Shepard - Ceramics for the Archaeologist.pdf

(3798 KB) Pobierz
204030789 UNPDF
204030789.002.png
CERAMICS FOR THE
ARCHAEOLOGIST
ANNA O. SHEPARD
Publication 609
CARNEGIE INSTITUTION OF WASHINGTON
WASHINGTON, D. C.
204030789.003.png
Manuscript submitted November 1954
Published 1956
Reprinted 1957,1961, 1963, 1965, 1968,1971, 1974, 1976, 1977,1980
/f
1985 reprinting: Braun-Brumfield, Inc., Ann Arbor
ISBN 0-87279-620-5
LC 56-4818
204030789.004.png
Foreword to Fifth Printing
Ceramic Studies, 1954 to 1964
ologist was written in 1954. The physicists have introduced sensitive and
rapid methods for compositional analysis; the availability of computers has given
statistical studies great impetus; there has been renewed interest in the potter's
natural resources; the classification of pottery has been pursued with fresh enthu-
siasm, taking different directions in different regions; ethnologists have been
inquiring into the place of pottery in the lives of village people; and the archaeol-
ogist, in his turn, has been challenged to think about the cultural significance of
pottery rather than content himself with its use for relative dating. A review of
these developments will point toward the fundamentals in ceramic studies and
suggest how the archaeologist can best obtain the aid he needs. Although this re-
view is not seasoned with either humor or satire, it is brief, and the points are car-
ried by illustration rather than by precept.
The physicist's contribution to ceramic studies. Physics has wrought and is
working a remarkable revolution in chemical analysis by the development of new
instrumental methods. In contrast to classic procedures, these methods are based
on the measurement of characteristic properties, especially atomic and electronic
properties. The methods are so varied that it is difficult to generalize about them,
but a number require relatively little material, and some are nondestructive—con-
ditions that are important to the archaeologist. They vary widely in sensitivity
and specificity; some are especially useful for the identification of impurities and
trace elements because of their extreme sensitivity; and some give a complete ele-
mental analysis whereas others can be used only for certain elements. A number
employ standards that must be prepared by classic methods, and their accuracy is
therefore dependent on the accuracy of those methods. .Most of them require ex-
pensive instruments, but in routine industrial testing the cost may be offset by the
reduced time for analysis. On the other hand, since standardization is required
when unknown material is tested, more time may be needed by the new than by
the standard method. In other words, the instrumental methods have great poten-
tialities but must be used with judgment. Reference to a few that have been ap-
plied in archaeology will illustrate their potentialities and limitations.
M ANY CHANGES in ceramic studies have occurred since Ceramics for the Archae-
204030789.005.png
IV
FOREWORD TO FIFTH PRINTING
Emission spectroscopy is a method that has been throughly tested over many
years; it offers a quick means of obtaining complete qualitative data, and its value
for quantitative analysis is well established for both routine work and research.
Its earlier applications in archaeological ceramics were mainly for pigment anal-
ysis, for which it is particularly well adapted. As an example, qualitative analyses
of Rio Grande glazes were performed in the laboratories of the Massachusetts In-
stitute of Technology to learn (a) whether production of glaze paint ware awaited
discovery of local sources of ore or whether lead ores were initially imported from
the western glaze paint centers; (b) how many sources of lead ore were known in
the Rio Grande Valley, and the influence of their location on the centers of produc-
tion and on trade (Shepard, 1942, p. 258).
In recent years spectrographic analysis has been used in the study of pastes,
with the aim of differentiating pottery from different sources. A study of Myce-
naean and Minoan pottery conducted by the Oxford Laboratory for Archaeology
and the History of Art is the most extensive and successful of these. Archaeolo-
gists of the British School of Archaeology in Athens and the Ashmolean Museum
outlined a sound and challenging problem of trade in the eastern Mediterranean
in Late Bronze Age times (Catling, 1961), and they collected an excellent sample
from sites considered centers of production in the Mycenaean-Minoan world and
from sites in the eastern Mediterranean where pottery, especially Mycenaean,
seemed to have been traded. Nine elements were chosen for quantitative spectro-
graphic anah sis because they had boon found useful in comparing and contrasting
various ty pes of pot ten (Blin-Stoyle and Richards, 1961). Initially, analysis of
10 sherds each of Mycenaean and knossian pottery showed that the composition
patterns (plot of percentages of elements) of the two were sufficiently distinct to
permit assignment of origin of sherds to one or the other if there seemed to be only
the two alternatives. Both the advantages and the limitations of the study are in-
dicated in the second report (Catling, Richards, and Blin-Stoyle, 1963) and the
review of archaeological interpretations (Catling, 1963). Eleven composition pat-
terns were recognized from the analysis of more than 500 sherds. In addition, there
were a few odd patterns that were not comparable to any of the others or to one
another. The largest group (composition pattern \) was represented by all except
a few of the- Peloponnesian sherds tested. Mycenaean-like pottery from three sites
in C\ [mis. one in north Syria, and Tell \marna, Kgypt, as well as pottery from
various sites in the \egean, had the same pattern. The second important group
'composition pattern B> occurred in most but not all the Minoan sherds. It was
much less common outside Crete than I lie Mycenaean pattern, although there are
occurrences within the Vegean area. Most of these may represenl trade, hut the
proportion of sherds with this composition pattern in Thebes in central (Jreece is
surprisingly high, making up nearly two-thirds of the sample analyzed. Catling
204030789.001.png
Zgłoś jeśli naruszono regulamin