the neo-assyrians at tell el-hesi: a petrographic study of ... · pdf filethe neo-assyrians at...
Embed Size (px)
The Neo-Assyrians at Tell el-Hesi: A Petrographic Study of Imitation Assyrian Palace WareAuthor(s): Christin M. A. EngstromSource: Bulletin of the American Schools of Oriental Research, No. 333 (Feb., 2004), pp. 69-81Published by: The American Schools of Oriental ResearchStable URL: http://www.jstor.org/stable/1357795 .Accessed: 17/08/2011 20:20
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].
The American Schools of Oriental Research is collaborating with JSTOR to digitize, preserve and extendaccess to Bulletin of the American Schools of Oriental Research.
The Neo-Assyrians at Tell el-Hesi:
A Petrographic Study of Imitation
Assyrian Palace Ware
CHRISTIN M. A. ENGSTROM
1810 Francisco Street Berkeley, CA 94703
The research described here was conducted to determine the geographical provenance of selected imitation Assyrian Palace Ware sherds unearthed at Tell el-Hesil in strata of the eighth and seventh centuries B.C.E. The goal was to establish if the unusually large numbers of Assyrian Palace Ware sherds found at Hesi were imported from Assyria proper or "imitation" pieces manufactured at or near Hesi. To that end, selected samples of imitation Assyrian Palace Ware from Hesi were thin-sectioned and their petrographic profiles compared with samples of Assyrian Palace Ware from Khirbet Qasrij, Iraq, with the geomorphology of the region surrounding Tell Jemmeh, and with petrographic descriptions of Tell Jemmeh pottery. Data presented here indicate that the Hesi Palace Ware samples were manufactured local to Hesi in the loess depositional region of the Beersheva basin.
rom approximately 745 B.C.E. until the mid- seventh century, the Neo-Assyrian empire was the most powerful polity in the Middle East.
Their hegemony spread swiftly, extending southward through the Levant and eventually into the very heart of Egypt. During this period, Assyria created a net- work of vassal states and client kingdoms out of the territories surrounding the Assyrian homeland and down through the Levant (Gitin 1997: 77). When politically expedient, the Assyrians annexed whole countries directly into the empire, forming new prov- inces ( Na'aman 1993; Oded 1970: 178).
Under Assyrian rule, the economies of the Levant prospered. The Philistine urban centers of Gaza, Ashkelon, Ashdod, and Ekron were particularly well situated to take advantage of the economic struc- tures brought about by Assyrian rule (Elat 1979: 29, 32; Gitin 1997: 79). The Philistines used the Assyr- ian trade markets and the Assyrian peace to expand
their large commercial interests by monopolizing the sea trade with Egypt and exporting local goods such as olive oil (Otzen 1979: 258; Gitin 1997: 80, 84).
Exactly who was in control of Hesi and the sur- rounding region during the eighth and seventh cen- turies is still being debated; however, an abundance of Assyrian archaeological material suggests that after 735 the area around Hesi and Tell Jemmeh was within the sphere of the Neo-Assyrian Empire, part of Philistine Gaza (Blakely and Hardin 2002; Oded 1970: 183). In contrast to Tell Jemmeh, where As- syrian-style buildings indicate that the settlement was a base of Assyrian regional power in the seventh century, Hesi had declined from a fortress settle- ment to an unfortified outpost in the border region between Philistine Gaza and the Judahite province identified as District 3 by Aharoni (fig. 1), an area that included the biblical city of Lachish (Blakely and Hardin 2002: 31; Van Beek 1983: 17; Aharoni 1997: 346).
Assyrian-style buildings, palaces, and material goods began to appear in the Levant during the late eighth century (Reich 1992). Assyrian architectural
1Hereafter referred to as Hesi. For a more complete discussion of Tell el-Hesi and a bibliography, please see Fargo 1997.
70 CHRISTIN M. A. ENGSTROM BASOR 333
"srij Gezer Carchet ish \ Jerusalem Nineve
dod. IEkron *
Ash elo• Nimru
Kh ire• - .\ aza~i• *Lac-hish Ashur
Tell -e Beit- \ 1el-Hes- Mir •m
Tell Haror Se
Tell\ Miles r~amascus el-F
•ah 0 1o 20
h) 0 10 20Km r
Sama a i;
/ rusalem Gaz
Miles N Zi Te el-Hesi 10 200
--- ----- 0 100 200
Fig. 1. Map of the Middle East with sites discussed in the text.
influence is apparent in the layout of public build- ings unearthed at Tell Jemmeh, Megiddo, and other, smaller sites, such as Ayyelet ha-Shahor/Hazor (Reich: 1975: 233; Lipschitz 1990: 96; Van Beek 1983: 12; Finkelstein, Ussishkin, and Halpern 2000: 6).2 Tell Jemmeh has a remarkably large corpus of Palace Ware for a site in the southern Levant. Much of Jemmeh's Palace Ware was recovered from the site's Assyrian palace complex (Van Beek 1983: 17; William Melson, personal communication).3
Hesi, 20 miles northeast of Tell Jemmeh, also has an unusually large amount of Palace Ware for a southern Levantine site; over 50 individual Palace Ware sherds have been excavated from Field I on Hesi's small 0.75-acre acropolis. Stratum VII is the earliest stratum in which Palace Ware has been found.
Structural remains at Hesi, in Stratum VII of Field I on the acropolis, have been dated to the Assyrian
period based on associated pottery, specifically the Palace Ware, as well as on the unique color and con- sistency of the clay used in the building construction. The color and texture of the mudbrick differs consid- erably from that of earlier and later strata (Blakely and Hardin 2002: 31, 32). Not enough of the building foundation remains, however, to conclusively identify the structure as architecturally Assyrian.
A NOTE ABOUT ASSYRIAN PALACE WARE
Assyrian Palace Ware was the luxury ware of the Neo-Assyrian Empire (Oates 1959). It was a wheel- thrown ware produced in a variety of forms includ- ing bowls, beakers, and miniature jars and bottles as well as goblets, all with everted, carinated rims. The common denominator among all these forms is the extreme thinness of the body walls and the fine levi- gation of the clay (Rawson 1954; Oates and Oates 2002). Potters highly skilled in their craft were es- sential to produce the ware's thin body walls, which
2For a more complete discussion of Tell Jemmeh and a bibli- ography, see Van Beek 1997.
3Referred to in this text as "Palace Ware."
2004 STUDY OF IMITATION ASSYRIAN PALACE WARE 71
was an advantage in the firing process. Thin body walls prevented the wares from exploding as water
escaped from the clay during firing, because the fine-
grained clays were not porous enough to allow water to efficiently vaporize (Rawson 1954: 169).
Rawson's classic account of Assyrian Palace Ware fabric describes a fine ware made in the area around Nimrud from two different kinds of clay (Rawson 1954). The first has a clay paste low in iron but rich in aluminum, with a coarse siliceous com- ponent. This ware was fired in an oxidizing envi- ronment that turned the body surface a pale pink (Rawson 1954: 169). The second type, more com- monly associated with the term "Assyrian Palace Ware," was manufactured from hyper-levigated clay, either riverine or artificially levigated in vats. This second ware also contained a high amount of alu- minous clay minerals. When fired, the ware body turned a light green to pale cream with a distinctive waxy sheen to the vessel surface. This unusual color and surface grain was consistently produced for over a century (Rawson 1954: 169). J. Oates maintains that it is not clear whether the two types are truly different clays or whether the differences in color and texture are due to differences in the firing tem- perature and kiln treatment. She describes the first type as a pinkish-buff, with a surface so pale as to be almost white, and a salmon-colored core. The second type is a buff yellow to pale green, with the surface color matching that of the core (Oates 1959: 131).
Both ware types are common to the Palace Ware corpus unearthed at Khirbet Qasrij, Iraq, which lies approximately 42 km northwest of ancient Nineveh (Curtis 1992: 154). A petrographic examination of both types in selected samples from Khirbet Qasrij shows the greenish Assyrian Palace Ware to be over 97 percent by volume fine-grained isotropic green clay paste, with the overwhelming majority of ac- cessory minerals (quartz, mica, amphibole, and some calcite) being < 10 p in size (Curtis 1989).4 The pink ware has an isotropic to anisotropic reddish brown, fine-grained, foliated paste (96 percent) with acces- sory grains of chert, quartz, and olivine (7-150 -t; the larger grains are angular); smaller grains of mica, amphibole, hornblende, orthoclase, and zircon are present in the grain fraction.
The manufacture of Assyrian Palace Ware outside of Assyria proper is well known (Courtois and Doray 1983). L. C. Courtois and A. M. Doray hypothesized that Assyrian Palace Ware forms produced in the Levant were not copied spontaneously for local in-
digenous consumption, but instead were produced locally by Assyrian exiles for Assyrian overlords and other exiles (Courtois and Doray 1983: 129, 131, 135). This conclusion was based partly on differ- ences of fine-grain mineral content in the petro- graphic profiles of Palace Ware sampled from sites scattered throughout Mesopotamia and the Levant. The petrographic profile of finely crafted Palace Ware from Nimrud, for example, was characterized by large flaked grains of altered mica (Courtois and Doray 1983: 128).
POTTERY PRODUCTION AND SOIL MORPHOLOGY OF THE NORTHERN
NEGEV, PHILISTINE COAST, AND SOUTHEASTERN SHEPHELAH
Recent petrographic studies of Levantine pot- tery have categorized the soil morphology of the southern Levant and northern Negev into broad petrographic profiles for clay types used in pottery production, and then checked how those profiles corresponded to the geography (Goren 1996: 50; Master 2003: 12). Loess-dominant profiles are most prevalent in the region of the northern Negev and southwestern Shephelah. North of Hesi, and at the northern edge of the Gaza-Beersheva basin, alluvial soils dominate the petrographic profile. Toward the coast, well-rounded sand-size quartz grains become more prominent, and south of the Gaza-Beersheva basin the loess soil profile has a dark silty paste that Master associates with the white/buff ware typical of the Negev, sometimes called "Negev Whiteware" (Beit-Arieh 1999).5
The geology of the area surrounding Hesi is predominantly characterized by succeeding deposi- tional strata of Pleistocene and Holocene loess blown in from the Sahara and the Sinai, separated by layers of carbonate nodulization (Koucky 1989: 19). Loess is an unstratified silt and clay deposit formed by wind deposition, covering up to 10 percent of the planet's land surface area (Pesci 1968). The largest concentration of loess deposition in the northern
4Thin-section samples studied by the author include: KQ117, KQ117b, KQ27, KQ132a, KQ132b, KQ121, KQ143, KQ142, and KQ120, listed in Curtis 1989: fig. 31. 5D. M. Master, personal communication.
72 CHRISTIN M. A. ENGSTROM BASOR 333
Terra Rosa Rendzina
Sand Dunes and Sandy Soils Jerusal i Loess Ashkelon
I I Sandy Regosols Tell ish .. . ..-.
. ...... .-. .. ........................ Calcareous Steppic Soils G zes Gaza
Alluvial and Colluvial Soils Dead
.. Sea Desert Soils and Lithosols
I1 Jem h
Saline Soils N h A " Tel
0 10 20 30 40km 8eersheva
Fig. 2. Soil profiles of the Negev (adapted from Horowitz 1979: 23).
Negev and southeastern hill country occurs in the Gaza-Beersheva basin, a roughly triangular feature bordered geographically by Lachish, Gaza, and Beer- sheva (see fig. 2; Goren 1996: 48). The typical modal profile for primary deposition loess from the area sur- rounding Beersheva is 40 percent coarse sand, 24.62 percent silt, and 13.87 percent clay (Reifenberg 1939: 307).
When viewed in thin section, samples of loess from the northern Negev have bimodal grain size, with the coarser fraction primarily composed of angular, well-sorted grains of quartz and accessory grains of plagioclase, hornblende, microcline, and other minerals. Grains range in size from 4 pm to > 500 pm; the average grain size is 50 pm. Also vis- ible in the groundmass are sand-sized (> 70 pm) nod- ules of fine calcite crystals, fragments of land snails, and well-rounded quartz grains. The groundmass is composed of < 4 pm grains of clay and silt in cal- cite cement (Melson and Van Beek 1992: 132). Quartz and calcite are the principal minerals; how- ever, due to the semiarid conditions of the area, weathered samples from paleosols usually contain primary feldspars and clay minerals such as micas. Small flakes of biotite and muscovite are visible in thin section (Melson and Van Beek 1992: 135, 136).
Along the Gazan coast, the soil profile is domi- nated by well-rounded sand-sized grains of quartz and accessory feldspars transported by Mediter- ranean currents from the Nile Delta and seasonal stream outlets of the Sinai. Further into the Philis- tine plains, the sand gives way to colluvial and al- luvial soil profiles interfingering with thick layers of loess deposition. To the east, the soils of the hill country are dominated by carbonate-rich rendzinas and terra rosa soils, derived from the underlying limestone (Horowitz 1979: 23). In general, source clays along the coast have a larger quartz sand com- ponent; those from the plain are loess dominant; and the clays of the hill country, to the west and north of the plain, contain more primary deposi- tional carbonates such as dolomite and limestone grain fragments. In semiarid regions, such as the area surrounding Hesi, redistribution of carbonate as a secondary deposit is often the major mechanism of soil formation. The resulting strata are rich with cal- cic horizons, nodules, and petro-calcic crusts (Reif- enberg 1939: 308).
Postdeposition, the quartz-dominant loess sedi- ment of the Negev is altered in some layers to a
CaCo3-rich eolianite known locally as kurkar (Yaa- Ion 1978: 202). In thin section, kurkar's cementation
2004 STUDY OF IMITATION ASSYRIAN PALACE WARE 73
is described as a "drusy Calcite mosaic" (Yaalon 1978: 201).
Levigation of Negev soils produces loess-rich clay higher in carbonate, and thus less plastic and less workable (Melson and Van Beek 1992: 132). But if treated properly with salt water during firing, or if the calcite component is very fine, the car- bonate can act as a flux. Carbonate in the form of micritic nodules is abundant in the sherds sampled for this study.
Aeolian, fluvial, and underlying bedrock quartz sand deposits are also found in many parts of the Gaza-Beersheva basin (Hesi itself is situated on a Barchan sand dune). In the basin itself, the aeolian sand deposition creates lenses of loess deposition with high sand-component percentages. But even in sand- and clay-rich loess sections, padogenic car- bonates will be present (Bruins and Yaalon 1979). Interfluvial loess layers sampled in the area of Neti- vot between Nahal Shiqma and Nahal Besor also contain fossil terrestrial snail mollusc fragments; their occurrence is commonly observed throughout the Gaza-Beersheva basin (Bruins and Yaalon 1979: 167).
Quantitative petrographic analysis of the Hesi sample group was done using point-counting.6 Sta- tistical data from the point-count was compared with soil profiles and integrated with typological, archae- ological, and historical information to infer a source location (Whitbread 1995: 98; Master 2001: 21).
The sample group consisted of 27 sherds from the Hesi collection (fig. 3; table 1); 17 sherds were pieces identified in the Hesi catalog as "imitation Assyrian Palace Ware" from strata of the eighth through fifth centuries. The remaining 10 were com- parison pieces such as kraters and cook pots of local origin ranging in date from the Chalcolithic to the Persian period. The determination of local origin was made through the integration of pot morphology,
typology, and fabric and stratigraphic data with data from D. Brooks's ground-breaking neutron activation analysis study of Hesi pottery (Brooks et al. 1974).7
The Palace Ware sherds unearthed at Hesi were most frequently pieces from beakers and small cari- nated bowls. The radii of the bowls averaged approx- imately 7 cm. Rim sherds determined to be from beakers had radii of around 6 cm. The more complete Palace Ware sherds from Hesi were all from com-
paratively small, shallow, carinated bowls, with the characteristic everted lip. The Hesi Palace Ware dif- fers markedly in color from the classic cream-buff and buff-green Assyrian Palace Ware of Nimrud. At Hesi the Palace Ware sherds range in color from gray-brown to orange-red (Oates 1959: 136).
The sample sherds were first grouped qualitatively into four broad petrographic profiles, with seven sub- groups. The subgroups were delineated based on color and optical properties of the paste, and the size, mineralogy, and estimated bulk percentage of the loess- and sand-sized grains (fig. 4; table 2).
In thin section, the alluvial loess group is char- acterized by a massive dark-sintered to red-sintered paste, lack of carbonate in the clay fraction, and linear foliation of void space. The grain fraction in- cludes argillaceous rock fragments, opaque inclu- sions, and loess-derived grains of quartz and feldspar.
The fabric associated with Negev Whiteware, the Negev loess profile, has a dark silty, carbonitic paste. Inclusions in the grain fraction include loess-derived quartz, land-snail fragments, and a significant mica component.
The loessial profile is dominated by loess-derived grains of silt-sized quartz, concertal quartz, micro- cline, and plagioclase. The paste is typically red, silty, and carbonitic, but in areas of the sample where the trace iron is less oxidized, the optically active, carbonate-rich paste appears tan to light brown. The grain fraction often includes land-snail fragments and other biocarbonitic material.
Moza marl is a soil type Goren associates with the Cenomanean section of the Judaean hills. The 6Point-counting is a technique that relies on the Delesse rela-
tion, which holds that areal proportions of minerals in thin sec- tions are equivalent to volumetric proportions of minerals in rocks (Stoltman 1991: 103-4). This study used the multiple intercept- true intersection method of point-counting. This method results in a frequency distribution of area covered by different-sized parti- cles expressed as percentage of the area (Middleton, Freestone, and Leese 1985: 71, 73).
7Point-count analysis for this project involved the use of a 1-mm counting interval over the entire ceramic area of the thin section. Three hundred fifty points were counted on each sherd; void space points were included. To facilitate maximum objec- tivity, thin sections were counted in groups of mixed types.
74 CHRISTIN M. A. ENGSTROM BASOR 333
2 3 1
7 ( 5 6
4 E ........ '
7 8 12...V...
7, 171.3 l4,3
185 *7541 6.1117
0 10 20
Fig. 3. Sample group of 27 sherds from the Hesi collection.
2004 STUDY OF IMITATION ASSYRIAN PALACE WARE 75
TABLE 1. Sample Group of 27 Sherds from the Hesi Collection, Illustrated in Figure 3
No. Catalog No. Object Stratum Exterior
1 H75.I.21.194.2946 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum VI 10YR 5/2
2 H77.I.11.314.2152 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum VI 10YR 7/2
3 H77.I.11.314.2960a, b Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum VI 2.5YR 6/2
4 H73.I.11.314.2494 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum VI 7.5YR 6/2
5 H77.I.316.3072 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum VIIId 7.5YR 5/2
6 H75.I.21.194.4258 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum Vb 10YR 7/3
7 H77.I.11.31.2946 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum VI 5YR 6/1
8 H75.I.21.194.3939 Imitation Assyrian Palace Ware bowl with exterior hematite bands Stratum Vd 7.5YR 5/4
9 H77.I.11.314.2146 Imitation Assyrian Palace Ware bowl with carinated, inverted rim Stratum VI 10R 6/6
10 H75.I.32.1813a Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum Vb 10YR 4/1
11 H75.I.32.141.1815 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum Vb 2.5YR 5/6
12 H77.I. 11.324.5085 Imitation Assyrian Palace Ware bowl with carinated, everted rim Stratum VI 7.5YR 5/2
13 H77.I.11.314.2959 Imitation Assyrian Palace Ware bowl with carinated rim and burnished exterior Stratum VI 5YR 4/1
14 H77.I.32.178.3318, 3317 Imitation Assyrian Palace Ware goblet Stratum Vd 10YR 7/3
15 H77.I.11.314.2327 Imitation Assyrian Palace Ware with carinated, everted rim Stratum VI 10R 5/6
16 H77.I.11.314.3185 Imition Assyrian Palace Ware bowl with an inverted rim Stratum VI 10R 6/6
17 H75.I.41.226.1117 Imitation Assyrian Palace Ware bowl, burnished inside Stratum V 10R 5/6
18 H77.I. 11.314.5370 Persian-period juglet bottom Stratum VI 7.5YR 7/4
19 H75.I.21.194.2941 Bowl with everted lip, burnished exterior Stratum Vd 10YR 5/2
20 H75.I.21.185.2208 Chalcolithic pinch handle Stratum III/IV 10YR 7/3
21 H77.I.11.315.2376 Bowl or pitcher rim Stratum VIId 7.5YR 6/4
22 H75.I.21.194.3630 Hematite banded bowl Stratum Vd 7.5YR 5/4
23 H77.II.314.2490 Bowl with everted lip, burnished exterior Stratum VI 10YR 7/2
24 H70.I.2.66.3107 Chalcolithic pinch handle and rim Stratum Va 5YR 7/4
25 H77.I. 11.324.5060 Krater with a burnished slip exterior Stratum VI 5YR 6/6
26 H77.I.11.314.2317 8th/7th century B.C.E. cook pot rim, scored on the exterior Stratum VI 5YR 4/3
27 H73.I.32.179.3305 8th century B.C.E. krater with anthropormorphic handles Stratum Vd 5YR 6/6
soil type has a paste that is red, silty, optically ac- tive, and highly carbonitic. The grain-sized fraction is usually dominated by gray rhomboid dolomite crystals ! 20 gi (Goren 1995: 301; 1996: 51).
Sherd H77.I.11.314.3185 (Group G) falls into a qualitative group lacking a loessial fabric defined by silt-sized quartz grains but matching Goren's defi- nition of Moza marl. The rim of sherd 3185 has an inverted lip that is not typical of the classic morpho- logic definition of Assyrian Palace Ware (Oates 1959: 136). These characteristics led to the conclu- sion that sherd 3185 may be an undefined fine-ware import of the Judaean hill country and not in fact Pal- ace Ware, although it was identified as "Imitation Assyrian Palace Ware" in the Hesi ceramics catalog.
Sherd H75.I.41.226.1117 (Group F) is also atyp- ical of the classic Assyrian Palace Ware morphol- ogy. The sherd rim is inverted until thickening and flaring out at the lip, and it is the only Palace Ware sherd in loessial Group F Because sherds 1117 and 3185 do not fit with classic Assyrian Palace Ware morphology and are from anomalous or imported petrographic groups, they were omitted from the quantitative analysis.
POINT-COUNT PETROGRAPHIC ANALYSIS
Gleaning meaningful statistical analysis from the point-count modal data was challenging due to the
76 CHRISTIN M. A. ENGSTROM BASOR 333
i...L. LJ_.... ..... (01... .)l
Fig. 4. Qualitative petrographic groups. (A) Alluvial loess, Group A. H126.96.36.1994.5060. (B) Alluvial loess, Group B, H188.8.131.526.3072. (C) Alluvial loess, Group C, H77.l.1.1.314.2960. (D) Negev loess, Group D, H184.108.40.2064.2146. (E) Loessial, Group E, H220.127.116.114.2152. (F) Loessial, Group F, H18.104.22.1689.3305. (G) Moza marl, Group G, H22.214.171.1244.3185.
2004 STUDY OF IMITATION ASSYRIAN PALACE WARE 77
TABLE 2. Qualitative Petrographic Groups, Illustrated in Figure 4
Alluvial Loess 1 Negev Loess [Loessial [Moza Marl
Group A Group D Group E Group G
Palace Ware Comparison Palace Ware Palace Ware Comparison "Palace Ware"*
H77.I.11.324.5085 H77.I.11.315.2376 H77.I.11.314.2146 H77.I.11.314.2152 H75.I.21.185.2208 H70.I.126.96.36.19907 H73.I.31.300.3939 H77.I.11.314.5370 H75.I.32.141.1815 H77.I.11.314.2959 H77.I.11.314.3185
H77.I. 11.324.5060 H73.I.11.314.2494 H77.I.11.314.2490 H75.I.21.194.2956
Palace Ware Comparison H77.I.11.316.3072 H75.I.21.194.2941 H75.I.21.194.4258 H77.I. 11.314.2946 H75.I.32.141.1813a H77.I.32.178.3318
"Palace Ware" Comparison H75.I.41.226.1117 H75.I.21.194.3630
H77.I. 11.314.2317 H75.I.32.179.3305
* Originally labeled as "Assyrian or Imitation Assyrian Palace Ware" in the Hesi catalog, now replaced with "unknown bowl."
small size of the sample group; however, a few trends in grain-size fractions stood out (tables 3, 4). Point-count modal analysis of the sample sherds showed that sherds from the alluvial group had an average silt-sized volume percentage of 21.70 per- cent, with a standard deviation of ?2.88. The Negev loess group had a comparable profile with an aver- age silt-volume percentage of 21.78 percent and a standard deviation of ?5.39. By contrast, the range of silt-volume percentages from loessial soil sam- ples at Tell Jemmeh is 25 to 50 percent, with actual percentage rates projected to be at least 10 percent higher (Melson and Van Beek 1992: 132). As in the Tell Jemmeh soil study, only hard minerals, princi- pally quartz and feldspar, were counted for modal grain size. Given the hypothesis that the Negev loess group soil profile is found south of the Gaza-Beer- sheva basin, the Negev loess would seem to be the most likely candidate for a profile matching Tell Jemmeh soil samples. Different data collection pro- cedures may account for the statistical difference, as multiple-intercept point-counting used on finely levigated pottery can create artificially high volume percentages for groundmass (Middleton, Freestone, and Leese 1985: 71).
Overall, modal analysis placed the Palace Ware in the lower range of the typical loess profile, where greater than 80 percent of the particles are silt-sized (Tsoar and Pye 1987: 139). The grain-fraction vol-
ume percentage of silt-sized particles averaged 78.45 percent for the Hesi Palace Ware, with a stan- dard deviation of ?9.98.
Palace Ware from Tell Jemmeh and Nineveh was observed by Melson and Van Beek to have chemically high concentrations of CaCo3 but to lack observable grain fractions of calcite and other carbonates, "sug- gesting firing temperatures in excess of 840-9000 C" (William Melson, personal communication). The majority of the Palace Ware sherds from Hesi appear to have been fired at temperatures lower than 8400 C. Of the 15 Palace Ware sherds, 11 have clearly de- fined calcite crystals and other carbonate grains larger than 4 mm across. None of the Palace Ware pieces have carbonate grains larger than 250 mm. Two Palace Ware sherds, H77.I.11.314.2327 (Group C) and H75.I.21.194.4258 (Group B), appear to have been higher fired than the other Palace Ware in the study. Their groundmass is darker and more vitreous with abundant parallel vesicles, and they lack visible carbonate grains in the coarse fraction. These two sherds also have similar color and styling, although H77.I.11.314.2327 is a much finer piece (the sherd itself is markedly thinner), and the paste is a more vitreous ware than its fellow.8
8A full set of data will be included with the Tell el-Hesi final report of Fields I and III.
78 CHRISTIN M. A. ENGSTROM BASOR 333
TABLE 3. Silt Grain Fraction Volume Percentages
Group A Palace Ware
Sherd Percentage H77.I. 11.324.5085 85.85
Standard Deviation ?5.63
Sherd Percentage H77.I. 11.315.2376 73.08
H77.I. 11.314.5370 53.85
H77.I. 11.324.5060 60.50
H77.I. 11.314.2490 78.85
Standard Deviation ?11.43
Sherd Percentage H77.I. 11.316.3072 85.33
H77.I. 11.314.2946 76.53
Standard Deviation ?4.606
Sherd Percentage H75.I.21.194.2941 69.41
Sherd Percentage H77.I. 11.314.2960 75.64
Standard Deviation ?4.16
Group D Palace Ware
H77.I. 11.314.2146 55.23
Standard Deviation ?14.77
Group E Palace Ware
Sherd Percentage H77.I.11.314.2152 57.14
H77.I. 11.314.2959 74.29
Standard Deviation ?12.13
Sherd Percentage H75.I.21.185.2208 39.54
Group F Palace Ware*
Sherd Percentage H75.I.41.226.1117 15.38
Standard Deviation ?27.10
Group G Palace Ware*
Standard Deviation ?20.30
* Originally labeled as "Assyrian or Imitation Assyrian Palace Ware" in the Hesi catalog, now replaced with "unknown bowl."
2004 STUDY OF IMITATION ASSYRIAN PALACE WARE 79
TABLE 4. Silt Percentages
Palace Ware Sherd Percentage
Group A H77.I.11.324.5085 26.00
Group B H77.I.11.316.3072 18.28
Group C H77.I.11.314.2960 16.85
Standard Deviation ?2.88
Palace Ware Sherd Percentage
Group D H77.I.11.314.2146 16.57
H73.I. 11.314.2494 26.85
Standard Deviation ?5.39
Palace Ware Sherd Percentage
Group E H77.I.11.314.2152 11.42
Standard Deviation ?7.68
Given that the loess-rich petrographic profiles of the alluvial, Negev, and loess groups do not match the fine-grained clay profiles of the Khirbet Qasrij Assyrian Palace Ware, it seems reasonable to as- sume that Hesi's Palace Ware is not Assyrian Palace Ware imported from Assyria. The Palace Ware from Hesi used in this study can be grouped into three petrographic categories: alluvial loess, loessial, and Negev loess.
Sherds from the alluvial loess group have a petro- graphic makeup consistent with the area along the northern edge of the loess depositional region of Gaza-Beersheva. This is the group most likely to be actually from Hesi, its immediate vicinity, or an- other site to the north of the basin with an Assyrian presence in the seventh century, such as Khirbet Hoga. The alluvial loess group includes compari- son sherd H77.I. 11.314.2490, a seventh-century bowl type categorized by Blakely and Gitin as "local ware" (Blakely, personal communication). Blakely has sug- gested that this bowl type was made in the region surrounding Hesi. This, along with the majority of the comparison sherds being from the alluvial loess groups, supports the suggestion of a local origin for the group.
Hesi and the surrounding area is also a possible source area for the loessial petrographic profile. The sherds in Group F have very high carbonate content
in the paste with very little sand or coastal inclu- sions, suggesting a more inland origin, inside the margins of the Beersheva basin, consistent with the Hesi locale.
The closest petrographic match for the Negev loess group is the Tell Jemmeh loess profile outlined by Melson and Van Beek (Melson and Van Beek, personal communication; Master 2003: 24). The pos- ited depositional environment for the Negev loess profile is geographically consistent with the Tell Jemmeh area, and qualitative descriptions of the Hesi Group D fabric match descriptions of loessial Palace Ware fabric from Tell Jemmeh (Melson and Van Beek, personal communication).
Tell Jemmeh was one of a series of sites along the Nahal Besor in the northwestern Negev that became outposts of the Assyrian Empire in the middle of the seventh century. During this period of Assyrian domination in the Negev, Hesi would likely have been a satellite of Tell Jemmeh or Gaza (Van Beek 1983). If so, it would seem more than likely that the Palace Ware found at Hesi was manufactured at or near Tell Jemmeh and exported the relatively short distance to Hesi. Alternatively, Group D wares could have come from Assyrian sites even closer to Hesi, such as Tel Sera, Tell el-Farah (South), or Tel Haror (Oren 1993).
It is clear that the Palace Ware from Hesi was manufactured in the loess depositional region of the Beersheva basin, or around its edges at the northern
80 CHRISTIN M. A. ENGSTROM BASOR 333
border of the western Negev and the southern edge of the coastal plain. The sherds delineated in Group D were likely made at Tell Jemmeh or a site close to it, and the Palace Ware sherds delineated in Groups A, B, C, E, and F were probably made at or
near Tell el-Hesi. Group G was an outlier group of imported fine ware incorrectly labeled as "Imitation Assyrian Palace Ware" in the Hesi catalog, originat- ing in the Judaean hill country near Jerusalem.
This paper would not have been possible without the guidance and advice of a great many people and the will- ingness of many more to lend expertise and research ma- terials. Among those the author wishes to thank are Jeffrey
Blakely, William Melson, Daniel Master, Jeffery Swope, Jonathan Tubb, John Curtis, Louise Joiner, Ian Freestone, Kevin O'Connell, S.J., John Spencer, Brian Hesse, and Alan Carroll.
Aharoni, Y. 1979 The Land of the Bible: A Historical Geography,
ed. A. E Rainey. Trans. A. E Rainey, from Hebrew. 2nd ed., rev. and enl. Philadelphia: Westminster..
Beit-Arieh, I. 1999 Tel 'Ira: A Stronghold in the Biblical Negev.
Monograph Series 15. Tel Aviv: Institute of Archaeology, Tel Aviv University.
Blakely, J. A., and Hardin, J. W. 2002 Southwestern Judah in the Late Eighth Century
B.C.E. Bulletin of the American Schools of Ori- ental Research 326: 11-64.
Brooks, D; Bieber, A. M. Jr.; Harbottle, G.; and Sayre, E. V. 1974 Biblical Studies through Activation Analysis of
Ancient Pottery. Pp. 48-80 in Archaeological Chemistry, ed. C. W. Beck. Advances in Chem- istry Series 138. Washington, D.C.: American Chemical Society.
Bruins, H. J., and Yaalon, D. H. 1979 Stratigraphy of the Netivot Section in the
Desert Loess of the Negev (Israel). Acta Geo- logica Academiae Scientiarium Hungaricae 22: 161-69.
Courtois, L. C., and Doray, A. M. 1983 Technologie des ceramiques Levantines au
temps de la domination Assyrienne. Pp. 125-36 in Comptes Rendus du 108e Congres National des Socidtes Savantes. Grenoble: Ministere de L'6ducation Nationale Comit6 des travaux his- toriques et scientifiques. Paris: CTHS.
Curtis, J. 1989 Excavations at Qasrij Cliff and Khirbet Qasrij.
British Museum Western Asiatic Excavations 1. London: British Museum.
1992 Recent British Museum Excavations in As- syria. Journal of the Royal Asiatic Society, Series 3, vol. 2, part 2: 147-64.
Elat, M. 1978 The Economic Relations of the Neo-Assyrian
Empire with Egypt. Journal of the American Oriental Society 98: 20-34.
Fargo, V. M. 1997 Hesi, Tell el-. P. 22 in The Oxford Ency-
clopedia of Archaeology in the Near East, Vol. 3, ed. E. M. Meyers. New York: Oxford University.
Finkelstein, I.; Ussishkin, D.; and Halpern, B. 2000 Megiddo III: The 1992-1996 Seasons. 2 vols.
Monograph Series 18. Tel Aviv: Institute of Ar- chaeology, Tel Aviv University.
Gitin, S. 1997 The Neo-Assyrian Empire and Its Western
Periphery: The Levant, with a Focus on Philis- tine Ekron. Pp. 77-103 in Assyria 1995: Pro- ceedings of the 10th Anniversary Symposium of the Neo-Assyrian Text Corpus Project, Hel- sinki, September 7-11, 1995, eds. S. Parpola and R. M. Whiting. Helsinki: Neo-Assyrian Text Project.
Goren, Y, 1995 Shrines and Ceramics in Chalcolithic Israel:
The View through the Petrographic Micro- scope. Archaeometry 37: 287-305.
1996 The Southern Levant in the Early Bronze Age IV: The Petrographic Perspective. Bulletin of the American Schools of Oriental Research 303: 33-72.
Horowitz, I., ed. 1979 The Quaternary oflsrael. New York: Academic.
2004 STUDY OF IMITATION ASSYRIAN PALACE WARE 81
Koucky, F L. 1989 The Present and Past Physical Environment of
Tell el-Hesi, Israel. Pp. 5-36 in Tell el-Hesi: The Site and the Expedition, eds. B. T. Dahl- berg and K. G. O'Connell, S.J. Joint Archaeo-
logical Expedition to Tell el-Hesi 4. Winona Lake, IN: Eisenbrauns.
Lipschitz, O. 1990 The Date of the "Assyrian Residence" at Ayye-
let ha-Shahar. Tel Aviv 17: 96-99. Master, D. M.
2001 The Seaport of Ashkelon in the Seventh Cen-
tury B.C.E.: A Petrographic Study. Ph.D. disser- tation, Harvard University.
2003 Trade and Politics: Ashkelon's Balancing Act in the Seventh Century B.C.E. Bulletin of the American Schools of Oriental Research 330: 47-64.
Melson W. G., and Van Beek, G 1992 Geology and the Loessial Soils, Tell Jemmeh,
Israel. Geoarchaeology 7: 121-47. Middleton, A. P.; Freestone, I. C.; and Leese. M. N.
1985 Textural Analysis of Ceramic Thin Sections: Evaluation of Grain Sampling Procedures. Ar-
chaeometry 27: 64-74. Na'aman, N.
1993 Population Changes in Palestine following Assyrian Deportations. Tel Aviv 20: 104-24.
Oates, J. 1959 Late Assyrian Pottery from Fort Shalmaneser.
Iraq 21: 130-46. Oates, J., and Oates, D,
2002 Nimrud: An Assyrian Imperial City Revealed. London: British School of Archaeology in Iraq.
Oded, B. 1970 Observations on Methods of Assyrian Rule in
Transjordania after the Palestinian Campaign of Tiglath-Pileser III. Journal of Near Eastern Studies 29: 177-86.
Oren, E. D. 1993 Ethnicity and Regional Archaeology: The West-
ern Negev under Assyrian Rule. Pp. 102-5 in Biblical Archaeology Today, 1990: Proceed- ings of the Second International Congress on Biblical Archaeology, Jerusalem, June-July 1990, eds. A. Biran and J. Aviram. Jerusalem: Israel Exploration Society.
Otzen, B. 1979 Israel under the Assyrians. Pp. 250-61 in
Power and Propaganda: A Symposium on An-
cient Empires. Mesopotamia, ed. M. T. Larsen.
Copenhagen Studies in Assyriology 7. Copen- hagen: Akademisk.
Pesci, M. 1968 Loess. Pp. 674-78 in The Encyclopedia of
Geomorphology, ed. R. W. Fairbridge. New York: Reinhold.
Rawson, P. S. 1954 Palace Ware from Nimrud: Technical Observa-
tions on Selected Examples. Iraq 16: 168-72. Reich, R.
1975 The Persian Building at Ayyelet ha-Shahar: The Assyrian Palace of Hazor? Israel Explora- tion Journal 25: 233-37.
1992 Palaces and Residences in the Iron Age. Pp. 202-22 in The Architecture of Ancient Israel: From the Prehistoric to the Persian Periods, eds. A. Kempinski and R. Reich. Jerusalem: Israel Exploration Society.
Reifenberg, A. 1939 The Loess Soils of the Beersheba Region of
Palestine. The Empire Journal of Experimental Agriculture 7: 305-10.
Stoltman, J. B. 1991 Ceramic Petrography as a Technique for Docu-
menting Cultural Interaction: An Example from the Upper Mississippi Valley. American Antiquity 56: 103-20.
Tsoar, H., and Pye, K. 1987 Dust Transport and the Question of Desert
Loess Formation. Sedimentology 34: 139-53. Van Beek, G.
1983 Digging up Tell Jemmeh. Archaeology 36, no. 1: 12-19.
1997 Jemmeh, Tel. Pp. 213-15 in The Oxford Ency- clopedia of Archaeology in the Near East, Vol. 3, ed. E. M. Meyers. New York: Oxford
University. Whitbread, I. K.
1995 Greek Transport Amphorae: A Petrological and Archaeological Study. British School at Athens, Fitch Laboratory Occasional Paper 4. Exeter: Short Run.
Yaalon, D. H. 1978 Geoderma-Continental Sedimentation, Cal-
crete, Desert Loess and Paleosols, Sand Dunes and Eolianite. Tenth International Congress on Sedimentology, Jerusalem, 1978. Post-Con- gress Excursion Y5. Jerusalem: International Conference of Sedimentologists (offprint).