Between Israel and Phoenicia: The Iron IIA–B Fortified Purple-dye Production Centre at Tel Shiqmona

Abstract

The history of Tel Shiqmona, on Israel’s Carmel coast, in the Iron Age has remained almost totally obscure since its excavation some 50 years ago. Recent analysis has revealed the site’s singularity—the only one around the Mediterranean that can be demonstrated to have produced the luxurious purple dye for half a millennium. This article is the first discussion of a central episode (three strata) in the site’s history. We argue that during the Late Iron IIA, the Kingdom of Israel, probably under the Omrides, replaced a small Phoenician village with a fortified casemate enclosure in order to control and institutionalise the production of the dye and other industries. These peaked under Jeroboam II, and subsequently the fort was ravaged during the period of unrest in Israel after this monarch’s reign. We discuss the historical and cultural picture emerging from a meticulous analysis of the stratigraphy and finds and address trade contacts and regional, historical and geopolitical contexts.

Keywords:

• Tel Shiqmona
• Casemate walls
• Kingdom of Israel
• Phoenician material culture
• Israelite material culture
• Purple dye
• Iron Age II

Introduction

The site, its excavation and significance

Tel Shiqmona is a very modest mound of less than one hectare, situated on a small rocky headland between the Carmel Mountains and the Mediterranean Sea (Fig. 1), on the southern outskirts of modern-day Haifa. Between 1963 and 1977 large-scale excavations were conducted on the mound by Joseph Elgavish on behalf of the Haifa Museum (800 m²; Elgavish 1994: 33). It produced a complex stratified architectural sequence with an unusual wealth of artefacts and ecofacts. Nevertheless, 50 years later, the site’s Iron Age sequence is mostly unpublished and under-studied and therefore has largely remained under the archaeological radar.

Fig. 1: Location map ofTel Shiqmona (map by Sveta Matskevich,processed by ltamar Ben-Ezra)

In order to reevaluate Elgavish’s stratigraphic sequence, excavations were renewed by Shay Bar between 2011 and 2013 (Bar 2021). In the current paper we integrate the relevant architecture of Bar’s excavations.

In 2016, we initiated the Shiqmona Early Periods Project under the auspices of the Zinman Institute of Archaeology at the University of Haifa in order to reconstruct the site’s occupational history and regional and inter-regional role during the Late Bronze and Iron Ages. We soon realised that dozens of (fragmentary) purple-stained clay vats (Fig. 2) could be attributed to all ten Iron Age strata, revealing the unique nature of the site. During most of the Iron Age, Tel Shiqmona’s inhabitants produced the lucrative purple dye—one of the most prized commodities of the ancient world (e.g., Marín-Aguilera, Iacono and Gleba 2019).¹ This role explains the establishment of the site near a high rocky submerged reef, which endangered any boat approaching the shore, and did not allow Tel Shiqmona to develop into an anchorage.² The rocky infrastructure is an excellent habitat mainly for the trunk murex (Hexaplex trunculus; Amar 2014: 35), the main species identified in chemical analyses of purple remains in Tel Shiqmona’s vats (Koren 2013; Sukenik et al. 2017; and yet unpublished chemical analyses).³ Indeed, Tel Shiqmona’s maritime environment is one of the best such habitats along the coast of the Southern Levant.

Fig. 2: Clay potsherds stained with purple dye (photo by Golan Shalvi, image processing by Sapir Haad; registration numbers and other markings on the sherds are by the original excavation expedition)

Aims and structure

We have previously dealt with Stratum 10 of the late 8th century BCE (‘the Four-Room House stratum’, hereafter 4RH; see Shalvi and Gilboa 2022a and more in Supplementary Material 2) and with the Iron IIC sequence, Strata 9–7 of the 7th century BCE (Shalvi and Gilboa 2022b). In the current paper we deal with the construction and developments of Tel Shiqmona’s casemate enclosure, in which we defined three strata (13–11), spanning most of the Late Iron IIA (Ir2a) and most of the Iron IIB (Ir2b), from the first half of the 9th throughout much of the 8th century BCE (for a comprehensive discussion of the chronology and chronological terms employed in this paper, see Supplementary Material 1). The main aim of this paper is to allow scholars for the first time to integrate this unique site into the archaeological and historical discourse regarding the Levant and more broadly the Eastern Mediterranean in the 9th–8th centuries BCE. In order to accord a firm archaeological basis to our arguments we present in the paper and in Supplementary Material 2 the stratigraphy of the casemate enclosure, its architecture and a preliminary discussion of its function, and we establish its chronology accurately on the basis of ceramic analysis.⁴ We selected comparanda for the pottery mainly from the best stratified sequences of the Kingdom of Israel—both far (Tel Reḥov, Tel Beth-Shean and Tel Hazor) and near (Tel Megiddo, Tel YoqneꜤam and Ḥorvat Rosh Zayit);⁵ from the south Phoenician sphere (Tell Abu Hawam,⁶ Tell Keisan, Tel Kabri, Tyre, Sarepta, Tell el-Burak and further up to al-Mina); and from nearby Tel Dor, the cultural and political affiliation of which changes through the Iron Age (Gilboa, Sharon and Bloch-Smith 2015). For parallels to the ceramics in Figs. 5–8, 10–11 and 15–17, see Supplementary Material 3.

Fig. 3: Plan of Stratum 13; the location of 4RH of Stratum 10 is marked by dashed lines

Fig. 4: a) Stratum 13 Casemates 513, 589 and 676 of W800, looking east; note how Room 676 is cut by 4RH of Stratum 10; b) Rooms 513, 589 and 676, looking south; c) assemblage of complete vessels from Room 676

Fig. 5: Main ceramic types of Stratum 13: 1, 3, s-8, 10-12) bowls; 2, 4, 9) Phoenician Bichrome bowls; 13-14) Cypriot BaR bowls (for parallels, see Fig. Sz in Supplementary Material 3)

Fig. 6: Main ceramic types of Stratum 13 (continued): 1-3) kraters; 4-8) cooking pots (for parallels, see Fig. S3 in Supplementary Material 3)

Fig. 7: Main ceramic types of Stratum 13 (continued): 1-2) cooking jugs; 3) 'Black juglet'; 4) dipper juglet; 5) Phoenician Bichrome juglet; 6-7) jugs; 8) Cypriot BaR jug; 9-10) chalices (for parallels, see Fig. S4 in Supplementary Material 3)

Fig. 8: Main ceramic types of Stratum 13 (continued): jars (for parallels, see Fig. Ss in Supplementary Material 3)

A major obstacle in the regional analysis of the Tel Shiqmona ceramics is the current near total lack of well-defined Iron IIA and IIB assemblages in the ꜤAkko Plain (cf. Arie 2020), including from Tel ꜤAkko itself. Somewhat restricted exceptions are Tell Keisan Strata 8–6 for the Iron IIA and the cremation burial near Tel Bira for the Iron IIB (Alexandre and Stern 2001).

Since the study of most material aspects of Tel Shiqmona is still underway, we refer only occasionally to data stemming from them. We also defer the full analysis of the purple-dye industry itself to a subsequent publication.

Finally, because of Tel Shiqmona’s location in a territory that during the 9th–8th centuries BCE may be defined historically as an interface between southern Phoenician entities and the Kingdom of Israel (e.g., Lemaire 1991: 152; Katzenstein 1973: 105), we also touch preliminarily upon geopolitical issues.

The early Iron Age at Tel Shiqmona: Previous studies

In his popular book, Elgavish presents his most up-to-date view of the stratigraphic sequence he excavated, which includes six Iron Age strata (1994: 34; here Table 1). Elgavish did not provide explanations for his stratigraphic sequence and dates (see details in Supplementary Material 2), nor did he provide architectural plans.

Table 1: The Iron I–IIB stratigraphic/chronological sequence at Tel Shiqmonaⁱ

i In the current paper we focus on our Strata 13–11.

Tel Shiqmona’s casemate enclosure, defined by him as Stratum 12 (called City A in Elgavish 1993), was the earliest occupation he assigned to the Iron II and dated to the 10th century BCE. Because of the limited data published regarding the Iron Age strata, very few other scholars refer to Tel Shiqmona during the Iron IIA–B. They usually accept Elgavish’s stratigraphic and chronological determinations and mostly discuss this ‘town’ or even ‘city’ in the context of trends in Israelite urbanism (e.g., Mazar 1990: 387, 389; Faust 2002a: 304, 312–313; 2003: 131, Fig. 6; 2012: 63–65; Faust and Bunimovitz 2003: 27) or olive-oil industries (Faust 2011a: 66, 69; Katz 2008: 8; Sader 2019: 293–295, 299). Others discussed the site’s marine aspects (Haggi 2006: 55; Galili 2009: 8–9; see further in Shalvi and Gilboa 2022a: 257–258). Arie (2020) was the only scholar to address Elgavish’s chronology critically—based on the few ceramic assemblages illustrated in Elgavish 1994 and on modern, quite consensual, relative ceramic chronology. He assigned Strata 14–12 to the Late Iron IIA and Elgavish’s Strata 11–10 (our Strata 10–11) to the Iron IIB. As we show below, his suggestions were quite accurate.

Prior to the establishment of the casemate enclosure, Tel Shiqmona was occupied by a series of villages, not discussed here (our Strata 15b, 15a and 14; Elgavish’s Strata 15–13; see Table 1). We date them to the late Iron I–Early Iron IIA (Ir1b–Ir1ǀ2). All produced some evidence of purple-dye production.

The casemate enclosure (Iron IIA–B)

Stratum 13, Late Iron IIA: Construction of the casemate enclosure

Architecture and industries

After the occupation of the Stratum 14 village ended (in abandonment or destruction), a casemate enclosure—of which Wall 800 in the south and Wall 801 in the east are known—was built over its remains (Fig. 3). The maximal extent of the enclosure was ca. 2,500 m². No architecture has been uncovered within it, but extensive in situ assemblages of artefacts were uncovered on its floors and in the casemates (Fig. 4). These include ceramics, such as jar concentrations and Cypriot BoR vessels, as well as ground-stone tools, loom weights, spindle whorls and several fragments of clay vats stained with purple dye, indicating production (see details in Supplementary Material 2). Phase 13 ended in a fiery destruction.

The ceramic assemblage

Aspects of relative and absolute chronology

Parallels to the main ceramic types of Stratum 13 can be cited mainly from Late Iron IIA strata in the Southern Levant (but also in Lebanon) (see Table 2 and Supplementary Material 3, Figs. S2–S5). These are primarily (from north to south): Sarepta Area II/ Trench Y/D2–D1, Tyre IX–VIII (some are later), Tel Hazor X–IX, Ḥorvat Rosh Zayit IIa–b, Tell Keisan 7, Tell Abu Hawam III, Tel YoqneꜤam XIV, the Ir2a horizon at Tel Dor (G/6a, D2/ 8b, B/8), Tel Megiddo VA–IVB, Levels Q-5–Q-4, H-7–H-5, K-3–K-2 and Tel Reḥov V–IV (cf. chronological discussions in Herzog and Singer-Avitz 2006; Arie 2013: 741–743; Gilboa 2018: 168; Panitz-Cohen 2020: 259–263).

Table 2: Relative chronology of the main sites considered

The Late Iron IIA date is also supported by the first and very extensive appearance of Cypriot BoR. This phenomenon is a significant characteristic of the Late Iron IIA in the Southern Levant (Gilboa and Sharon 2003: 62–64; Kleiman et al. 2019: Fig. 8; Waiman-Barak, Georgiadou and Gilboa 2021).

The Late Iron IIA ceramic horizon generally falls in the 9th century BCE and ends during its last third (Supplementary Material 1, Table S1). According to the analysis of Stratum 12 (see below), it seems that Stratum 13 had been destroyed before the end of the 9th century BCE.

Lastly, we note that of all the sites we examined, the site that best correlates in terms of its ceramics with Tel Shiqmona Stratum 13 is Ḥorvat Rosh Zayit, in Strata IIb–a, especially the destruction contexts. However, while the Ḥorvat Rosh Zayit fort ceased to exist then, the architecture and activity in Tel Shiqmona’s casemate enclosure were rapidly restored.

Regional aspects

Most of the bowls, both plain (Fig. 5:1,3,5) and decorated (Fig. 5:2,4,9) have significant parallels in the ꜤAkko Plain and on the southern Lebanese coast (see Supplementary Material 3, Figs. S2–S5). In contrast, there are other unadorned bowls (Fig. 5:6–8) with identical parallels mainly in inland sites of the Kingdom of Israel and there are others (Fig. 5:10–12) with parallels almost exclusively in such sites, such as Ḥorvat Rosh Zayit, Tel Hazor, Tel YoqneꜤam, Tel Megiddo, Tel Beth-Shean and Tel Reḥov. While simple kraters (Fig. 6:1) are common throughout the Southern Levant, red-slipped kraters (Fig. 6:2) are common exclusively inland. In contrast to the clear regional characteristics of the bowls, the situation regarding the cooking pots (Fig. 6:4–8) is more ambiguous. Generally, they correspond to cooking pots found in the vicinity of Tel Shiqmona, on the Carmel coast and in the ꜤAkko Plain, as well as in the Israelite inland sites mentioned above, and they are markedly different from those uncovered in south Lebanon. Black Juglets (Fig. 7:3), frequent at Tel Shiqmona, are typical—in varying quantities—of inland Israelite sites (and farther south mainly in Judah), and they are nearly absent, for example, in the extensive Ir2a (= Late Iron IIA) assemblages of nearby Phoenician Dor (but note one at Tell Abu Hawam). Phoenician Bichrome jugs are represented in the assemblage both by complete vessel (Fig. 7:5) and by sherds, but they appear in much smaller quantities than at Dor. Despite being a distinctly Phoenician characteristic, these jugs are attested throughout the Southern Levant in this period, but beyond the coast usually as solitary examples.

The most common (oval) jars in Stratum 13 (Fig. 8:3–8) represent a distinct category, probably locally produced, as indicated by petrography. Beyond a very general resemblance to oval jars known only from inland sites of the Kingdom of Israel, they do not cluster with any jars known elsewhere.

The elongated jars with a short rim, represented by a few examples (Fig. 8:1–2), are best attested at Israelite Ḥorvat Rosh Zayit (see Supplementary Material 3, Fig. S5). The identification of this jar type when vessels are fragmentary is difficult; nevertheless, it seems that such jars (Lehmann et al., in press: TJ 1) are common mainly in coastal Late Iron IIA sites on the Carmel coast, in the ꜤAkko Plain and in south Lebanon.⁷

Most of the open vessels (and others) in Stratum 13 were left without any special surface treatment, similarly to Phoenician coastal sites (Gilboa 2018: 161). In contrast, in the above-mentioned Israelite inland sites, red slip and/or burnishing of open vessels constitute the ‘most obvious separating factor between the pottery of the Iron I and that of the Iron IIA’ (Arie 2013: 730; and cf. Faust 2002b for red slip and burnishing as a component of emic identity in Israelite society).

Lastly, the phenomenon of large quantities of Cypriot BoR during the Iron IIA, mostly open shapes (e.g., Figs. 5:13–14, 7:8) can also be viewed from both wide and focused regional perspectives. As noted by Schreiber (2003: 28–34), such assemblages during the Iron IIA typify only the Southern Levant. In fact, large BoR assemblages characterize primarily centres of the Kingdom of Israel (including the estate at Ḥorvat Rosh Zayit; Gal and Alexandre 2000: 68), but also at Tell Abu Hawam (Herrera and Gomez 2004: Pl. 1–7). However, among the rich assemblage of nearby Phoenician Dor during the Late Iron IIA (Ir12a), specifically BoR vessels are a minority among the many Cypriot imports (Waiman-Barak, Georgiadou and Gilboa 2021: Fig. 2).

Stratum 12, Final Iron IIA:

Reconstruction of the casemate enclosure

Architecture and industries

After the destruction of Stratum 13, new casemate walls were built on the outline of W800 and W801, slightly deviating in orientation, at least partially constructed as an offset-inset wall (Fig. 9). The space inside the enclosure was now densely built, with straight alleys running between series of buildings with various stone installations, including—on the east—two pillared ‘three room-houses’ (for details, see Supplementary Material 2).

Fig. 9: Plan of Stratum 12, integrating Bar’s excavation (grey shaded area); the location of the Stratum 10 4RH is marked by dashed lines

As in Stratum 13, several purple dye-stained industrial vat fragments were found in this stratum, mainly in Bar’s excavations (e.g., Fig. 2). These indicate the continuation of the industry.

Stratum 12 underwent gradual architecture development into what we term Stratum 11.

The ceramic assemblage

Because of the gradual change that Stratum 12 underwent, only a few complete vessels have been preserved. The pottery sherds we present are from deposits above well-defined floors located under clear surfaces of Stratum 11. The best representative context comes from the pavement of Building 283, sandwiched between two well-identified surfaces of Strata 13 and 11. More pottery came from surfaces of Stratum 12, for example in the ‘Three-Room House’ R2261 and in Room R2287.

Aspects of relative and absolute chronology

The architectural and ceramic continuity between Strata 13 and 12 indicates that they were close in time. Many of the types in Stratum 12 already occurred in Stratum 13 (Late Iron IIA), such as the simple bowls (Fig. 10:1–6), kraters with rounded carination, usually with a slightly thickened rim (Fig. 10:11–12), triangular-rim cooking pots (Fig. 11:1–2), cooking pots with thick, horizontal external ridges under their rim (Fig. 11:3) and cooking jugs with vertical rim (Fig. 11:5). A prominent Stratum 12 feature, originating in Stratum 13 and distinguishing it from Stratum 11, are the oval jars with high neck and thickened rim (Fig. 11:10–13; cf. Fig. 8:3–8). This is their final occurrence. On the other hand, several vessels and ceramic phenomena already characterize the Iron IIB in both inland and coastal sites and are more typical of the subsequent Stratum 11. These include the first appearance of Phoenician-style bowls with a pronounced horizontal ledge rim (Fig. 10:7), simple ledged-rim bowls (Fig. 10:9), goblet-like bowls (Fig. 10:10), rounded kraters with a horizontal oblong rim (Fig. 10:13), cooking pots with a bulbous grooved rim (Fig. 11:4) and an increasing amount of red-slipped and well-burnished fine wares (the so-called ‘Samaria Ware’; not illustrated). On the other hand, several Iron IIB ceramic markers, which are typical of Stratum 11, are missing, a notable example being the absence of the Phoenician elongated jars (TJ2; see below in Stratum 11). Compared to Stratum 13, there is a drastic decline in Cypriot BoR vessels.

Fig. 10: Main ceramic types of Stratum 12: 1-10) bowls; 11-14) kraters (for parallels, see Fig. 56 in Supplementary Material 3)

Fig. 11: Main ceramic types of Stratum 12 (continued): 1-4) cooking pots; s) cooking jug; 6-7) juglets; 8) Phoenician Bichrome jug; 9-15) jars (for parallels, see Fig. S7 in Supplementary Material 3)

This phenomenon—namely, an occupation combining ceramic forms typical of both the Late Iron IIA and IIB—has been identified at other sites in the Southern Levant. A nearby example is Tel YoqneꜤam Stratum XIII. Like at Tel Shiqmona, along with Late Iron IIA types, it is characterized by ledged-rim bowls (Zarzecki-Peleg, Cohen-Anidjar and Ben-Tor 2005: Type B II; 255) and cooking pots with a bulbous grooved rim (ibid.: Fig. II.22:1–2; Types CP VIA–B; 279). Cypriot BoR here too decline in quantities (ibid.: 251).⁸ At Hazor, the combination of Iron IIA and IIB pottery types occurs in Stratum VIII (e.g., Zarzecki-Peleg 2005b: 372; Sergi and Kleiman 2018: 4, Table 2; Shochat and Gilboa 2019: 379–380), which is also typified by an increase in fine wares (e.g., Ben-Ami 2012: Fig. 3.9:12–13). We assign this ceramic phase to the decades flanking the year 800 BCE (Supplementary Material 1, Table S1).

The ceramic types of Stratum 12 are discussed above (Late Iron IIA) and below (Iron IIB).

Stratum 11, Iron IIB: The heyday of Tel Shiqmona

Architecture and industries

Stratum 11 (Figs. 12–13) amalgamates several changes through time to the Stratum 12 architecture. Some Stratum 12 buildings remained practically unchanged, while others were rebuilt, including new stone installations, tabuns and at least one oil press (Fig. 14a; for details, see Supplementary Material 2). Notably, construction now extended over some of the casemates to the east and west, annulling their function as fortifications.

Fig. 12: Plan of Stratum 11, integrating Bar’s excavations in the grey shaded area (Bar 2021: Fig. 6); the Stratum 13 casemate wall is marked by dotted lines and the Stratum 10 4RH by dashed lines

Fig. 13: Aerial view of Stratum 11; top right: Stratum 10 4RH; looking east

Fig. 14: Building 283, Stratum 11: a) olive-oil press; b) assemblage of complete vessels

Fig. 15: Main ceramic types of Stratum 11: 1-5, 7-21) bowls; 6) Phoenician Bichrome bowl; 22) mortarium (for parallels, see Fig. 58 in Supplementary Material 3)

Fig. 16: Main ceramic types of Stratum 11 (continued): 1-3) kraters; 4-11) cooking pots; 12) cooking jug;13) decanter;14-16) juglets;17) jug (for parallels,see Fig. S9 in Supplementary Material 3)

Stratum 11 ended in total destruction, leaving large assemblages of artefacts on its floors, including pottery (e.g., Fig 14b), several fragments of vats and other stone and pottery items stained with purple dye (Fig. 2; see below, Fig. 17:11–13), loom weights, spindle whorls and more.

Fig. 17: Main ceramic types of Stratum 11 (continued): 1-3, 5-10) jars; 4) Phoenician Bichrome jar; 11-13) vats (for parallels, see Fig. S10 in Supplementary Material 3)

The ceramic assemblage

Aspects of relative and absolute chronology

Stratum 11 corresponds to substantial Iron IIB strata in the Kingdom of Israel, as can be seen in Table 2 and in Supplementary Material 3, Figs. S8–S10. These include mainly Tel Hazor VII–V, Tel YoqneꜤam XII, Tel Megiddo IVA, Tel Beth-Shean P-8–P-7 and Tel Reḥov III–II, as well as most of the buildings in the village of Ḥorvat Rosh Zayit Areas A–C, Sarepta C2, Tell el-Burak Phase E and al-Mina Stratum 8.⁹ All the above-mentioned Iron IIB occupations in Israel are defined as terminating in Assyrian destructions in the 730s BCE (e.g., Singer-Avitz 2014; Ben-Tor and Zarzecki-Peleg 2015: 135) and we concur (Shalvi and Gilboa 2022a: 265–269; 2022b).

Dating the beginning of Stratum 11 depends to a large extent upon our understanding of the previous strata. If indeed, as argued above, Stratum 13 was destroyed at the end of the Iron IIA, in the last third of the 9th century BCE and Stratum 12 dates around the turn of the 9th century BCE, this would mean that Stratum 11 begins around the early8th century BCE.

Dating the end of Stratum 11 is somewhat more straightforward. Beyond the similarity to sites in Israel ending in late 8th-century BCE destructions, the prevalent Phoenician transport jars in Stratum 11 (TJ2) indicate that it could not have continued after the conquests of Tiglath-Pileser III (as shown in Shalvi and Gilboa 2022a). Furthermore, since the subsequent Stratum 10—the large 4RH with olive press—embodies a phenomenon familiar in the Kingdom of Israel only before the Assyrian conquest of the 730s BCE but never later, we have proposed dating the end of Stratum 11 somewhat before the conquests, ca. 740 BCE (discussed in ibid.). Stratum 11 is therefore dated between ca. 790–740 BCE.

Regional aspects

As can be seen in Supplementary Material 3 (Figs. S8–S10), the Stratum 11 bowl assemblage shows clear polarity. On the one hand, the many bowls with various pronounced ledge rims (Fig. 15:1–4) and, for example, the quantity and variety of fine-ware bowls (Fig. 15:5,7–13) cluster with south Lebanese coastal sites. On the other hand, carinated bowls with simple rim (Fig. 15:15–17), deep bowls with profiled rim (Fig. 15:18–19), rounded kraters with inverted upper body and thickened rim (Fig. 16:1–2), bulbous grooved-rim cooking pots (Fig. 16:9–11), ridged-neck jars (Fig. 17:8–9) and holemouth jars (Fig. 17:10) are all types that group with inland sites of the Kingdom of Israel.¹⁰ Note, however, that the Tel Shiqmona ceramic assemblage lacks several bowl types that are very common in the Israelite sphere—for example, flat bowls with straight walls (Ben-Tor and Zarzecki-Peleg 2015: Pl. 2.2.2:26,38–40) and deep rounded ones with an external ridge under the rim (Ben-Tor and Zarzecki-Peleg 2015: Pl. 2.2.1:31–32).

The jars assemblage in Stratum 11 changes drastically compared to the previous strata. The numerous local oval jars vanish entirely and the elongated south Lebanese ‘Phoenician’ transport jars with tall ridged rim (TJ2 ‘Torpedo’ jars; Fig. 17:1–3) become dominant by far. Also frequent are large oval jars with an inclined ridged neck and a thick everted rim (Fig. 17:8–9), as well as many holemouth jars (Fig. 17:10), which clearly cluster with sites in the territory of the Kingdom of Israel.

With regard to other imports, the number of Cypriot fragments diminishes significantly in comparison to Strata 13–12, like at many other 8th-century BCE sites in the Southern Levant (Gilboa 2015: 488–489; Waiman-Barak, Georgiadou and Gilboa 2021: 255). Nevertheless, 8th-century Tel Shiqmona did produce a relatively large assemblage of Cypriot pottery, mainly closed shapes (to be published separately).

Discussion: The casemate enclosure from emergence to termination

Stratum 13

At a certain point in the first half of the 9th century BCE a casemate enclosure was constructed over and through the Stratum 14 village, establishing a small fortified compound, probably no more than 0.2 hectares in extent. Whether the construction of the casemates terminated the village or the latter had been abandoned earlier is currently unknown. Even though it was devoid of inner architecture, intensive and varied activity took place here. The fortifications, then, were intended first and foremost to protect the activity rather than intra muros people or institutions. In addition to the purple-related industries, Tel Shiqmona Stratum 13 produced one of the richest assemblages of Cypriot BoR vessels beyond the borders of Cyprus in this period, perhaps pointing to one of the destinations of the purple products.

The construction of the fortified enclosure and the expropriation of the luxury industries most probably indicate a takeover by a micro- or macro-regional power that controlled (and taxed) the compound. Possibilities are Dor, Megiddo, or ꜤAkko, or, alternatively, Tyre or Samaria. The identification of this power and the characterization of the people who operated in Stratum 13 are, of course, among the most difficult tasks in any attempt to understand this occupation.

During the Late Iron IIA (the 9th century BCE) the Kingdom of Israel exhibits the first significant urban phase, manifest, for example, in monumental architecture. This period overlaps, at least in part, with the Omride dynasty, which became a significant political and military force in the region (e.g., Finkelstein 2000; 2011; Grabbe 2012; Na’aman 2007; Sergi 2016). This dynasty established the capital in Samaria and constructed fortified administrative centres such as Jezreel, Megiddo Stratum VA–IVB and YoqneꜤam Stratum XIV. Historical and biblical sources—the Mesha and Tel Dan stelae (cf. 2 Kings 3:4)—indicate extensive territorial expansion (see also Finkelstein 2011: 235–239). Close to Tel Shiqmona, the material evidence from Tel Dor suggests that during the Late Iron IIA (Ir2a), this Phoenician town also came under the control of the Kingdom of Israel, either by force or through political/ economic treaties. This event has been attributed to the Omride era (Gilboa, Sharon and Bloch-Smith 2015: 62–66), but could be slightly later (Shochat 2018; for a peaceful transfer of the town, see also Na’aman 2016). During this period, Israel paid tribute to the Assyrians, at least under Shalmaneser III (Hasegawa 2012: 55, with sources).

Meanwhile, in Tyre, most of the 9th century BCE was a period of prosperity and possibly expansion (all according to Josephus Flavius, quoting Menander; see Sader 2019: 77, 126). Ithobaal I’s murder of his predecessor, Phelles—who was king for about one year—ended a period of dynastic unrest and ushered in a period of stability during his reign and the reigns of his son and grandson, BaꜤlꜤezer and Mathan I, all three reigning between 888–821 BCE (Katzenstein 1973: 166–167). During this time Tyre may have governed Sidon (or vice versa, e.g., Khreich 2020: 12–13). Lehmann (2021: 302, 304) argued that under Ithobaal I, around the mid-9th century BCE, Tyre expanded to the ꜤAkko Plain. Throughout most of this period the Tyrians too paid tribute to the Assyrians—Ashurnasirpal II and then Shalmaneser III (Katzenstein 1973: 139–142, 162–185; Fales 2017: 228; Elayi 2018: 62, 122, 165, 131–133, 135).

Does the material culture of Stratum 13 provide clues regarding the geopolitical and/or cultural affiliation of the operators of the fortress? We first discuss architecture. Scholarship has most frequently associated casemates in the Levant with the Kingdoms of Israel and Judah, either with their administrative centres (‘cities’) or with strategically situated defensive fortresses (e.g., Shiloh 1987; Herzog 1987; Cohen and Bernick-Greenberg 2007; Ben-Ami 2009; Finkelstein 2014: 217; Arie 2020, and see more below; see Vico 2020a for a definition of constructional/functional subtypes in casemate constructions).

In the Kingdom of Israel—according to the absolute chronology we advocate—the first to adopt this fortification method were the Omrides (Finkelstein 2000; Sergi and Gadot 2017). The best examples are at Tel Hazor Strata X–IX, Tel YoqneꜤam Stratum XIV and probably Jezreel, though at the capital Samaria itself the casemates may be later (Franklin 2004).¹¹

Casemate enclosures, however, were more widespread in the Southern Levant regionally and temporally, even before they became common in Israel and Judah, for example in Galilee (Katz 2021), Transjordan (Routledge 2000; Swinnen 2009; Emswiler 2020) and the Negev Desert (Cohen and Cohen-Amin 2004) during the Iron IB–Early IIA. Most of the Negev ‘fortresses’—which are not in fact fortresses—date to the Early Iron IIA (ibid.; Herzog and Singer-Avitz 2004, see Table 1).¹²

No casemate enclosures are known from Lebanon during the Iron IIA, but several occurrences in the Phoenician West are attested as early as the second half of the 8th century. The best examples are at Carthage (Docter et al. 2003; 2006) and sites in the Iberian Peninsula, for example El Cabezo Pequeño del Estaño and Tavira (summary in Vico 2020b: 367–370 and Table 8 on p. 1034). These, alongside the late Iron Age fortress at Kabri in the ꜤAkko Plain (Lehmann 2002), suggest that this practice may have existed in the Tyrian–Sidonian sphere earlier too.¹³

The casemate enclosure of Tel Shiqmona Stratum 13 neither protects a town/‘city’ nor is it located at a dominant controlling high point, and it housed no buildings. The archaeological evidence from Stratum 13, the longevity of the purple-dye production (and dyeing) at the site and the favourable marine environment for collecting and possibly cultivating the shells suggest that the primary motivation was the protection of the profitable industries. It may not be the only fortified industrial Iron Age site per se. At Carthage too, casemates housed metallurgical activity (Docter et al. 2006: 39–41), and the fortified enclosure at Khirbet el-Mudēyine ath-Thamad, central Jordan, has been argued to have been redesigned as an industrial site specialising in textiles, at least in the late Iron Age (Daviau 2006: 28).¹⁴

With regard to pottery, as shown above, the ceramic assemblage exhibits a unique equivocal geographical affiliation, with some types clustering with ‘Phoenician’ sites along the coast in northern Israel and south Lebanon, including nearby Dor, and others with inland Israelite sites. This holds true both for ordinary vessels and for others. Although media other than architecture and pottery are beyond the scope of this article, we note, for example, Arie and Bar’s (2021) suggestion that a stone altar (out of context) may reflect Israelite cultic traditions.

Based on all the above, the most plausible scenario would be to attribute the takeover of Tel Shiqmona and the establishment of the fortified industrial complex to the economic interests of the Kingdom of Israel with its consolidation and expansion under the House of Omri. In this regard, the absence of ‘hippo’ and holemouth jars should be explained by the specific role of the site (and the agricultural functions it did not fulfill; cf. Butcher et al. 2022).

Nevertheless, highland Samarian dynasties did not possess the necessary knowhow to run coastal industries. Knowledge regarding the procurement (and cultivation?) of the snails, the production of dye and the dyeing of fibres with it, and, generally speaking, engagement with the sea and coast certainly lay with locals of the Carmel coast (‘Phoenicians’), who operated the industry, by coercion or cooperation (for coastal vs. inland habitus along the Carmel coast in the Iron IIA, see Gilboa, Sharon and Bloch-Smith 2015: 71–72; Gilboa 2022a: 44; 2022b; see also Faust’s ‘cultural attitudes’, 2011b: 121). We suggest that this situation prevailed until the destruction of the enclosure.

This complex situation may offer an explanation for the unique mixture in the site’s material culture; just what that explanation is, however, is not yet entirely clear.¹⁵

With regard to maritime and possibly terrestrial exchange networks, intensive contacts during this period are at present attested only with Cyprus, most probably with specific centres on the island’s east and south, but there is no unequivocal evidence of trade with the Lebanese coast. It is also quite safe to assume that Tel Shiqmona’s purple-dyed fibres (and perhaps textiles) were continuously distributed to elite institutions and individuals in Israel and Judah and perhaps elsewhere, but this, of course, is nearly impossible to substantiate.

Finally, we zoom in on Israel’s takeover of the Carmel coast. As previously mentioned, Dor’s excavators date the transition from a ‘Phoenician’ town to an Israelite administrative centre to the Late Iron IIA (Ir2a in Dor terminology). Shochat (2018: 125–127) argues for some Iron IIB pottery types in the final phase of the Phoenician sequence and for a somewhat later date for the transition, during a final Late Iron IIA stage, Ir2a late in the terminology of the Dor excavators. In any event, it seems that Tel Shiqmona was the first to be brought into the auspices of Israelite regulation, and only later was the takeover of the Carmel coast completed with the annexation of Dor and the deepening of Israelite involvement at Tel Shiqmona (below). We suggest that the fort at Khirbet es-Suwweida on the Har Ḥorshan ridge of the Carmel, a site that controls a large area of the southern Carmel ridge and coast, was established as part of the Israelite expansionist aspiration, before the annexation of Dor.¹⁶

The violent destruction and utter annihilation of Stratum 13 could indicate punitive activity and deterrence with no apparent interest in preserving the site’s industries. The Stratum 13 pottery assemblage corresponds with some occupations in the Southern Levant, the destructions of which have been attributed to Hazael’s campaign(s) (Finkelstein 2016; Kleiman 2016: Table I). However, there are other potential causes for destruction in the late 9th century BCE, possibly related to the military coup and replacement of the ruling dynasty (from Omride to Nimshide; cf. Ussishkin and Woodhead 1997: 70).

Stratum 12

Shortly after the destruction of Stratum 13, around the end of the 9th century BCE, the fortification was re-established, now enclosing a well-planned layout of buildings and alleys. The organised manner of construction indicates, to our minds, an attempt to develop and stabilise the main industry on the tell—purple-dye manufacturing and dyeing.

Since most of the Stratum 12 buildings continue in Stratum 11—their final use phase— we are most familiar with them (and their functions) in this latter stratum and discuss them below. Regarding pottery, the same mixture of traditions evident in Stratum 13 largely continues into Stratum 12.

In calendric/historical terms, the reconstruction of Stratum 12 must belong somewhere in the early Nimshide era—the last part of Jehu’s reign, Jehoahaz’s or Joash’s reigns. Historically, we suggest that the days of Joash (805–790 BCE; Galil 1996: Tables 14–16) are the most probable. This was a period of strength and prosperity in Israel (Hasegawa 2012: Chapter 5 and summary on p. 122; Na’aman 2019: 16–18). This king’s supremacy is expressed in the Bible, for example, in the account of his victories over Ben Hadad son of Hazael (2 Kings 13:25) and the fable of the thistle and the cedar attributed to him (2 Kings 14:9–10). In contrast, it seems that in this period, the ꜤAkko Plain sites did not recover from the devastation they suffered around the end of the Iron IIA, which the Aramaeans may have caused (except perhaps ꜤAkko, of which very little is known, cf. especially Arie 2020, with references). The Stratum 12 buildings gradually developed to become Stratum 11.

Stratum 11

This stratum, which is dated on ceramic grounds to ca. 790–740 BCE, represents the Golden Age of the Iron Age tell, in which both purple dye and olive oil were extensively produced. Some of the new buildings extend over and beyond the casemates. Beyond the need for more space, this suggests that their protection was not required anymore.

Apart from the casemates, the best definable structures in Stratum 11 are the two three-room houses first built in Stratum 12. Despite the common association of such houses with Israel and Judah, they are spatially much more widespread, and in the Iron Age their ‘ethnic’/geopolitical association not straightforward (see recent discussion of the 4RH in Shalvi and Gilboa 2022a: 270–271, with extensive references). Still, such pillared structures, in the Iron IIB, are indeed prevalent in the Kingdom of Israel and Judah (Faust and Bunimovitz 2003: 29–30 with many examples in Faust 2012) and are so far unknown on the Lebanese coast.

Stratum 11 is the first to produce evidence for olive-oil production. Weaving continues, and direct and indirect evidence for the purple-dye industry has been found in and near almost all the Stratum 11 buildings excavated. These indicate that the production of the dye remained the main function of the site.

In Stratum 11, short-distance (and thus, most probably, direct) maritime ties with south Lebanon start to be conspicuous. This is reflected in the very large number of south Lebanese transport jars that seem to have served for any function requiring large containers¹⁷ and in the many fine-ware vessels that originated there (evidenced by yet unpublished petrographic studies). Among other commercial functions, the jars may have carried oil to south Lebanon.

The decline in Cypriot imports, as already hinted above, does not necessarily reflect a decline in the intensity of exchanges, but may suggest, for example, changes in production and distribution of BoR ware in Cyprus (see Georgiadou 2016: 95, 100). The exact definition of the provenance of the Cypriot vessels is bound to answer this question.

As we argued for Stratum 13, products of the purple-dye industry were most probably distributed to clientele in Cyprus, the Southern Levant, and at least from Stratum 11 to south Lebanon too, in light of the intensity of traffic.

As in Strata 13 and 12, the ceramic assemblages remain ‘mixed’, but the Phoenician component becomes more dominant.

The archaeological date of Stratum 11 largely coincides with Jeroboam II’s reign (ca. 790–750 BCE), the last peaceful and prosperous period in the Kingdom of Israel prior to the Assyrian conquest (Hasegawa 2012; Chapter 6, p. 147; Knauf 2017). This king has been claimed to have taken over territories in the ꜤAkko Plain (Finkelstein 2017: 284; see, in the same vein, Sergi 2017: 339; Arie 2020). Compromising the fortification of such an important asset when more space was needed would fit such a scenario, in which, among other factors, the kingdom’s borders moved farther north.

The destruction of Stratum 11 terminated the Tel Shiqmona casemate enclosure, never to be rebuilt. This, we assert, happened before the Assyrian conquests of 734–732 BCE— at ca. 740 BCE. Thus, we suggest relating this event to the days of unrest, rebellion and wars of succession after Jeroboam II, between 750 BCE and Tiglath-Pileser’s campaigns, which saw the replacement of six kings in less than 15 years, most of them assassinated (2 Kings 15–17).

Summary

During the Late Iron IIA, a fortified complex was constructed at Tel Shiqmona for the production of dye. After its destruction, the compound was restored and reached its peak during the Iron IIB. It was finally destroyed toward the end of this period.

The combination of the archaeological parameters, also viewed through an historical lens, indicates an Israelite takeover of the production site and its fortification, probably by Ahab. However, there are reasons to assume that production was carried out by local ‘Phoenicians’. After the destruction at the end of the 9th century, the site was restored by Joash and reached its peak under the rule of Jeroboam II. At this stage too, local ‘Phoenicians’ likely produced the purple dye. Subsequently, the industry was resumed under Assyrian hegemony.

Research into Tel Shiqmona’s Iron Age is in its infancy, and we have barely scratched the surface. A plethora of questions remains regarding the way(s) the site functioned, and more. In this article we have brought to light a historical ‘blind spot’ regarding the Israelite–Phoenician frontier and the sources of wealth of the Kingdom of Israel, and we hope to have paved the way to salvaging this unparalleled site from obscurity.

Contributors

Golan Shalvi: The Zinman Institute of Archaeology, University of Haifa, and the Department of Bible, Archaeology and Ancient Near Eastern Studies, Ben-Gurion University of the Negev; ORCID: https://orcid.org/0000-0003-1071-7825; corresponding author’s email: [email protected]

Ayelet Gilboa: The Zinman Institute of Archaeology and School of Archaeology and Maritime Cultures, University of Haifa; email: [email protected]

Acknowledgements

The Tel Shiqmona project is conducted with the significant help of our dedicated volunteers: Sandy Katz, Moshe Diengott and Edna Avrahami, the manager of the Tel Shiqmona laboratory at the University of Haifa, Sonia Pinsky, and the past and present staff of the National Maritime Museum in Haifa, especially Oren Cohen, Adi Shelach, Filip Vukosavović and Ron Hillel. We are deeply grateful to the Israel Antiquities Authority for granting us the publication rights for the Late Bronze and Iron Age levels of Elgavish’s excavations at Tel Shiqmona. Unconditional help and support was received from Israel Antiquities Authority staff, especially Gideon Avni, Avshalom Karasik, Zvi Greenhut, Debi Ben-Ami and Rebecca Cohen-Amin. Part of the pottery had been drawn by unknown artists in the 1960s and 1970s, and part of it was 3D-scanned in the Computational Archaeology Laboratory of the Israel Antiquities Authority by Argita Gyerman Levanon and in the Institute of Archaeology of the Hebrew University of Jerusalem by Ortal Harush and Tamara Mkheidze. We thank Sapir Haad for processing the figures and plans, Sveta Matskevich for her work on the map and Sonia Pinsky for producing the pottery plates. Some of the figures originate from the photo albums of the Elgavish excavation, taken by an unknown photographer. We are grateful to Shay Bar for allowing us to incorporate an unpublished plan. The purple-dye residues were analysed by Naama Sukenik, David Iluz, Alexander Verwak and Zohar Amar from the life sciences laboratories at Bar-Ilan University, Ramat Gan. We thank Maria Eugenia Aubet and Mariano Torres Ortiz for discussing fortifications in the Phoenician West and Ilan Sharon for discussing the radiocarbon data of Ḥorvat Rosh Zayit. Petrographic analysis was recently conducted by Paula Waiman-Barak and is still mostly unpublished. Finally, we owe gratitude to two anonymous reviewers for their critiques and valuable suggestions.

The research and publication of Tel Shiqmona is part of a broader research program regarding southern Phoenicia, supported by an Israel Science Foundation grant awarded to Gunnar Lehmann and Ayelet Gilboa (Grant No. 596/18), conducted in cooperation with Paula Waiman-Barak, Harel Shochat and Revital Golding Meir. Research of Tel Shiqmona is funded by a Shelby White and Leon Levy Program for Archaeological Publication grant awarded to the authors; by a collaborative initiative of the Haifa Municipality, the Rector’s Office and the Research Authority at the University of Haifa; by the Hecht Foundation in Haifa; and by the Zinman Institute of Archaeology at the University of Haifa. A Ph.D. Rotenstreich Scholarship was awarded to Golan Shalvi. We gratefully acknowledge this support.

Disclosure statement

The authors report that there are no competing interests to declare.

Supplementary material

Supplemental data for this article can be accessed at https://doi.org/10.1080/03344355.2023.2190283

Additional information

Notes on contributors

Golan Shalvi

Golan Shalvi: The Zinman Institute of Archaeology, University of Haifa, and the Department of Bible, Archaeology and Ancient Near Eastern Studies, Ben-Gurion University of the Negev

Ayelet Gilboa

Ayelet Gilboa: The Zinman Institute of Archaeology and School of Archaeology and Maritime Cultures, University of Haifa.

Notes

1 Elgavish was aware of the evidence for purple-dye production, but attributed it to his Stratum 10 (our Stratum 11) only.

2 In a pioneering interdisciplinary study, Joseph Breman (1999) showed categorically that the Tel Shiqmona maritime interface was unsuitable for mooring and that access to the site from the sea would have been possible only during especially calm conditions and by small boats (cf. Galili, Arenson and Rosen 2021: 172).

3 Conducted by Naama Sukenik (Israel Antiquities Authority), David Iluz, Zohar Amar and Alexander Varvak (all from Bar-Ilan University).

4 We did not find organics fit for radiometric dating.

5 We consider Ḥorvat Rosh Zayit to be an Israelite centre, as argued convincingly in Kleiman 2017; cf. Mazar 2022: 208; Shalvi and Gilboa 2022a: 275.

6 Whether this site can indeed be defined as ‘Phoenician’ remains to be determined.

7 With few examples from inland sites. Nava Panitz-Cohen has discussed similar jars from Tel Reḥov and other inland sites and has argued for a coastal origin by visually examining their fabric (2020: 198; Type SJ56a), which seems very plausible; to our minds, however, the definition of this jar type at Tel Reḥov is too broad.

8 The few Phoenician elongated jars attributed to YoqneꜤam Stratum XIII (Zarzecki-Peleg 2005a: Fig. I.73: 21–23) originated from a pit of uncertain stratigraphic affiliation (L.1725). As clarified by Zarzecki-Peleg (ibid.: 171), with regard to L.1725 and L.1732: ‘No floors were detected in either of these loci. Their attribution to Stratum XIII, based upon their resemblance to other pits of that stratum, is uncertain’.

9 At Tyre, the range is wider (Strata VIII–III).

10 Regarding holemouth jars, in addition to the comparanda in Supplementary Material 3, see Butcher et al. 2022.

11 At Megiddo Stratum VA–IVB, the northern ‘casemates’ are associated only with Palace 6000 and do not comprise a civic defence system. The Iron IIA casemate fortification at Gezer may prove earlier, as suggested in Ortiz and Wolff 2021, based on radiometric dating (see Fig. 2), but no conclusion can be drawn until the exact contexts/deposition of the (few) samples are published.

12 To our minds, the geopolitical affiliation of the casemate wall (and associated structures) at Tel ꜤEn Gev on the eastern shore of the Sea of Galilee is not sufficiently well established yet, nor can the dates offered be critically assessed, since hardly any ceramics have yet been published by the various expeditions to the site. For various suggestions, see Sergi and Kleiman 2018 and Hasegawa 2019, both with references. We also maintain that there are still insufficient data to date the Iron Age casemate enclosures at ꜤEn Hazeva and Tell el-Kheleifeh in the Arabah accurately (but cf. Finkelstein 2014, with references).

13 However, we do not accept that the complex dubbed ‘casemates’ at Beirut (Badre 1997: Fig. 40a–b) merits this definition (similarly, Sader 2019: 164). In addition, the (mostly reconstructed) Iron IIB casemate wall at Tell Arqa (Thalmann 1998: 132) is questionable. Further, we do not agree that there is sufficient evidence to place the beginning of the Kabri fortress in the Iron IIA (earlier than claimed by the excavators), as suggested in Arie 2020. Forty years ago, Kochavi (1984) was the first to suggest that the casemate fortresses of Upper Galilee were, in fact, strongholds in Tyre’s hinterland, a suggestion that has not been widely accepted; but for the close coastal, possibly Tyrian, connections of the fortress at Har Adir, see Pagelson, Katz and Goren 2022.

14 Other industrial ‘metallurgical fortresses’ were protected by solid, rather than casemate, walls, e.g., Khirbat en-Naḥas in lower Edom (southern Jordan; Levy et al. 2014).

15 We, of course, do not mean that ‘local workers’ were eating from coastal bowls in their own little corner while ‘Israelite overseers’ cooked their meals in pots of inland tradition. The factors underlying the (fluctuating) ceramic combinations at Tel Shiqmona are much more complex, and as a first step the production locales of the main ceramic types should be determined. Rather than accepting approaches such as Kleiman’s (2021: 4), who defines any consideration of ceramics for aspects of geopolitics and identity as a deplorable ‘pots and people’ attitude—throwing, in our opinion, the baby out with the bathwater—one should adopt a more nuanced approach.

16 Indeed, Arie hinted that several ‘hippo jars’ found at the site suggest an Israelite affiliation, though in conclusion he refers to it as a Phoenician fort (Arie 2020: 7, 9). Ongoing excavation at this site, led by Meir Edrey, may permit a reconsideration of these suggestions.

17 This phenomenon continues at Tel Shiqmona to the end of the Iron Age (Shalvi and Gilboa 2022b).