sources—although this information is valuable in itself.
NAA provides precision and accuracy for a large number of elements; the method uses instruments (no chemical preparation of the sample is required; hence, I[nstrumental]NAA is often used in tests in place of NAA); many element abundances are measured simultaneously; measurements can be automated; errors of measurement can be precisely determined; and large numbers of samples can be measured in relatively short time round the clock with little human attendance.
[See also Analytical Techniques.]
- Perlman, I., and F. Asaro. “Pottery Analysis by Neutron Activation Analysis.” Archaeometry 11 (1969): 21–52.
- Yellin, Joseph, I. Perlman, F. Asaro, H. V. Michel, and D. F. Mosier. “Comparison of Neutron Activation Analysis from the Lawrence Berkeley Laboratory and the Hebrew University.” Archaeometry 20 (1978): 95–100.
- Yellin, Joseph, and A. M. Maier. “Origin of the Picotrial Krater from the ‘Mycenaean’ Tomb at Tel Dan.” Archaeometry 34 (1992): 31–36.
- Zorn, Jeffrey, Joseph Yellin, and J. Hays. “The M(W)SH Stamp Impressions and the Neo-Babylonian Period.” Journal of the Israel Exploration Society 44 (1994): 161–183.