PREVIOUS   NEXT   CONTENTS   HOME 

2.0 Key southern Levantine ceramic materials

2.1 Groups derived from clay formations

2.1.1 Moza clay (Figure 1)

Characterisation: The clay colour is yellowish, turning to pink as the firing intensity increases and to dark grey at high temperatures. The inclusions are homogeneous, densely spread, well-sorted fine sand of whitish particles. In thin section, the Moza group is characterised by fine clay, somewhat calcareous and rich in iron oxides, and sand-sized inclusions of euhedral dolomite crystals.

Interpretation: Based on the extensive body of reference material (below), this petrographic group is identified as originating from clay of the upper member of the Moza formation, mixed with dolomitic sand that was quarried from the 'Amminadav' formation.

Distribution: This petrographic group is well known from pottery assemblages from sites of different periods spread throughout the Judean-Samarian hills. In the Chalcolithic period, it apparently typifies the southern Judean sites (Goren 1995), but is very rare in other regions. It has been recorded from Early Bronze Age I sites in central Israel extending from the Beer-Sheva valley to Aphek (Porat 1989b, 47-48). In the Early Bronze Age IV - Middle Bronze Age the Moza group dominates sites in the vicinity of Jerusalem (Goren 1996a).

In the Middle Bronze and the Late Bronze Ages, it continued to be one of the common raw materials of the Judean ceramic assemblages (Glass et al. 1993). This group is also known from the Iron Age I assemblages of Tell en-Nasbeh and Radana, from the collared-rim pithoi at Giloh (Goren 1996b) and from the pottery assemblage of Shiloh (Glass et al. 1993, 278). It was also found to be common in the Iron Age II assemblage from the City of David (Franken and Steiner 1990, 79-85) and the contemporary sites of Ramot 06 and Moza (Goren 1996b).

Although dolomite rocks and the dolomitic sand derived from them can be found in other parts of the southern Levant, it seems that its use as inclusions was restricted to the Judean hills. This was confirmed by the results of a neutron activation analysis study that attributed pottery of this petrographic group to the Jerusalem area even when it was found in the Negev and Sinai (Gunneweg et al. 1985, 273). Many of the vessels from Kuntillet 'Ajrud, Beer-Sheva, and Giloh, analyzed by Gunneweg et al. (1985), have also proved to belong to this petrographic group.

2.1.2 Taqiya Marl (Figure 2)

Characterisation: The fabric is whitish or yellowish in colour and turns into greenish-grey at higher firing temperatures. Microscopically it is characterised by a pale, calcareous marl containing sparse microfossils and densely spread iron oxides. The microfaunal assemblage within the matrix, when identified, is usually of Pliocene-Eocene age. Fine, fibrous carbonate crystals, sometimes exhibiting weak optical orientation, are abundant in the matrix. Well-sorted, sparsely distributed silty quartz appears in many cases.

Interpretation: Based on its mineralogical and palaeontological affinities, this clay is identified as marl of the Taqiya Formation of Paleocene age (Bentor 1966, 72-73). It is exposed in the Jordanian Plateau, in the northern and central Negev, in the Judean Desert and along the western slopes of the Judea-Samaria anticline. This formation is almost constant in its stratigraphic position and even in details of its composition. Equivalent beds appear in Egypt (Esna shales) and even in Morocco and Turkey (Bentor 1966, 73). However, the use of Esna shales for pottery production has never been recorded in Egyptian assemblages.

Distribution: In Chalcolithic contexts Taqiya marl was commonly used at sites adjacent to its exposures (Goren 1995). The use of Taqiya marl with the addition of various non-plastic materials was observed by Porat (1989a, 177-78) in the Early Bronze Age II assemblages of the Negev, especially at Arad, for the production of jars and holemouth jars. In EBIV contexts this group dominates the assemblage of Har Yeruham site (Goren 1996a), the only site in the central Negev where it proved to be the primary group. Taqiya marl was the main source for pottery production at Tell Gezer (Bullard 1970). In the Iron Age I it was recorded as one of the main raw materials for the production of collared-rim pithoi (Glass et al. 1993).

2.1.3. Lower Cretaceous clay and shales (Figures 3 and 4)

Characterisation: This group has been described in detail by Greenberg and Porat (1996), and Goren (1995; 1996a). It is characterised by argillaceous, ferruginous, shale-rich clay, or fine non-calcareous, smectitic clay with pronounced optical orientation. The first type usually has a relatively high content of characteristic ferruginous ooliths. In most cases quartz sand is present, usually as subrounded grains. Other indicators of this group are diversified shale fragments, most of which are ferruginous while others tend to be more clayey. Pellets, tuff and weathered basalt fragments, and typical rhombohedral limonitic pseudomorphs after dolomite occasionally occur in some cases.

Interpretation: A large body of comparative data enables us to determine that in this case, the lower formations of the Lower Cretaceous lithological section were used as a source for both matrix and inclusions. These formations outcrop widely in Lebanon, from the southern Beq'a through the Lebanon Mountains to the Lebanese coast between Beirut and Tripoli. In the southern Levant they are exposed in the eastern Jordan Valley, between the southern Dead Sea and Wadi Zarqa, and the eastern Upper Galilee. Smaller outcrops appear in Wadi Malikh and Wadi Farcah in eastern Samaria and in the northeastern Negev craters (Greenberg and Porat 1996; Goren 1995; 1996a).

Distribution: This clay is usually considered to be of superior quality for pottery production since its high iron content made it possible to achieve a high degree of sintering at lower firing temperatures owing to the fluxing properties of the iron. Therefore, Lower Cretaceous ferruginous shales were often used to produce resistant vessels, especially liquid containers such as pithoi, jugs and storage jars. At the Chalcolithic site of Teleilat Ghassul, for example, most of the locally-made pottery of Tell 1 is formed of this iron-rich clay (Goren 1991a, Appendix 2; Edwards and Segnit 1984). Similar raw materials were used during the Early Bronze Age II-III in order to produce the high quality 'Metallic Ware' as well as other superior vessels (Porat 1989a, 71-74; Greenberg and Porat 1996). In the EBIV this petrographic group is frequent at all the EBIV sites of the central Negev, and at some it constitutes a major part of the assemblage (Goren 1996a, Table 2). Recent examination of Iron Age I 'collared-rim' and 'Galilean' pithoi from several sites in the Samaria and Galilee regions (Glass et al. 1993; Cohen-Weinberger and Goren 1996) demonstrated that in several cases such clays were preferred by the potters in order to produce high quality vessels.

2.2 Groups derived from soils

2.2.1 Terra Rossa soil (Figure 5)

Characterisation: This group is usually characterised by its dark reddish-brown colour, often with a darker core, and silty appearance that can be observed with a magnifying glass or stereomicroscope. Under the petrographic microscope, it appears as silty, non-carbonatic, rather ferruginous matrix, that commonly exhibits isotropic properties. The inclusions commonly represent wadi sand, usually of limestone, chert and quartz in differing proportions, and are frequently accompanied by vegetal matter such as straw.

Interpretation: This petrographic group is identified as terra rossa soil, mixed with sieved wadi sand and/or chaff. Terra rossa is widely exposed over the mountainous regions within the Mediterranean climatic zones of the southern Levant. Therefore, the provenance of vessels belonging to this petrographic group cannot be determined on the basis of their clay alone. The reference material from Levantine sites, however, together with the inclusions that appear in the vessels, may indicate a more specific provenance.

The inclusions in most of the recorded cases contain quartz, chalk, and chert. This indicates that the sands were collected in wadis that drain areas where either Senonian or Eocene formations are exposed, since these are the two major ages when chalk and chert were deposited in the southern Levant. The quartz sand was probably swept inland by aeolian deposition. The combination of terra rossa, and Senonian or Eocene chalk and chert formations strongly suggests that the upper Shephelah should be preferred as the main origin of this group. In this area, terra rossa soil appears alongside Eocene chalk rich in chert horizons. Wadis that cross this area drain the Senonian chalk and chert formations that are exposed on the slopes of the Judean ridge to the east.

Distribution: The use of terra rossa soil as clay for ceramic vessels begins with Early Bronze assemblages (Goren 1996b). In most cases, pottery belonging to this group can be attributed on typological grounds to Judea or the upper Shephelah. All the published examples are related to the Iron Age.

Terra rossa soil, mixed with wadi sand, crushed calcite crystals or grog, was frequently used by Iron Age potters in Judea as a ceramic raw material, especially for the production of cooking pots. In Jerusalem, the City of David excavations produced numerous ceramic figurines, most of which were made locally from terra rossa soil (Goren et al. 1996). More relevant is the case of the LMLK (The King's) stamped jar handles in which 180 examples were examined using NAA (Mommsen et al. 1984). The results suggest that the jars bearing these handles were produced at a single site, perhaps in the Shephelah. In the most recent petrographic study of these vessels it has been shown that the fabric was made of terra rossa soil with chalk, quartz and chert temper.

2.2.2 Hamra soil (Figure 6)

Characterisation: In this group, sand from the Coastal Plain of Israel is mixed with highly ferruginous, fine clay. The quartz sand is usually accompanied by sand-sized grains of accessory minerals including hornblende, epidote, zircon, feldspars and augite (Nahmias 1969).

Interpretation: The nature of this group, together with its geographical distribution in Levantine sites, may point to a coastal origin. It is most likely that the Hamra red soil of the central littoral areas of Israel was used here, perhaps after some purification by dilution of the sand component. Hamra soil is spread along the Coastal Plain of Israel from the Ashdod area northwards.

Distribution: Pottery produced from Hamra is distributed in Israel at sites located mainly along the central Coastal Plain. It is known from some Chalcolithic assemblages of the central Coastal Plain (Goren 1991b), and from Early Bronze Age assemblages at Palmahim (Braun 1995). A Middle Bronze Age kiln site containing vessels only of this petrographic group was excavated near Yavneh Yam (Singer-Avitz and Levy 1992).

2.2.3 Loess soil (Figure 7)

Characterisation: A matrix of silty, rather calcareous clay, and inclusions of sand of either quartz or calcareous rock fragments define this group. The silty component contains mainly quartz, but also a recognisable quantity of other minerals including hornblende, zircon, minerals of the mica group, feldspars, tourmaline, augite and, rarely, garnet, epidote, and rutile. The silt is relatively well-sorted and comprises about 10%-20% of the matrix. The non-plastic assemblage includes either well-sorted, well-rounded sand-sized quartz grains, sometimes with the addition of other minerals (feldspars, hornblende, zircon, augite), or limestone sand with occasional chert. The two categories can appear together in various proportions. In other cases, chalk sand may appear as the dominant or the sole non-plastic component.

Interpretation: Based on a large body of published data, the matrix is readily identified as the loess soil which occurs in Israel, mainly in the northern Negev and the southern Shephelah (Porat 1989a, 50-52; Goren 1996a; Gilead and Goren 1989, 7; Goldberg et al. 1986). The inclusions that accompany this matrix are variable, and indicate different depositional environments within the loess soil distribution in the southern Levant (see below).

Distribution: Several studies of ceramics assigned to this group demonstrate that the composition of the inclusions used varies with the geographic location of each site (Gilead and Goren 1989, figure 2; Goren 1991a, 118-20, figure 13; 1995, figures 3-8). The use of loess with inclusions in which limestone is the dominant component is prevalent mainly at sites northeast of the Beer-Sheva Valley and in the southern Shephelah, whereas at northwestern Negev sites quartz is the major constituent (Gilead and Goren 1989, figure 2). In the inner southern Shephelah, chalk sand is commonly the sole inclusion. In any of the periods investigated, the overall distribution of pottery of loess-calcareous sand/quartzitic sand groups does not extend significantly beyond the limits of the triangle formed by Lachish, Gaza and Beer-Sheva, which includes Tel Hesi.

The main references to the use of loess for pottery production is based on the study of Chalcolithic ceramic assemblages in southern Israel (Gilead and Goren 1989; Goren 1991a; 1995). It is also known from EBIV contexts (Goren 1996a). In a recent study, Byzantine ceramic workshop wasters from numerous production sites where loess was used for the manufacture of vessels (especially jars) were examined. The sites are located between Gaza, Ashdod, and Beer-Sheva. The variants within this group differ geographically and can be isolated and defined petrographically (Fabian and Goren in press). This is due to the fact that the selection of potential ceramic raw materials in the northern Negev is very limited. Consequently, in most cases, as petrographic analysis has demonstrated, loess and local sands were used throughout the duration of pottery production in this area.


 PREVIOUS   NEXT   CONTENTS   HOME 

© Internet Archaeology URL: http://intarch.ac.uk/journal/issue9/goren/sect2.html
Last updated: Tue Oct 24 2000