The Kiln Dump

by Aleydis Van de Moortel

A large mass of about 26,000 pottery fragments, weighing more than 450 kg, was found in and around the Kommos kiln, covering a large part of the South Stoa of Building T and spilling onto the court (Figs. 8, 13-17, 19, 21, 26-28, 30-31).1 Several features of this deposit, apart from its location, indicate that the bulk of it represents waste from the kiln operation. First, there are more than 300 ceramic wasters distributed throughout the deposit, as well as small numbers of burned and overfired sherds and badly deformed vessels.2 Some vases could not be fitted together properly, pre-

  1. I would like to thank Joseph W. Shaw and Jeremy B. Rutter for permission to publish the kiln pottery. A warm debt of gratitude is due also to J. B. Rutter for his continuous guidance during the course of my pottery studies. The interpretation of the kiln pottery has also benefited greatly from advice given by Gloria London and William D. Glanzman. Needless to say, the responsibility for any remaining oversights and inadequacies is entirely mine. Professor V. La Rosa and Nicola Cucuzza have graciously given me permission to see the unpublished pottery from Seli di Kamilari. Cucuzza's dissertation on this pottery has been published jointly with La Rosa (La Rosa and Cucuzza 2000). The discussions of burned kiln superstructures, wasters, and coarse slabs owe much to information collected by Doniert Evely in his forthcoming book on Minoan crafts (part 2), and by Beatrice McLoughlin in her honors thesis (1993) on ancient Greek kilns. My sincere thanks to Cucuzza, Evely, and McLoughlin for permission to refer to their unpublished manuscripts. Finally I would like to thank the Classics Department at the University of Washington in Seattle for enabling my access to the University library.
  2. A few much-deformed vases from the Kommos dump have been inventoried: C 10335 (trench 95A, pails 143, 144) and C 10294 (trench 95A, pails 119, 139, 143, 144, 150, 152). The term "pail" at Kommos refers to a basic excavation unit. The term "waster" is used specifically to refer to badly overfired sherds that are deformed and often cracked, are gray to black in color, and have surfaces marked by projecting inclusions or lime-spalling; occasionally the surfaces may have a shiny brown color. Wasters have been reported from elsewhere at Kommos, but always in small quantities. One was found in MM III-LM IA rubble located in and above Room CH 16 (Kommos II, p. 164, no. 1381). A second vase fragment dated to MM IIB is insufficiently overfired to be considered a waster according to the definition given above (contra Betancourt [Kommos II, p. 92, no. 416]). Its interior shows a grayish hue, whereas its exterior preserves a polychrome decorative scheme. Its shape is somewhat warped, but as a result of pressures applied when the handle was attached rather than from overfiring. Two other wasters are known from a LM I fill on the hill (Kommos III, pp. 4, 9, nos. 147, 152). Concentrations of wasters have been found in association with Minoan kilns at Aghia Triada (Levi and Laviosa 1986, pp. 18-28, figs. 17, 20, 23, 26), at Kavousi (Gesell, Day, and Coulson 1988, p. 291; McLoughlin 1993, p. 15), and elsewhere (Evely in press). Mac-Gillivray (1987, p. 276) reports seeing tens of thousands of wasters, mostly dating to MM IIIB and LM IA, with the remains of a kiln at Silamos, but because of the huge numbers quoted it appears that he uses the term "waster" to refer to general kiln waste rather than in the sense defined above.

sumably because they had broken during firing, and their parts subsequently had been warped (Fig. 25).3 Second, the large majority of the pottery is highly repetitive in fabric, shape, and decoration, and it looks strikingly fresh, as if it had never been used.4 Finally, even though the restorable vases represent common household shapes, the absence of mendable cooking pots, lamps, and braziers would be unusual for an ordinary household assemblage at Kommos, but fits well within the kiln dump interpretation (see below).5

In the upper parts of the kiln and throughout the dump, small quantities of earlier sherds have been found, which are datable to MM IB, MM II, MM III, and early LM IA. These sherds are small in size and are worn at the edges as well as on the surfaces. Some may derive from the kiln superstructure (see above, p. 23), but most of them, since they were distributed widely and not just in areas of fallen superstructure debris, are likely to have served as fire supports. During the loading of the kiln before firing, they would have been set in between the unfired vases or between those and the clay-lined superstructure of the kiln to prevent them from

  1. Only three such examples have been recognized at present, but their number may increase as more vessels are mended. Two (C 10306, C 10531) are conical cups found inside the kiln, in channels 2 and 3, respectively. The third example is a bell cup (23) from the dump.
  2. The fresh condition of most of the pottery is indicated by the minimal wear of surfaces and fractures, and especially by the absence of wear patterns on rims, handles, or the undersides of bases that could be related to use. Another aspect of their fresh condition is the large average size of the sherds, as illustrated by their great average weight (see Tables 2-4 and Figs. 28, 30-31).
  3. For MM III household assemblages from the Central Hillside at Kommos, see Wright 1996. According to recently developed chronological criteria (see below, pp. 90-91), a few deposits from this area may now be datable to early LM IA.

Figure 25. Vases thought to have broken during firing. Arrows indicate the warped areas, which do not join properly anymore. T. Dabney touching during firing.6 During the emptying of the kiln after firing, these supports would have been discarded together with the broken pottery. These earlier sherds will not be considered in the present study.

The distribution of joins among the pottery pieces found inside the kiln suggests that these vases were part of the last firing load (see below, p. 28). If this interpretation is correct, the fact that pottery from within the kiln joins with fragments from the dump would mean that part of the last load had been thrown on the dump.7

Because of the substantial size of the deposit, only a preliminary study has been conducted thus far, and that is what is presented here. Its purpose is to establish the range and frequency ofvessel shapes and decorative motifs within the kiln output, as evidenced by the finds, and to propose a date range for the kiln's operation. In addition, some aspects of LM IA pottery production at Kommos will be discussed. The catalogue includes representative examples of each shape thought to have been fired in the kiln (1-59). Shape frequencies have been estimated on the basis of countable diagnostic features, such as spouts, handles, bases, and rims (Tables 2-4).

Mending the kiln pottery was not as straightforward as it first seemed it would be. From the distribution ofjoining sherds it appears that parts of the dump had been moved about, while unknown portions had been carried off in the LM IA phase to be used elsewhere on the site (see below, pp. 40-41). These circumstances greatly increased the difficulty of finding joins, and as a result few vases could be restored to near completeness. For the purposes of this preliminary study, therefore, vase types have been accepted as kiln products on the basis of consistency in fabric, decoration, and state of preservation with the bulk of the vases in the dump and kiln. They include large fragments or restorable examples in a fresh condition, and each type is usually represented by at least five examples.8

Cooking vessels have been eliminated from consideration because they are very fragmentary and quite rare, representing only 6% of the sherds in the dump (by count). Also rejected were the highly fragmentary remains of some twenty vases of different shapes carrying dark-painted motifs on a lustrous ground (e.g., 62-70). This associated pottery, not actually produced in the kiln, nevertheless had been thoroughly mixed with the kiln material, so it appears that the kiln dump had been used in antiquity for other kinds of ceramic waste as well.

  1. For ethnographic examples of the use of sherds as fire supports, see Voyatzoglou 1974, p. 23; Blitzer 1984, p. 156; 1990, p. 696. Blitzer (1984, p. 156, note 39) stresses that the Kentri potters use spacers only to keep the pots from touching the kiln walls; the pots are allowed to touch one another. The same practice is found among traditional Cypriot potters (London 1989b, p. 221).
  2. Such joins occurred in basins 54

55 (see below), bridge-spouted jars

C 10490 and C 10567, cylindrical vase C 9957, and jar C 10591. None of the other presently inventoried vases from the kiln have joins in the dump, but it is likely that more joins will be discovered in the future as more vessels are assembled.

8. The only exception made to the last criterion is for three piriform rhyta, which in fabric, decoration, and condition of preservation are entirely consistent with the accepted kiln dump pottery (see Table 4 and p. 80 below).

Was this article helpful?

0 0

Post a comment