2.1 Geography and Environment

The region of Edom is commonly identified with the highland region of southwestern Jordan, south of the Wadi al-Hasa (biblical Zered) and north of Ras en-Naqb that marks the southern delineation of the highlands. To the east lies the arid Syro-Arabian Desert, and to the south, the Hisma Desert, which extends through the Hejaz region of northwest Arabia. This highland region of Edom is framed along its western extent by a mountainous range, with the Wadi Arabah beyond. The term Edom (אדם) itself, meaning “red,” appears to be derived from the reddish color of the Nubian Sandstone massifs running north-south along the eastern side of the Wadi Arabah (Reference Bartlett and FreedmanBartlett 1992: 287; Reference DanielsonDanielson 2020a: 74–84). The term “Seir” (שעיר) also often appears in relation to Edom, at times appearing to designate a mountainous subset of the greater region of Edom (Genesis 14:6, 32:3, 36:8), and at other times functioning as a near synonym (Genesis 32:3; 36:8; Reference Knauf and FreedmanKnauf 1992a; Reference Edelman and EdelmanEdelman 1995: 7–11).

While these topographical features in effect served to delineate and constrain human movements, they by no means function as a strict “border” for the region as in fact there is significant evidence of material culture and elite Edomite activity beyond this region, most notably at Tall al-Khalayfi (formerly Tell el-Kheleifeh) at the southern end of the Wadi Arabah along the Red Sea, and in the northeastern Negev. In fact, while the label Edom can be understood as first a regional term, later co-opted as a political and gentilic designation, its geographic contours likely shifted at different times and included regions from the Negev to the west (Reference Edelman and EdelmanEdelman 1995: 7–11; Reference Ben-YosefBen-Yosef 2021: 161–166).

The core of the region of Edom on the highland plateau was located in a Mediterranean vegetation zone, characterized by a Mediterranean maquis forest featuring species of oak, olive, pine, and terebinth. Dwarf shrubs and other herbaceous flora are also common, with ideal rainfall patterns reaching 200 to 400 mm per year (Figure 2;Footnote ³ Reference CordovaCordova 2007: 90–94; Reference Langgut, Neumann and SteinLanggut et al. 2014, Reference Kurtz2015). While the Mediterranean vegetation zone at the heart of Edom was well suited for dry farming and horticulture, it was geographically limited. Surrounding it was the more arid Irano-Turanian zone, a drier landscape suited to more limited rainfed agriculture, but ideal for range pastoralism (Figure 3). Beyond, the remainder of southern Jordan was situated in the Saharo-Arabian desert zone, in which only a limited range-based pastoralism – dependent on a specialist knowledge of water sources – was possible (Reference CordovaCordova 2007: 95–104; Reference AbabsaAbabsa 2013: 64–90; Reference Kagan, Langgut, Boaretto, Neumann and SteinKagan et al. 2015).

Figure 2 Average rainfall patterns of Edom and the Negev

(map by author; data after Reference IsraelIsrael 1985: Map 12; Reference AbabsaAbabsa 2013: Fig. I.12)

Figure 3 Vegetation zones of the southern Levant

(map by author, after Reference Langgut, Finkelstein, Litt, Neumann and SternLanggut et al. 2015: Fig. 1)

The kingdom of Judah to the northwest broadly comprised the regions of the Shephelah, Judean Highlands, and northern Negev. To the east and southeast of the political and ideological center at Jerusalem, the arid Judean Desert presented a difficult and hostile landscape, contrasted with the more advantageous environmental situation of the Shephelah to its west and southwest. This more fertile region of Judah consistently received approximately 300–600 mm of rainfall per annum (Reference KurtzLanggut et al. 2015), with its resultant Mediterranean vegetation zone well suited for agriculture and horticulture, particularly viticulture and oleiculture (Reference Finkelstein, Gadot and LanggutFinkelstein et al. 2022).

The northeastern Negev of southern Judah, however, was more marginal than central Judah, receiving only about 100 to 300 mm of rainfall per year. Similar to large parts of Edom, vegetation in this nearly treeless Irano-Turanian zone was restricted primarily to species of grasses and shrubs (Reference CordovaCordova 2007: 95–104; Reference Langgut, Finkelstein, Litt, Neumann and SternLanggut et al. 2015: 219). For subsistence purposes, it was suited to less reliable rainfed farming, but better suited to pastoralism. In the Negev to the south, and the Wadi Arabah that lay between the northeastern Negev and Edom, the landscape shifted again into desert (Saharo-Arabian Zone), suitable for limited range pastoralism, though reliant on a specialist knowledge of the region (Reference CordovaCordova 2007: 104–121).

Perhaps most significant to the northeastern Negev, was the east-west oriented Beersheba-Arad Valley that crossed it, with its perennial springs, settlements, and forts, establishing it as an ideal route by which to cross from the Wadi Arabah to the Coastal Plain. At its eastern end, the valley was entered from the northern Arabah likely via the Nahal Heimar, a route guarded by the fort at Horvat ʿUza and its subsidiary towers (Reference Cohen-Sasson, Varoner, Frieman and HerriottCohen-Sasson et al. 2021), with a secondary route entering the Beersheba-Arad Valley from the south in the environs of Tel ʿAroer (Reference Ben-Yosef, Najjar, Levy, Levy, Najjar and Ben-YosefBen-Yosef et al. 2014: 546, Fig. 6.39). At its western end, the Beersheba-Arad Valley linked with the Nahal Gerar to the northwest and the Nahal Besor to the southwest, both of which intersected on the Coastal Plain at the site of Tel Jemmeh, a supremely important junction in the southern Levant.

2.2 Negotiating Autonomy in an Imperial World

The independent nature of southern Levantine kingdoms, particularly Judah and Edom, was never a given. Their small size and marginality often resulted in influence from neighboring kingdoms and empires, if not outright control. During the late eighth century BCE, the geopolitical situation in the southern Levant was fraught. The expanding Assyrian Empire had threatened the southern Levant for more than a century, and now under the vigorous leadership of Tiglath-Pileser III began to seize more direct control. In particular, the conquest of Damascus in 732 BCE ushered in an era of southern Levantine subservience to Assyria, with nearly every king subsequently recorded as providing tribute and entering into “client” status (Reference WestbrookWestbrook 2005). The Assyrian sources identify a named king of Judah – Ahaz – presenting tribute to Tiglath-Pileser III at this time (Reference Koch, Keimer and PierceKoch 2022: 694), and name Qausmalak of Edom as similarly entering into client status through tribute payments (Reference CrowellCrowell 2021: 111–114). From this period onward, the southern Levant, in particular Judah and Edom, became formally entangled in the Assyrian Empire, with Assyrian expectations of tribute and access to strategic sites.

The close of the eighth century BCE was punctuated by a significant disruption, at least in relation to Judah. Following the shocking and untimely death of Sargon II in Anatolia, Hezekiah of Judah sought to break free from Assyria, encouraged and aided by Egypt. Despite the “victory” claimed by Judah as the city of Jerusalem was not destroyed, Sennacherib’s 701 BCE suppression of the rebellion is widely attested archaeologically in destruction levels at numerous sites throughout Judah, marking a clear shift in the settlement system of the region (Reference FaustFaust 2008; Reference Koch and Sapir-HenKoch and Sapir-Hen 2018: 439). Archaeologically, these destructions are convenient as they permit a diachronic comparison between the late eighth and seventh century BCE phases of the settlements (Reference DanielsonDanielson 2020a: 133–140). Edom does not exhibit the same pattern of destruction as it does not appear to have participated in the rebellion. Despite the rebellion, throughout the seventh century BCE and until the gradual dissipation of Assyrian influence ca. 630 BCE, Judah and Edom continued to function as client kingdoms of the Assyrian Empire.

In fact, in the case of Edom, the arrival of Assyrian influence may have served as a catalyst for the establishment of a ruling elite at Busayra. As previously discussed, it was not until the late eighth century BCE that the trappings of elite authority could be detected in the sedentary settlements of the highlands, especially visible in the construction of the palatial and temple complexes on the acropolis of Busayra (Reference PorterPorter 2004: 384–387). These structures have long been noted for their Assyrian influence, most notable in their use of podia (tamlû), and the architectural forms of the temple and palace (Reference Reich, Kempinski and ReichReich 1992: 219–220; Reference BienkowskiBienkowski 2002a: 94–95, 199–200, 478–482). Their late eighth century BCE date is thus quite striking when considered in tandem with Assyrian records of tribute, and with the named Qausmalak drawing Edom into client status with Assyria.

Assyrian influences in construction, however, need not imply formal Assyrian presence, as there remains little compelling evidence for Assyrian occupation or even the presence of Assyrian officials (Reference BennettBennett 1982: 187; Reference CrowellCrowell 2021: 356–357). Considering the unique ways that the “Assyrian” features are not exact replicas of Assyrian elements and the ways by which southern Levantine traditions were interwoven with them, more compelling explanations highlight local architects influenced by Assyrian architects or visits to Assyrian courts, with the Assyrian features lending prestige to local contexts of authority (Reference Reich, Kempinski and ReichReich 1992: 214–215; Reference PorterPorter 2004: 385–386; Reference BrownBrown 2018b: 66–70; Reference Danielson and FesslerDanielson and Fessler 2023). The contemporaneity of the creation of political authority at Busayra with the advent of client status to Assyria is not meant to imply that the creation of sedentary sociopolitical complexity necessitated Assyrian influence. Rather, at this time, a confluence of circumstances created an opportune moment in which ambitious elites in the region could formalize and expand their own power. Notably, elite symbols of dominance, readily visible from the expanding, alluring, and seemingly inevitable Assyrian Empire, presented an accessible and impactful repertoire by which to communicate such authority.

Similarly, this influence and prestige can be detected in certain pottery forms, forms that were inspired in part by Assyria but thoroughly indigenized in local tradition. In the southern Levant, and in Edom in particular, the presence of “Assyrian-style” pottery has long been noted (Reference GlueckGlueck 1967; Reference MazarMazar 1985), often overstating their “Assyrian” elements and implying the local varieties to be derivative, with influence and prestige unidirectional. More recent studies have since emphasized the social and regional contingencies of the specific forms found in the corpus from Edom (Reference ThareaniThareani 2010; Reference TebesTebes 2011b), and the ways by which the unique painted decorations on many of the vessels evoke a more North Arabian tradition (Reference TebesTebes 2013, Reference Tebes2015) that could be linked to certain contexts and modes of consumption (Reference DanielsonDanielson 2020a: 244–251, Reference Danielson, Arbuckle MacLeod and Hamm2022: 128–133).

Indeed, when considering the “Assyrian-style” forms from the southern Levant more generally, Alice Reference HuntHunt (2015) has stressed the ways that local agency resulted in specific modes of use for certain forms that are not a reflection of their use in the Assyrian heartland. In particular, Hunt’s analysis of the corpus from nearby Tell Jemmeh demonstrated significant differences in form, surface treatments, and usage patterns from Assyria, concluding that they served different social and semiotic functions that reflected the local population’s perception of Assyria, and their relation to its empire (Reference HuntHunt 2015: 146–181). In this way, as we see in the pottery from Edom, the selective adoption of certain forms of consumption vessels that are integrated in a broader corpus, and are further decorated in motifs and patterns that are specific to the region, demonstrate a local adaptation of prestige objects that evoke a particular relationship with Assyria while serving the needs of the local population.

Returning then to the larger narrative, the formalization of sociopolitical complexity in a sedentary form centered at Busayra appears to have coincided with the arrival of Assyrian dominance in the southern Levant, and as will be discussed presently, with factors including the flourishing of the South Arabian trade. Assyrian interest in the Levant, beyond ideological conquest and the proximity of Egypt, lay in its ability to extract wealth and resources as is evident in the tribute lists associated with conquest and client subservience (Reference Dalley and FrahmDalley 2017; Reference Koch, Keimer and PierceKoch 2022: 701–702). As such, one of the goals of Assyria in the southern Levant appears to have been the creation of a “frictionless” corridor for Assyrian military personnel, as well as the transfer of goods back to Assyria (Reference FesslerFessler 2016; Reference Danielson and FesslerDanielson and Fessler 2023). In this model, the southern Coastal Plain and the Assyrian investment at Tell Jemmeh were quite significant as they stationed Assyrian authority at a position where they could benefit from the confluence of Arabian, Levantine, and Mediterranean trade networks, at the border with Egypt (Reference KochKoch 2018: 371–375). Targeted control over this region, with a particular emphasis on sites including Tell Jemmeh (Reference Ben-ShlomoBen-Shlomo 2014), allowed Assyria to maximize influence and wealth with the least amount of investment (Reference ThareaniThareani 2016: 94–95).

Provided tributes were paid, Assyria presented something of a laissez-faire approach to local polities, positioning agents of their empire in strategic locations, and relying on local elite collaborators to manage the diverse polities and populations throughout the region (Reference FaustFaust 2021: 131–132, 209–210). This situation is particularly apparent in late Iron Age Transjordan – especially Edom – where Assyria only ruled indirectly through local proxies who navigated a tenuous position between their local populations and external imperial rulers, ensuring loyalty and tribute payments to Assyria (Porter 2004; Reference YamadaYamada 2005: 79–80; Reference Faust, Aster and FaustFaust 2018: 30–36). In such a context, Craig Tyson’s emphasis on the idea of the peripheral elite as imperial collaborators is particularly cogent (Reference Tyson, Tyson and Herrmann2018). Using the kingdom of Ammon as an example, Tyson has explored this idea to emphasize the agency of the local elite, analyzing the ways by which they maximized their own positions within society while serving as intermediaries with the Assyrian Empire (Reference Thareani, Lipschits, Gadot and AdamsThareani 2017: 417–418; Reference Tyson, Tyson and HerrmannTyson 2018). This heuristic can similarly well explain the diverse forms of Assyrian “influence” seen throughout Edom, while positioning them within a framework that highlights the role of the local elite in navigating transregional politics, utilizing their own local, specialist knowledge to maximize their position.

4.1 Foodways: Maintaining and Trespassing Tradition

Food stands at the center of social interactions. To begin a broader discussion on culinary practices in the northeastern Negev, a cultural memory from the biblical text provides a unique window into some of the theoretical concepts associated with foodways studies. Discussed in further detail in Section 5.1, the Jacob and Esau narratives from the latter portion of Genesis are notable for their etymological and etiological understandings of the origins of Judah/Israel and Edom as from these legendary patriarchs (Genesis 25, 27, 32–33). The setting of the narrative is in this region, and its redaction is contemporary to this late Iron Age context and the centuries following (Finkelstein and Römer 2014).

The Jacob and Esau narratives center on the competitive, fraternal relations between the twin figures of Jacob and Esau, with a portion of the text devoted to Esau somewhat unwittingly losing his birthright to the younger Jacob. Amidst this contention, two meals take place. The first consists of hunted game prepared in a “savory” (מטעם) way that was presented with bread and wine to the patriarch of the family, Isaac, in a setting of social significance, indicating a degree of prestige placed on the (hunted) meat (Genesis 27:17). The second meal consists of a red lentil stew (נזיד עדשים; Reference Pace, Altmann and FuPeters 2016: 117–119), prepared by Jacob, but desired in a significant enough way by Esau to “trade” his birthright for it (Genesis 25:29–34). The text plays on the meaning of “red,” seeking to etymologically and etiologically associate the meal with Esau and Edom. From these meals consumed by the legendary patriarchs of these regions, we see first a prestige placed on the consumption of meat, and second, that an iconic food dish literally stood at the center of the interactions between Jacob and Esau, who stand in the narrative as metaphors for Judah/Israel and Edom.

This identification of food as at the center of interactions between the brothers, but at the same time more abstractly, the metaphorical social groups they represent, highlights a key feature of foodways study. That is, the importance placed on food beyond its role for subsistence. Food stands at the center of diverse and scaled social engagements, from routine family meals to celebrations of the creation of social bonds (e.g., marriage), and to large commemorations of religious events. Due to repetitive consumption primarily in group settings, food can become imbued with emotion, standing at the forefront of ideas of heritage and holding a powerful semiotic role in its social milieu (Reference AppaduraiAppadurai 1981; Reference Brulotte and Di GiovineBrulotte and Di Giovine 2014).

Understanding food then as socially sensitive as well as culturally conservative, allows it to serve as a readily identifiable marker of identity, creating distinction between groups, with those who eat similar food viewed positively, and those with unrecognizable foods viewed with distrust or even disgust (Reference TwissTwiss 2012, Reference Twiss2019). Numerous studies have engaged with the ways that cuisine can be used to express association with a certain social identity, particularly among immigrant communities or in culture contact situations (Reference Brighton, Metheny and BeaudryBrighton 2015; Reference Franklin, Metheny and BeaudryFranklin 2015; Reference DanielsonDanielson 2021; Reference Damm, Battini, Brody and SteadmanDamm 2022). In such contexts, numerous examples can be seen where traditional foodways are maintained over successive generations, even when separated from places of origin (Reference Ben-Shlomo, Shai, Zukerman and MaeirBen-Shlomo et al. 2008; Reference Fantalkin, Spataro and VillingFantalkin 2015). Likewise, as food procurement and preparation are often a gendered activity, they can allow for a more nuanced examination of aspects of the household economy (Reference MeyersMeyers 1988: 146–147, Reference Meyers, Edwards and McCullough2007; Reference Gero and ConkeyGero and Conkey 1991).

The Jacob and Esau narrative also highlights another key feature of foodways that can be traced archaeologically – cooking pots. In the text, these material objects are highlighted by the act of “boiling” (נזיד), or cooking different ingredients to produce a distinctive and desirable meal (Reference Shafer-ElliottShafer-Elliott 2013: 148–156; Reference PetersPeters 2016: 117–119). In the archaeological record of the southern Levant, cooking pots are ubiquitous. Due to daily thermal stresses, cooking pots tend not to survive for long periods of time. Their short-lived nature accentuates their ubiquity, and allows for the tracking of morphological variations. More importantly, in the absence of residue analyses, the fabrics and forms of vessels can provide insight into different cuisines, recipes, and ingredients. For example, different sizes of vessel openings permit differently sized ingredients and different levels of liquid evaporation, with certain fabrics ideal for higher or lower temperatures. Cumulatively, these factors result in diverse textures and tastes of the food prepared in certain types of pots (Reference RiceRice 1987: 239–240; Reference TebesTebes 2011b: 87–88; Reference Magness, Altmann and FuMagness 2014: 50–51; Reference DanielsonDanielson 2020a: 241–244). As coarse wares used in non-performative settings, cooking pots are likewise seldom the objects of trade between different cultural groups. Combined with their socially sensitive nature, and the fact that in pottery production the rigid nature of complex motor habits required to produce certain forms are only attained through years of immersive apprenticeship, cooking pots tend to serve as strong indicators of the presence of their bearers and/or producers (Reference Dietler, Herbich and StarkDietler and Herbich 1998; Reference Gosselain and StarkGosselain 1998; Reference Nelson, Metheny and BeaudryNelson 2015). Lastly, the identification of standardized locales of production for cooking pots in the late Iron Age emphasizes the understanding of their regionalized use (Reference Ben-Shlomo, Bouzaglou, Mommsen and SterbaBen-Shlomo et al. 2023).

The culinary landscape of the southern Levant was dominated by the so-called Levantine triad: bread (wheat, barley), wine (grapes), and oil (olives; Reference MacdonaldMacdonald 2008: 19–24; Reference Pace, Altmann and FuPace 2014: 187–189). Evidence for the centrality of these foods in the diet of the northeastern Negev can be found in the Arad Ostraca where various combinations of flour (קמח) or bread (לחם), oil (שמן), and wine (יין) are listed as the rations provided for persons associated with the military administration in the region during the late seventh and early sixth centuries BCE (Reference AharoniAharoni 1981: 12–38; Reference Pace, Altmann and FuPace 2014: 188). While forming staple pillars of the diet, however, these were supplemented by a variety of vegetables, particularly legumes (e.g., lentils, peas, vetch), and fruits (e.g., fig, date, pomegranate, apple), dairy as a secondary product from sheep and goats, meats from domesticated animals (especially sheep and goat), a variety of hunted species (e.g., gazelle, deer, ibex, hare, boar, etc.), and imported fish (Reference MacdonaldMacdonald 2008; Reference DanielsonDanielson 2020a: 191–193, 221–224).

Various combinations of these foods, based on preference and availability, would result in dishes that were both flavorful and familiar. Certain ingredients, such as meat, would have been restricted to more festive occasions, elite contexts, or ritual activity (Reference King and StagerKing and Stager 2001: 68; Reference Magness, Altmann and FuMagness 2014: 35). In an average house, up to three meals would be eaten daily, with the main meal in the evening consisting of a type of vegetable (and meat?) stew, flavored with herbs and sopped up with bread (Genesis 25:29–34; Ruth 2:14; Reference King and StagerKing and Stager 2001: 67–68). Hunted game or imported fish could supplement the diet for those with the means by which to acquire them (Genesis 25, 27; Reference Pace, Altmann and FuPace 2014: 193–194). Consumed foods, however, would also be constrained by geography, social status, and gender, with those at lower social levels lacking access to greater nutritional diversity, and likely women consuming less prestige food, like meat, than their husbands and sons (Reference MacdonaldMacdonald 2008: 91–93). Nonetheless, the continued preparation and consumption of such foods over time would result in a network of entangled recipes, cuisines, and practices perpetuated by different social groups in different regions.

4.1.1 Food Production and Cooking Vessels

While a single type of cooking pot was not necessarily restricted to a single cuisine, and apart from residue analyses determining ingredients, different cooking pot types do signal variant practices, and through differences in size and form, variant culinary traditions. Geographic patterns in the distribution of specific cooking pot types can then reveal insights into the culinary traditions that were preferred in certain regions or at certain sites. During the late Iron Age, no less than fifteen distinct cooking pot forms can be identified in the northeastern Negev and southern Jordan (Figure 6). Each form presents its own history and distribution resulting in unique social narratives. For example, Type CP10 is represented by only a single exemplar at Tel Malhata, but possesses a form and fabric that originated in the Aegean (Reference Freud, Beit-Arieh and FreudFreud 2015: 201; Reference DanielsonDanielson 2020a: 164, 482). Alternatively, Type CP13 was dominant at the site of Tel Beersheba in the eighth century BCE, but appears to not have been well-used beyond this site (Reference DanielsonDanielson 2020a: 165, 485). As detailed analyses of these cooking pot forms have been presented elsewhere (Reference DanielsonDanielson 2020a: 147–251, 420–488, Reference Danielson2021, Reference Danielson, Hensel, Zvi and Edelman2022: 122–128), the following discussion will track only some of the most dominant forms from the late eighth through the early sixth century BCE, presented in Table 2.

Figure 6 Cooking pot types of the late Iron Age northeastern Negev

(figure by the author; vessel forms adapted from Reference Singer-AvitzSinger-Avitz 2002: Fig. 14:6, 15:9; Reference Freud and Beit-AriehFreud 2007: Fig. 3.41:9, Reference Freud, Beit-Arieh and Freud2015: Fig. 4.132:2)

Table 2 Descriptions of major cooking pot (CP) types of the late Iron Age northeastern Negev

Type	Description	Regional patterns and date rangeFootnote 11
CP1	Neckless CP with an everted rim and two handles; often called a “Judahite” CP.	Common in Judah, Iron IIC.
CP2	Ridged-neck CP with two handles; often called a “Judahite” CP.	Common in Judah, Iron IIC.
CP3	Stepped-rim CP with an out-flaring neck with two handles; often called a “Coastal” CP.	Common in the Beersheba–Arad Valley, but better-known in the southern Coastal Plain, Iron IIC.
CP4	Neckless or short-neck CP with a ridged rim, and two to four handles; often called an “Edomite” CP.	Dominant in southern Jordan, but also well attested in the northeastern Negev; predominantly Iron IIC, although it also appears in late Iron IIB.
CP5	Simple rim or holemouth CP with incipient or no handles. It is often handmade and presents numerous localized subvariants. It is often called an “early” Edomite CP, and is similarly at times called “Negevite,” though differentiated from CP6 in this typology by its lack of a flat base.	Attested throughout southern Jordan and the northeastern Negev; dates predominantly to the Iron IIB, but is also present in Iron IIC.
CP6	Handmade CP with a flat base; also called a “Negevite” CP or krater.	Common in Iron IIB and IIC, as well as in earlier centuries; attested in southern Jordan, at Tall al-Khalayfi and at Kadesh Barnea, although not attested in the northeastern Negev.

To demonstrate distinct culinary practices at different sites, we can turn to two forts from the northeastern Negev: Horvat Tov and Horvat ʿUza. Both of these sites date to the seventh and early sixth centuries BCE, and on the basis of geography, material remains, and the rich inscriptional corpus from Horvat ʿUza, are securely identified as operated by the Judahite military administration (Reference Beit-AriehBeit-Arieh 2007b; Reference ItkinItkin 2020). Beginning with Horvat Tov, a near-uniform picture of culinary practices emerges. The fort is dominated by cooking pots of the CP1 and CP2 variety (Figure 7). These forms present both an open (CP1) and closed (CP2) cooking pot type that is dominant in the region at this time, demonstrating why the label “Judahite” has often been applied to them. The shared geography of these open and closed forms suggests that they reflect different aspects of a related cuisine. The open, wider-rimmed CP1 pots could have been used to produce drier stews with the option of larger ingredients (bones, meats?), whereas the closed CP2 form was better suited toward moisture retention, but with its restricted opening limiting the size of ingredients. The near ubiquity of these forms at Horvat Tov is broken by one vessel dissimilar to those found in the northeastern Negev, and by a single exemplar of Type CP4. Excavated in the pillared structure in the southern corner of the site and in the same context as two exemplars of both CP1 and CP2, the Type CP4 cooking pot is notable not only in its isolation at the site, but in its own context of excavation.

Figure 7 Cooking pot types at Horvat Tov

Turning to the site of Horvat ʿUza, a rather divergent portrait emerges. Whereas, similar to Horvat Tov, types CP1 and CP2 are the most common, types CP3 and CP4 are also present, accounting for 20 percent and 15 percent of the overall assemblage respectively (Figure 8). The picture of culinary practices here is clearly more complex. The form CP3 with its out-flaring neck represents a type that is well attested in the Coastal Plain to the west. Similarly, while only present in a single exemplar at Horvat Tov, CP4 is common at Horvat ʿUza. Notably, however, both types CP3 and CP4 are not restricted to certain areas in or outside the fort, but are found scattered throughout. In fact, in contexts such as L.741 in the south-central area of the site, we find that all four types of cooking pots were excavated in the same domestic structure. If their location of deposition matches their use context, then diverse foods were prepared in the same space. Despite their contemporaneity and operation by the same political entity, Horvat Tov and Horvat ʿUza present starkly different portraits of culinary practice.

Figure 8 Cooking pot types at Horvat ʿUza

Here, geographic location and our understanding of routes through the northeastern Negev is of importance. Notably, while Horvat ʿUza was located on the major east-west route running from southern Jordan to the Mediterranean, Horvat Tov was located a distance northward, on a north-south branch off this highway.Footnote ¹² Horvat ʿUza’s position along a route connecting eastern and western regions – that each presented its own specific traditions – appears to have afforded it greater culinary diversity. In contrast, Horvat Tov’s location along a less-traversed route, and more deeply embedded in Judah, may have resulted in a more restricted culinary repertoire. Indeed, if one takes a broader perspective of the geographic distributions of cooking pots, an interesting pattern emerges.

When the relative quantities of excavated cooking pots are mapped spatially, a strong correlation between CP1 and CP2 emerges (Figure 9). Both forms are present in relatively similar quantities, and are prominent in the northeastern Negev and at Kadesh Barnea, but have not been excavated to the east. A single exemplar of CP1 has been published from Tall al-Khalayfi, where it was clearly a rare form. Type CP3, discussed previously in relation to Horvat ʿUza, is fairly well attested in the northeastern Negev, though at a lesser scale than CP1 and CP2 (Figure 10). In contrast, its distribution is accentuated to the west, where it is prominent at Tel Jemmeh, and while not depicted in Figure 10, is dominant in the Coastal Plain to the north, comprising for example, more than 90 percent of the cooking pot assemblage at seventh century BCE Ashkelon (Reference Stager, Master and SchloenStager et al. 2008: 86–87). This type is also well represented at Kadesh Barnea, intimating a high degree of connection between this site and the Coastal Plain.

Figure 9 Regional distribution of cooking pot type CP1 (left) and CP2 (right)

Figure 10 Regional distribution of cooking pot type CP3 (left) and CP4 (right)

In contrast to the previously discussed forms that are prominent in the northeastern Negev, CP4 is overwhelmingly the most dominant type in southern Jordan, present at every site for which there is available data: from ash-Shorabat in the north to Tall al-Khalayfi in the south (Figure 10). Alongside the presence of CP4 in southern Jordan are other well-known types (e.g., CP5, CP6, CP7, and CP8), though all are infrequent forms in comparison (Reference DanielsonDanielson 2020a: 158–163, 172–193, 475–480). Yet, CP4 is also attested at nearly every site in the northeastern Negev, though most prominently at the religious site of Horvat Qitmit, and the fortified towns of Tel Malhata and Tel ʿAroer. A similar, though less accentuated pattern can be identified for CP5, which similarly is known throughout southern Jordan and also the northeastern Negev at Tel Malhata and Tel ʿAroer, where it derives primarily from eighth century BCE contexts (Figure 11; Reference DanielsonDanielson 2020a: 435, 444–446). Its similarity in form and chronological relation to CP4 has led Liora Freud to suggest that the latter form developed from, and replaced, the former (Reference Freud, Beit-Arieh and Freud2015: 196). To highlight a final cooking pot type, CP6 is common in southern Jordan, particularly at Tawilan, Ghrareh, and Tall al-Khalayfi, and is also known at Kadesh Barnea, though not in the northeastern Negev (Figure 11). The long-lived nature of this form and its correlation with the more arid locales suggests its use among communities active in the desert regions (Reference Haiman, Goren, Bar-Yosef and KhazanovHaiman and Goren 1992; Reference Martin and FinkelsteinMartin and Finkelstein 2013).

Figure 11 Regional distribution of cooking pot type CP5 (left) and CP6 (right)

While the distributions of these forms suggest restricted regional use for certain cooking pot types (e.g., CP2), contrasted with wider use by other forms (e.g., CP4), it is necessary to further examine the contexts of production. As pottery production areas or kilns are lacking from the excavations in the northeastern Negev, we can turn to petrographic analysis of the clay fabrics to determine likely clay sources, and Instrumental Neutron Activation Analysis (INAA) to determine chemical similarity between vessels. Of the cooking pots excavated in the Negev, petrographic analysis of types CP1 and CP2 has identified predominantly terra rossa soils, likely originating in the Shephelah or central Judean Highlands to the north (Reference Iserlis, Thareani and ThareaniIserlis and Thareani 2011: 181; Reference Freud and TebesFreud 2014: 302). Comparisons to similar vessels throughout Judah have led David Ben-Shlomo et al. to suggest specialized production centers for these and other cooking pots, perhaps in the vicinity of Jerusalem (Reference Ben-Shlomo, Bouzaglou, Mommsen and SterbaBen-Shlomo et al. 2023; Reference Bouzaglou and Ben-ShlomoBouzaglou and Ben-Shlomo 2023). As such, these cooking pots stand in contrast to many of the other non-cooking pot vessels from the northeastern Negev that were instead made from the local loess (Reference Iserlis, Thareani and ThareaniIserlis and Thareani 2011: 179–183; Reference Ben-Shlomo, Bouzaglou, Mommsen and SterbaBen-Shlomo et al. 2023: 100).

The petrographic analysis of Type CP3 cooking pots identifies many as belonging to the Coastal Hamra soil group, indicating an origin on the Coastal Plain to the west, while others petrographically belong to the sandstone group of the Hazeva formation, suggesting they were produced in the environs of the northeastern Negev or northern Arabah Valley (Reference Freud and TebesFreud 2014: 286–287, 292). Petrographic analysis of Type CP4 vessels excavated at Tel Malhata, Horvat ʿUza, and Horvat Qitmit have determined that they similarly belong to the sandstone Hazeva formation, produced in the environs of the eastern Negev or northern Arabah (Reference Freud and TebesFreud 2014: 285–286, 289–291). Notably, several Type CP4 vessels from Tel ʿAroer have been associated with Lower Cretaceous shales that may have originated in the highland region of southern Jordan (Reference Iserlis, Thareani and ThareaniIserlis and Thareani 2011: 180–182).Footnote ¹³ From southern Jordan, only INAA results are available (from Busayra, Tawilan, Umm al-Biyara, and Ghrareh), and only in limited numbers for Type CP4 cooking pots. Nonetheless, the INAA results have shown that the vessels from southern Jordan are chemically dissimilar from those in the northeastern Negev, indicating different locales of production (Reference Gunneweg and MommsenGunneweg and Mommsen 1990, Reference Gunneweg, Mommsen and Beit-Arieh1995: 281, 285–286; Reference Gunneweg, Beier, Diehl, Lambrecht and MommsenGunneweg et al. 1991; Reference Gunneweg, Balla and BienkowskiGunneweg and Balla 2002).

While less petrography has been done on Type CP5 vessels, examples from Tel ʿAroer and Tel Malhata have similarly identified them as deriving from the sandstone Hazeva formation, suggesting again, production in the vicinity of the northeastern Negev or northern Arabah Valley (Reference Iserlis, Thareani and ThareaniIserlis and Thareani 2011: 181; Reference Freud and TebesFreud 2014: 302). INAA analysis of a single Type CP6 “Negevite” cooking pot amidst a greater sample from Kadesh Barnea, has tentatively suggested an origin in Edom, with the greater “Negevite” corpus showing numerous examples deriving from both the Negev and southern Jordan (Reference Gunneweg, Beier, Diehl, Lambrecht and MommsenGunneweg et al. 1991). Petrographic analyses of this type of vessel from earlier centuries (e.g., Reference Martin and FinkelsteinMartin and Finkelstein 2013) have established diverse contexts of production and use, highlighting their association with mobility.

The emergent portrait of cooking pot distributions and contexts of production is striking. Both CP1 and CP2 vessels appear closely related to one another, and signify the culinary traditions most deeply rooted in the late Iron Age landscape of the northeastern Negev and regions to the north, where they were produced. The diverse assemblage at Horvat ʿUza appears to exemplify the east-west connections of this site, with the CP3 forms linking with traditions to the west and the CP4 forms to common traditions in southern Jordan. Yet, notably, many of the CP4 forms from the northeastern Negev – including those from Horvat ʿUza – were produced locally (Reference Freud and TebesFreud 2014). Thus, CP4, and its likely forebear CP5, represent traditions local to the northeastern Negev for at least a century and a half, highlighting a complex series of overlapping and intersecting culinary traditions. And yet, a diversity of practices was not attested at every site.

Just as Horvat Tov and Horvat ʿUza presented different culinary traditions, we can turn to two additional sites to compare: Tel ʿIra and Tel Malhata. These latter sites present multiple strata from the eighth through sixth century BCE, allowing for diachronic analyses of their culinary traditions. To begin, Tel ʿIra presents two major relevant strata: Stratum VII dated to the late eighth century BCE, and Stratum VI dated to the seventh through early sixth centuries BCE. The cooking pot forms are presented in Figure 12, where the center “mixed” stratum identifies the cooking pots in contexts that could not confidently be assigned to either Stratum VII or VI. Both of these strata, however, contain a fairly restricted number of cooking pot forms, with the earlier level presenting an overwhelming number of types CP12 and CP14, and Stratum VI presenting an overwhelming abundance of types CP1 and CP2, with the mixed strata maintaining the same general pattern, albeit mixed. While not discussed previously, the prominent Stratum VII types CP12 and CP14, represent the major open and closed cooking pot forms of the eighth century BCE northeastern Negev (Reference DanielsonDanielson 2020a: 165–166, 484, 486), effectively mirroring the subsequent role of CP1 and CP2. Accordingly, due to the inherent “local” nature of these earlier forms, Tel ʿIra appears to exhibit considerable culinary stability over successive centuries, evolving to a degree, but remaining remarkably consistent. The assemblage is not entirely homogenous, however, as three exemplars of Type CP3 and three additional exemplars of CP4 are attested, but these latter forms comprise a minority.

Figure 12 Cooking pot types at Tel ʿIra

Tel Malhata, located less than 4 km to the southeast, presents a stark contrast. Here, three main strata are available for analysis: Stratum IV of the eighth century BCE, Stratum IIIB, and Stratum IIIA dated to the seventh through early sixth centuries BCE. While types CP12 and CP14 are similarly well represented together with an additional, local, open form CP11, other forms are common (Figure 13). A significant number of Type CP5 are attested, together with a significant number of Type CP4, signifying both its longevity in the northeastern Negev and its popularity. In the early seventh century BCE (Stratum IIIB), there is continuity in the use of local cooking pot types CP11, CP12, and CP14, as well as the introduction of the prominent seventh century BCE CP1 and CP2 forms (Figure 14). Notably, however, during this phase Type CP4 became the most prominent, accounting for 33 percent of the overall assemblage (Reference DanielsonDanielson 2021: 101–102). This portrait of CP4 prominence increases in the subsequent Stratum IIIA until the destruction of the site in the early sixth century BCE. In this final phase, Type CP4 accounts for 42 percent of the overall assemblage, with the newly introduced Type CP3 similarly prominent (24 percent). Types CP1 and CP2 are also attested, though as part of a distinct minority. The diversity of cooking pot types is also identifiable in the same activity (or discard) areas at Tel Malhata exemplifying that these traditions existed adjacent and with one another, rather than in isolation (Reference DanielsonDanielson 2020a: 206–211, 444–456, Reference Danielson2021: 101–103).

Figure 13 Cooking pot types at Tel Malhata, Stratum IV

Figure 14 Cooking pot types at Tel Malhata, Stratum III

Despite their proximity to one another, Tel Malhata presents a vastly different culinary story than Tel ʿIra. The contrast seen between Horvat Tov and Horvat ʿUza, presented in a contemporaneous moment, can be teased out diachronically at Tel ʿIra and Tel Malhata, where the former’s relatively stable homogeneity can be contrasted with the increasingly diverse set of culinary traditions at the latter. These differences likely indicate variant social preferences and activities occurring at diverse sites. In particular, we can also determine that geographic positioning as well as site functionality played a role in the patterning of culinary practices. For example, Horvat Tov’s restricted assemblage can be compared to Tel Arad, another Judahite fort located on the north-south road. With these forts, we might also consider Tel ʿIra, whose similarly restricted assemblage of cooking pots is identifiable over successive strata. This large fortified site was located at a higher elevation along the northern side of the valley, and if its identification as Ramat Negeb is correct, may have served an important military/administrative role for the region during the seventh century BCE, perhaps assuming the role held by Tel Beersheba in the preceding century (Reference AharoniAharoni 1981: 46–49; Reference ThareaniThareani 2011: 5). While also a military fort, Horvat ʿUza presents the same major types but with increasing diversity. Here, its location along the major east-west route is likely a central factor driving the diversity that approaches the degrees of complexity seen at Tel Malhata and Tel ʿAroer. These latter, culinarily diverse sites, represent significant nodal points along the major east-west trade route through the region.

At an immediate practical scale, these divergent cooking pot forms represent connections between their users and the locations where the vessels were made. The aforementioned petrographic analyses of the cooking pots are now beginning to identify regional production centers, likely due in part to the higher technical skill necessary to produce durable cooking wares, and the surrounding regions that imported them (Reference Ben-Shlomo, Bouzaglou, Mommsen and SterbaBen-Shlomo et al. 2023). This situation especially applies to types CP1 and CP2, whose terra rossa soil type and similarity to other forms suggest a locale of origin likely in the Jerusalem area (Reference Ben-Shlomo, Bouzaglou, Mommsen and SterbaBen-Shlomo et al. 2023: 99–101; Reference Bouzaglou and Ben-ShlomoBouzaglou and Ben-Shlomo 2023). As such, connection, exchange, and movement between these locales were responsible for the presence of these forms, exemplifying that this directionality in culinary (and social) orientation was particularly prominent at sites such as Tel Arad, Horvat Tov, and Tel ʿIra. Here, their association with the military administration of Jerusalem, and their nearness to the north-south road appears to have affected their culinary traditions.

In contrast, vessels like CP4, produced in the northeastern Negev or northern Arabah, represent an entirely different set of connections and exchange as well as production techniques (Reference Bouzaglou and Ben-ShlomoBouzaglou and Ben-Shlomo 2023). Despite their formal similarity to corresponding vessels in southern Jordan, CP4 vessels from the northeastern Negev appear to have been largely, though not exclusively, locally produced (Reference Freud and TebesFreud 2014). Nonetheless, their shared forms suggest a broader shared culinary tradition (Reference DanielsonDanielson 2021). Yet, this tradition did not suddenly “appear” in the northeastern Negev, but was established already as early as the late eighth century BCE as seen at Tel Malhata (see Figure 13; Reference Singer-Avitz and TebesSinger-Avitz 2014). Further, if Liora Freud’s hypothesis holds, that Type CP5 forms a precursor that evolves into Type CP4 (Reference Freud, Beit-Arieh and FreudFreud 2015: 196), then our perspective on the rootedness of this tradition in the northeastern Negev becomes even more accentuated.

And yet, as certain imported forms suggest, and as the rapid increase in popularity of CP4 and CP3 in the seventh century BCE indicates, these traditions were supplemented through natural population growth as well as the introduction of new persons to the region (Reference DanielsonDanielson 2021). In such contexts of immigration, the maintenance of familiar culinary traditions is not only unsurprising, but to be expected. Beyond the maintenance of traditional culinary practices, the cooking pot forms from southern Jordan are found amidst a broader web of diverse practices and preferences that intersected with local traditions in a complex and entangled network that did not always fully overlap or integrate (Reference Danielson, Hensel, Zvi and EdelmanDanielson 2022). Just as Horvat Tov and Tel Arad maintained a preference for CP1 and CP2 forms, the inverse can be seen at the religious site of Horvat Qitmit. Here, 80 percent of the overall forms were of Type CP4, which together with iconography and inscriptions, identify the site as closely linked with traditions from southern Jordan (see Section 4.2), and appearing to serve a large southern Jordanian culturally affiliated population in the northeastern Negev (Reference Beit-AriehBeit-Arieh 1995a; Reference DanielsonDanielson 2020b: 122–127, Reference Danielson, Hensel, Zvi and Edelman2022: 133–137).

The integration of these cooking pots – to varying degrees at varying sites – is best understood as relating to movement and cultural interaction. Notably, the sites with the greatest culinary diversity were located on the major east-west road from Edom to the Mediterranean. At these sites, diversity and its integration are attested even in individual domestic contexts, as seen at the fort of Horvat ʿUza. Further, when textual descriptions and artistic depictions of food preparation are considered, it is possible to associate this activity with persons of a lesser perceived social status, including women, as a part of the broader household economy (Reference Gero and ConkeyGero and Conkey 1991; Reference EbelingEbeling 2010; Reference PetersPeters 2016: 69–81).Footnote ¹⁴ In this way, it is possible that degrees of culinary diversity as seen in a military fort garrisoned by Judahite male soldiers, was perhaps the result of social alliances created through intermarriage, or the result of prisoners of war or enslaved people holding to specific culinary traditions. Notably, reference to trade in humans is preserved in the Hebrew Bible, where texts set in the eighth century BCE record the transfer of forcibly “exiled communities” between Gaza and Edom (Amos 1:6–12),Footnote ¹⁵ with the road connecting these regions traveling directly past these sites in the northeastern Negev. As outlined in Section 3, the roles played by trade and pastoralism underscore how the region was inherently linked to mobility and movement, modes by which we might expect evidence of culinary diversity.

4.1.2 Food Consumption and Iconic Tableware Traditions

Beyond food preparation and the vessels involved, we can also focus on the practice of food consumption and related tableware assemblages. Local tableware of the northeastern Negev is largely undecorated (Figure 15A), and can be contrasted with a diverse and often decorated tableware assemblage common to southern Jordan, which also extends selectively to certain locations in the northeastern Negev. This tableware is known as “Southern Transjordan-Negev Pottery”, or “Busayra Painted Ware” (BPW) after one of its earliest and most prominent contexts of discovery (Figure 15B). At Busayra, vast amounts of this ware were excavated atop the acropolis in the vicinity of the palatial and temple complexes (Reference Bienkowski and BienkowskiBienkowski 2002b). The corpus of BPW consists predominantly of bowls and kraters – forms used in the presentation and consumption of food – and stands out for its distinctive painted patterns and motifs, consisting of ladder, triangle, net, triglyph, and metopes applied in black, brown, red and white colors (Reference ThareaniThareani 2010; Reference TebesTebes 2011b; Reference Bienkowski and GitinBienkowski 2015). This tableware is further marked as distinct from the northeastern Negev assemblage on the basis of form as well as the not-uncommon presence of plastic denticulations.

Figure 15 (A) Tableware common in the northeastern Negev of southern Judah (figure adapted from Reference Beit-Arieh, Freud and GitinBeit-Arieh and Freud 2015: pls. 3.4.1, 3.4.3); (B) examples of Busayra Painted Ware

(BPW; Figure adapted from Reference Bienkowski and BienkowskiBienkowski 2002b: Figs. 9.1, 9.2, 9.5, 9.7, 9.25, 9.26, 9.29, 9.30; Reference Cohen and Bernick-GreenbergCohen and Bernick-Greenberg 2007: Pl. 11.79: 12; Reference ThareaniThareani 2010: Fig. 1; Reference Beit-Arieh, Freud and GitinBeit-Arieh and Freud 2015: Pl. 3.4.1, 3.4.3; Reference Freud, Beit-Arieh and FreudFreud 2015: Fig. 4.130)

The iconic painted decorations of the BPW corpus, as found through southern Jordan, appear to follow a northwest Arabian tradition. Indeed, at each of the major oasis sites to the south of Edom (e.g., Tayma, Dedan, Qurayyah, etc.), iconic decorative patterns mark the pottery – and tableware – as distinct (Reference TebesTebes 2013, Reference Tebes2015; Reference Hausleiter, Luciani and HausleiterHausleiter 2014; Reference Luciani and AlsaudLuciani and Alsaud 2018). The BPW corpus has also often been (mistakenly) identified as related to, or derivative of, Assyrian Palace Ware (APW) and its southern Levantine adaptations and imitations. While the BPW corpus does include some APW forms, these selections are part of a much broader formal repertoire, with the imitative APW forms fully indigenized through their decorative motifs (Reference Brown, Tyson and HerrmannBrown 2018a). Further, this decorated assemblage is neither homogenous, nor linked to a single locale of production (Reference Iserlis, Thareani and ThareaniIserlis and Thareani 2011), yet the iconic decorations on them, particularly when contrasted with other non-decorative forms, stand out.

The consumption of food is inherently social, and tableware is necessary to facilitate dining. The consumption of food in groups creates situations in which social bonds are fostered or created anew, with the opportunity for performativity and the potential in these instances for material objects to serve a communicative role. Indeed, much anthropological literature has focused on the importance of communal consumption and feasts, highlighting them as contexts in which kinship or alliance relationships are fostered, obligations are created, and the aesthetics of distinction enforced (Reference Fox, Harrell, Baker, Allen, Middle and PooleFox and Harrell 2008; Reference PollockPollock 2012; Reference Altmann and FuAltmann and Fu 2014). In such contexts, status inequalities, and demonstrations of opulence can serve as a means by which to create and replicate power (Reference Dietler, Dietler and HaydenDietler 2001; Reference BrayBray 2003; Reference Jiménez, Montón-Subías and RomeroJiménez et al. 2011).

In contexts of performative commensality where social relationships and alliances are structured (Reference Dietler, Dietler and HaydenDietler 2001: 103–104), tableware can serve a subtle though key visual role. For example, it has been demonstrated in Near Eastern and analogous contexts that open areas with lowlight or firelight in which ritual feasting occurred, served to visually emphasize the elaborate decorations on serving vessels, highlighting their visual association with the performative aspects of feasts and the prestige of the hosts (Reference MillsMills 2007; Reference PorterPorter 2011: 41–46). Thus, whether or not locations where BPW was used can be definitively identified with feasting, the highly iconic decorations would nonetheless serve as an indexical marker of the prestige of the ware, and the individuals and contexts associated with its use (Reference KeaneKeane 2003). Moreover, through its semiotic association with elite commensality, when clusters of BPW are excavated in areas that can be linked to feasts, they may be understood as a form of political power (Reference KurtzKurtz 2001: 35–36).

The semiotic significance of such vessels is visible at the site of Busayra. Here, not only have the largest quantities of this decorated tableware been excavated, but also its most elaborate and highest quality versions. These wares have been found in significant quantities on the acropolis (Figure 16), with the vast majority excavated in Area B, between the palatial compound (Area C) and the temple complex (Area A; Bienkowski 2002b; Reference DanielsonDanielson 2020a: 172–176, 421). Of the limited available data related to foodways from the original excavations at the site, the faunal data from the acropolis has indicated a young age at slaughter for the most prevalent animal species: sheep (Ovis aries) and goat (Capra hircus; Reference BienkowskiBienkowski 2002a: 471–474). This culling pattern suggests that herd management strategies here were not wholly related to subsistence, but likely suggest elite consumption of prized (young) specimens. Moreover, the initial interpretations of the faunal data from the renewed excavations atop the acropolis have similarly confirmed this pattern, but also point to a higher-than-average culling of female sheep and goats, emphasizing the idea of conspicuous consumption in this space at Busayra (Reference BrownBrown 2018b: 117–121). While it is not possible to directly link the BPW tableware with the faunal data in a specific context indicating patterned feasts, the accumulation of this data atop the acropolis certainly points in the direction that elite consumption in communal settings occurred, and that BPW was featured.

Figure 16 Busayra Painted Ware vessels at Busayra

Elsewhere in Edom, clusters of BPW similarly appear, despite the paucity of excavation data from southern Jordan. At Tawilan, for example, BPW is found throughout the site, thoroughly integrated into diverse contexts (Reference Bennett and BienkowskiBennett and Bienkowski 1995; Reference DanielsonDanielson 2020a: 176–179, 423). Further south, at Ghrareh, a site understood as the southernmost node of Edomite authority in the highlands, we similarly see clusters of this ware (Figure 17). Notably at Ghrareh, the BPW is clustered in the large central structure, interpreted as an elite residence (Reference HartHart 1988; Reference DanielsonDanielson 2020a: 181–183, 426). While faunal data from Ghrareh is lacking, these indications suggest communal consumption using BPW. Similarly, at Ghrareh Type CP4 cooking pots were most prominent (52 percent; n = 45), with Type CP6 as the second most common (30 percent; n = 24). Indeed, when mapped spatially, the distribution and clusters of BPW show their prominence at Busayra, with different clusters of the ware seemingly extending from Busayra to various nodes throughout the established trade network (Figure 18).

Figure 17 Busayra Painted Ware vessels at Ghrareh

Figure 18 Regional distribution of Busayra Painted Ware

These wares are also attested to the west, throughout the northeastern Negev, though most prominently at Tel Malhata, Tel ʿAroer, and Horvat Qitmit, sites that demonstrate strong cultural links to southern Jordan. The locations where they were excavated at Tel ʿAroer are particularly intriguing. Here, we first see that BPW was present already in the eighth century BCE (Stratum IV–III) and spread throughout the majority of the site (Figure 19; Reference DanielsonDanielson 2020a: 198–201). The prominence of the ware also appears to increase in the seventh century BCE (Stratum II), and while it is still found in all areas of the site, it is most strongly clustered in the extramural contexts of Area D and especially Area A, which was interpreted as a caravanserai (Reference Thareani-SusselyThareani-Sussely 2007; Reference ThareaniThareani 2010). In this caravanserai context, numerous other items associated with food preparation and consumption have been identified (e.g., clay ovens, grinding stones, cooking pots). Likewise, amidst a broader program of herd management aimed toward secondary product exploitation, the limited faunal data – especially from the caravanserai – indicates a high percentage of animals culled in their first year, similar to Busayra, suggesting more conspicuous consumption (Reference Motro and ThareaniMotro 2011: 275–279; Reference ThareaniThareani 2011: 115–178, 301–311).

Figure 19 Busayra Painted Ware vessels at Tel ʿAroer

To further complicate the picture of cross-cultural interaction in the northeastern Negev, while it might be assumed that tableware aligned with traditions in southern Jordan would be used to consume food prepared in a cooking pot common to southern Jordan (e.g., CP4), in the caravanserai at Tel ʿAroer this does not appear to have been the case. Rather, while several CP4 vessels were excavated here, the vast majority of cooking was done in CP2 vessels (Reference DanielsonDanielson 2020a: 198–201, 435–437), exemplifying a unique entanglement between diverse forms of food preparation and consumption. Similarly, the inverse situation is also attested. For example, at Horvat ʿUza, while a sizable minority of CP4 vessels were excavated, the nearly nonexistent examples of BPW at the site indicate this was not a recognizable form of food consumption at this site, or indeed at any of the other military forts of the region (Horvat Tov, Tel Arad, Horvat Radum).

Ultimately, the distribution of BPW reveals a shared practice of food consumption that was clearly centered on Busayra and emanates throughout southern Jordan and the northeastern Negev, particularly at nodal sites of the region’s trade network. This is not necessarily reflective of an “ethnic” component, but rather reveals an association with prestige and status, echoing practices that are well attested in Edom. Furthermore, it is important to emphasize that similar to the petrographic and INAA analysis of CP4 vessels, BPW appears to have been produced in several major groups, one aligned with southern Jordan, and another with the northeastern Negev or northern Arabah (Reference Gunneweg and MommsenGunneweg and Mommsen 1990, Reference Gunneweg, Mommsen and Beit-Arieh1995; Reference Gunneweg, Beier, Diehl, Lambrecht and MommsenGunneweg et al. 1991; Reference Gunneweg, Balla and BienkowskiGunneweg and Balla 2002; Reference ThareaniThareani 2010; Reference Iserlis, Thareani and ThareaniIserlis and Thareani 2011). This pattern appears to indicate different points of production, signifying that it was shared in specific elite contexts throughout the region. Lastly, similar to the cooking pot data, we can highlight both the longevity of this practice and its differential integration into sites in the region, suggesting unique narratives for different sites and different contexts.

4.2 Living with Gods in Contested Lands

Turning toward religious practice in the northeastern Negev, analyses have traditionally focused on the juxtaposition of two deities – Yahweh and Qos – and on an assumed homogenous religious orientation of Judah toward Yahweh, and Edom toward Qos. In particular, the discovery of the religious shrine at Horvat Qitmit, with its iconic religious artifacts and numerous inscriptions referencing Qos, has worked to highlight religious differences between Judah and Edom in the northeastern Negev (Reference Beit-AriehBeit-Arieh 1995a). While these deities were indeed singularly prominent in these regions, and despite the analytical expedience of such contrasts, this type of approach tends to essentialize the complexities of religious traditions and ritual behavior, highlighting and promoting the idea of homogenous singularities between cultural groups. Moreover, from another angle, this approach tends to reflect more the Deuteronomic orthodoxy as presented in the Hebrew Bible, rather than explaining the complexities of relationships between different deities and their adherents in this region. An alternative approach centers the house as the primary locus of religious activity (Reference Smoak, Keimer and PierceSmoak 2022) and then considers the effect of larger regional shrines and external temples. In the following discussion, patterns of religious practice are examined first at the household level, and subsequently at increasing geographic and social scales. Structuring the analysis in this way avoids the expedience of contrasting large-scale “cultural” groups, focusing instead on the realia of everyday religious life and the effect that different social structures or elite agents may have upon it.

Recent scholarship on the archaeology of religion and the household has emphasized its embeddedness in everyday life, and the inability of modernist thought to necessarily understand its nuances (Reference Edwards, Díaz-Andreu, Lucy, Babić and EdwardsEdwards 2005; Reference Nakhai, Albertz, Nakhai, Olyan and SchmittNakhai 2014; Reference Zevit, Albertz, Nakhai, Olyan and SchmittZevit 2014). Deities and the supernatural were indeed an integral and real component of the landscape and communication with or reaction to them was not separate from everyday activities (Reference Stowers, Bodel and OlyanStowers 2008: 8–9). As such, many of these practices would likely not have been recognized by their performers as a separate class of “religious” activity, but rather merely the status quo of what life entailed when the divine was a constant in the physical world. It is difficult then, and perhaps inappropriate to a degree, to identify some material culture as “religious” and others as not since it all held the potential for use in religious activity. Nonetheless, in focusing on individual household structures, it is possible to highlight areas where ritual activity can be most easily identified according to modernist understandings, and to examine the associated material culture (Reference Daviau, Daviau, Wevers and WeiglDaviau 2001; Reference Bodel and OlyanBodel and Olyan 2008; Reference Albertz and SchmittAlbertz and Schmitt 2012; Reference Albertz, Nakhai, Olyan and SchmittAlbertz et al. 2014). The objects from such contexts most commonly identified with household religion include altars, stands, zoomorphic or anthropomorphic figurines (including Judean Pillar Figurines; hereafter JPFs), specialized vessels, as well as a variety of unmarked objects, or faunal remains, whose use would be dependent on local beliefs, ritual requirements, and the availability of resources (Reference Daviau and KamlahDaviau 2012; Reference Nakhai, Albertz, Nakhai, Olyan and SchmittNakhai 2014: 57).

At Tel ʿIra, for example, a variety of material culture traditionally associated with household religion was excavated, including two small incense altars and numerous figurines (Reference Beck and Beit-AriehBeck 1999; Reference Goldsmith, Ben-Dov, Kertesz and Beit-AriehGoldsmith et al. 1999: 469–470; Reference Kletter and Beit-AriehKletter 1999). One of the altars was excavated in a room of the casemate wall system (Room/Locus 191) that also contained a variety of domestic ceramics including seven bowls (including one BPW), three kraters, three cooking pots (types CP12 and CP14), two juglets, four lamps, two jars, a flask, and a JPF (Reference Freud and Beit-AriehFreud 1999: 262–263; Reference Kletter and Beit-AriehKletter 1999: 375, No. 4). The figurine assemblage is largely of the JPF variety, with additional zoomorphic figurines, and apart from several fill or surface finds, the remainder were excavated in domestic contexts. Several variant figurines were also excavated, one from an anthropomorphic vessel in a domestic context (Reference Kletter and Beit-AriehKletter 1999: 376, 384, No. 7), and another fragment of a female figurine that appears to have formerly been applied to a cultic stand-type vessel (Reference Kletter and Beit-AriehKletter 1999: 375, 384, No. 8). Somewhat divergent stylistically from this assemblage was an anthropomorphic plaque that bore both male and female genitalia and held a tambourine, whose closest stylistic parallels are at Horvat Qitmit, Tel Malhata, and in Transjordan (Reference Beck and Beit-AriehBeck 1999; Reference KletterKletter 1999: 375, Reference Kletter, Beit-Arieh and Freud2015: 548).

Nearby at Tel Masos, the late Iron Age contexts from Area G similarly emphasize the integration of ritual activity into everyday life. In Room 708, a context that appears linked with food preparation, the excavated material culture remains included four cooking pots (Type CP1), a bowl, juglets, a storejar, two lamps, and the base of a JPF (Reference Albertz and SchmittAlbertz and Schmitt 2012: 130–131). Additional figurine (JPF) fragments were found in areas where there were no indications of formal domestic shrines or ritual spaces, indicating that ritual activity was embedded in domestic life.

Shifting focus regionally to southern Jordan, the site of Tawilan provides some intriguing similarities. First, the areas of the site that have been excavated appear to indicate primarily domestic residences in an agropastoral subsistence-based village (Reference Bennett and BienkowskiBennett and Bienkowski 1995: 103–105). Moreover, the distribution of material culture remains relating to domestic activities is spread throughout the site, indicating that while some variances in the organization of household activities between different kin groups are likely, overall aspects of domestic life were relatively evenly patterned (Reference BrownBrown 2018b: 89–92). Figurine fragments, traditionally associated with ritual, were found in greater quantities in Area III to the west, an area also associated with food and textile production (Reference BrownBrown 2018b: 91, 190–192; Reference DanielsonDanielson 2020a: 307, 422–423). Two additional small altars were found in other domestic contexts. It is difficult to identify any specific deities or practices, or ways that material culture indicators of “religion” could be isolated to particular areas of the site. Rather, Tawilan demonstrates that interaction with the divine could not be separated from the everyday contexts, especially in cooking and textile production areas.

In the more culturally diverse towns of Tel ʿAroer and Tel Malhata, similar patterns are present. At Tel ʿAroer all the ritual material culture derives from domestic or nonformal ritual contexts, integrated with items of everyday use (Reference DanielsonDanielson 2020a: 299–300). As one example, in Locus 339 in a small room in Area D west, a JPF was excavated with a zoomorphic figurine, bowls (including BPW), a cooking pot (Type CP2), a seal with a bull head and two spatulas (Reference ThareaniThareani 2011: 86, Pl. 125). These contexts do well to illustrate the integration of ritual acts in the everyday acts of cooking and weaving, but also to highlight the entangled culinary traditions previously described. In the intramural context of Locus 1621 in Area H, a zoomorphic vessel, a fragment of a kernos, two rattles, a spatula, and a jar stopper were excavated. This context was interpreted as a rubbish depository, and based on the nature of the finds, may suggest they came from a more formalized ritual space in the domestic structure (Reference ThareaniThareani 2011: 107, Pl. 177).

Tel Malhata presents further insights. First, five stone altars were excavated in Pillared Building 1564 in Area H (Stratum IIIA), with three of the altars clustered together in the southwest corner of the room together with wooden furniture, bone inlays, and a 23 cm long phallus-shaped clay object (Reference Beit-Arieh, Beit-Arieh and FreudBeit-Arieh 2015; Reference Freud, Reshef, Beit-Arieh and FreudFreud and Reshef 2015: 591). The size of the structure and the nature and clustering of the finds suggest that this corner may have served a more formalized ritual purpose, perhaps associated with food preparation and the textile industry operating in the surrounding rooms, at a level above that of a single household (Reference Schmitt, Albertz, Nakhai, Olyan and SchmittSchmitt 2014: 270–271; Reference Koch and Sapir-HenKoch and Sapir-Hen 2018: 435). Regarding the remainder of the ritual artifact assemblage, particularly figurines, we find them present and integrated into the everyday contexts and activities of different households, though few JPFs have been excavated (Reference DanielsonDanielson 2020a: 304–306). What is particularly intriguing, however, is that while they depict elements that would be considered common in a “Judahite” assemblage (e.g., females, horses), they are created in a technological style that is more commonly seen in southern Jordan than the Negev. Other examples present additional stylistic features (“grotesque” style heads) or present females in the plaque style similar to Busayra, but with some unique features (Reference Sedman and BienkowskiSedman 2002: 375–376; Reference Kletter, Beit-Arieh and FreudKletter 2015: 572).

Ultimately, all of these contexts demonstrate the rootedness of ritual practice in everyday life and domestic contexts. In creating a typology of ritual spaces for the southern Levant, Rüdiger Schmitt has categorized these domestic spaces as of two types, one relating to religious practice that is wholly integrated with everyday activity, often in contexts of food or textile production, and a second, where a more formalized space in a house (corner, bench, rooftop, etc.) is the location of a cluster of specialized objects (Reference Schmitt, Albertz, Nakhai, Olyan and SchmittSchmitt 2014: 267–269). What all of these sites highlight, however, is that the human practices associated with household religion are highly similar across the broader region. While figurines may have been made with slightly different technologies or appear in slightly different styles (Reference Sedman and BienkowskiSedman 2002; Reference Kletter, Beit-Arieh and FreudKletter 2015), the function and use of these objects suggest similar practices. Such a similarity can be seen more broadly as well, where the categories of objects and their locations of use suggest similar patterns, despite the potential that they were oriented toward different deities or aligned with different ideologies.

The deities to whom this ritual activity in the household was directed are not entirely clear. Certainly, Yahweh and Qos were well known and perhaps revered in the household, but more compellingly, the prevalence of female figurines (e.g., JPFs), suggests potentially a physical model of a female deity, or more likely apotropaic practice or supplications toward a female deity (Reference KletterKletter 1996; Reference DarbyDarby 2014; Reference Dever, Albertz, Nakhai, Olyan and SchmittDever 2014). Often associated with Asherah, this common practice in the house likely centered on a goddess who may have also been identified as a consort of Yahweh or Qos. In viewing this data then, the strongest evidence of a deity in the material culture of the house features neither Yahweh nor Qos. Religious practice, however, did not cease at the household level. While the house was undoubtedly the most frequented religious space, larger, regional, and community ritual spaces played an important role, and it is in these spaces that starker differences in practice are identified.

At the community or regional scale, the site of Horvat Qitmit was the most prominent religious site in the northeastern Negev. Located in the south-central portion of the Beersheba-Arad Valley, it was visible to nearly every major site in the region. Horvat Qitmit consisted of two structure complexes and several stone enclosures. Complex A consisted of a rectangular structure at its northern end that contained three parallel rooms, each measuring approximately 2 × 4 m (Figure 20A). Each room contained a bench along its east wall, and at a right angle to the entrance of each room, a wall appeared to function as a podium or table (Reference Beit-AriehBeit-Arieh 1995a: 9–12). In the structure, substantial amounts of pottery and animal bones, as well as several figurines were excavated, with evidence that the rooms were periodically cleaned, perhaps making use of a favissa (Reference Beit-AriehBeit-Arieh 1995a: 12–13, 26). To the south was an open-air space with a raised stone platform, interpreted as a bāmāh, surrounded by a stone enclosure wall. To the east of this delineated space was a poorly preserved small wall surrounding a flint-slab-topped altar, a round basin, and a small pit (Reference Beit-AriehBeit-Arieh 1995a: 13–20).

Figure 20 Horvat Qitmit: (A) reconstruction of Complex A; (B) reconstruction of Complex B; (C) head of female deity; (D) anthropomorphic male figures

(figures after Reference Beit-AriehBeit-Arieh 1995a: Fig. 9.1, 9.2; Reference Beck and Beit-AriehBeck 1995: figs. 3.17, 3.19, 3.22, 3.23, 3.53, 3.54)

Complex B at Horvat Qitmit was located approximately 15 m to the north of Complex A and consisted of a square-shaped structure measuring 8 × 8.5 m (Figure 20B). Several rooms were located on the west, north, and northeast sides, surrounding an open-air courtyard. A small standing stone, interpreted as a maṣṣēbah was excavated here, with significant evidence in the complex for food preparation (Reference Beit-AriehBeit-Arieh 1995a: 20–24). At Horvat Qitmit, Type CP4 cooking pots accounted for the overwhelming majority of types, and significant amounts of BPW were similarly excavated. Two additional stone enclosures, consisting of delineated elliptical spaces with only a single surviving course of stones in its wall, were also uncovered. Their purpose is not entirely clear, but they may have functioned as animal enclosures or served an unknown ritual function (Reference Beit-AriehBeit-Arieh 1995a: 24–26).

Horvat Qitmit is best known, however, for its iconic statuary and its association with the god Qos. Numerous fragmentary inscriptions preserve the name Qos, suggesting this deity to have been an important focus at the site, though not the sole deity (Reference Beit-Arieh and Beit-AriehBeit-Arieh 1995b). The head of a horned goddess was also excavated and was likely a prominently featured deity (Figure 20C). Further, the tripartite division of Complex A suggests that a third deity may also have been present. Numerous other anthropomorphic figures likely representing priests or worshippers were excavated (Figure 20D), together with male, female, and zoomorphic figurines, as well as a sphynx, musical instruments, architectural stands with human, floral, and pomegranate decorative features (Reference Beck and Beit-AriehBeck 1995). Notably, the assemblage appears divergent from what has been excavated elsewhere in Judah, with stronger influences associated with central and southern Jordan. Together with the association of Qos to southern Jordan, Horvat Qitmit has been compellingly argued as related to the cultural and religious practices common in Edom and particularly among the elite of Edom.

Due to its location along the east-west road, Horvat Qitmit has long been interpreted as a wayside shrine for diverse groups moving through the region (Reference FinkelsteinFinkelstein 1992). However, petrographic and INAA studies on material from Qitmit have also identified a local signature for ceramic production (Reference Gunneweg and MommsenGunneweg and Mommsen 1995; Reference Freud and TebesFreud 2014). In other words, the material culture was produced in the environs of the northeastern Negev, and thus the site was also frequented by people local to the region. Similarly, further research on the material culture from Tel Malhata has identified strong similarities that indicate a close relationship between the two nearby sites, beginning by at least the early seventh century BCE (Tel Malhata Stratum IIIB; Reference Beit-Arieh, Freud, Tal, Beit-Arieh and FreudBeit-Arieh, Freud and Tal 2015: 742). In this way, Horvat Qitmit must be viewed as an inherently local feature of the religious landscape, and at home in the northeastern Negev. The fact that Horvat Qitmit stood in this region for approximately a century without evidence of hostility or even destruction at the time of its abandonment, demonstrates that the site was not viewed in an overly negative way by the inhabitants of the region, or presumably the Judahite administration whose fortresses were scattered throughout the valley.

On the highway to the east, at the northern end of the Arabah Valley and adjacent to the fort at ʿEn Hazeva, is an additional space of broader community religious significance. The religious space is unfortunately poorly preserved and yet awaits its final publication, but appears to consist of a small rectilinear U-shaped structure interpreted as a shrine (Reference Cohen and YisraelCohen and Yisrael 1995, Reference Cohen and Yisrael1996). From a nearby favissa dug near the fortress walls, a significant number of iconic ritual artifacts were excavated, including anthropomorphic and non-anthropomorphic stands, altars, pedestalled bowl incense burners, chalices, perforated tripod cup incense burners, incense shovels, bowls, and pomegranates (Reference Cohen and YisraelCohen and Yisrael 1995: 224–225). While no deity can be explicitly identified from the assemblage, the similarity of the artifacts to Horvat Qitmit and southern Jordan, with the discovery of an incised goring bull suggesting a weather deity, has led to the suggestion that here too, Qos was the primary deity of focus (Reference BeckBeck 1996).

It would be tempting to contrast the apparent singularity of Horvat Qitmit in the northeastern Negev (and nearby ʿEn Hazeva), with the famed temple located at Tel Arad. There, a small tripartite temple measuring 12 × 18 m was excavated (Reference HerzogHerzog 2002), whose character and association with the Judahite administration compellingly suggest Yahweh was the deity of focus. However, two factors prevent the religious practices at the Tel Arad temple from being compared to those at Horvat Qitmit. First, and most glaringly, is the fact that the small Tel Arad temple was decommissioned in the late eighth century BCE, prior to the construction of Horvat Qitmit (Reference HerzogHerzog 2002: 14, 49–50). Thus, the temple at Tel Arad and the sanctuary at Horvat Qitmit did not exist at the same time. Second, is the fact that as an intramural temple in a military fort, its public accessibility is in question. Similarly, a larger, formal religious space has been determined to have existed at Tel Beersheba, though again, this space was decommissioned prior to the end of the eighth century BCE, and thus was not contemporary with Horvat Qitmit (Reference ZevitZevit 2001: 171, 301–302).

Beyond, we are confronted with the larger state-promoted religious centers of Judah and Edom that were both nearly equidistant from the northeastern Negev. At Jerusalem was a large temple linked to Yahweh (and a consort?) that was constructed and promoted by the ruling elite (Reference Bloch-Smith and GittlenBloch-Smith 2002; Reference Lemaire, Finkelstein and Na’amanLemaire 2011). Similarly, at Busayra, adjacent to the palatial center was a large temple complex oriented toward Qos (and a consort?),Footnote ¹⁶ with its close proximity to the palace demonstrating its entangled relationship with the ruling elite, similar to Jerusalem (Reference BienkowskiBienkowski 2002a: 94–95; Reference PorterPorter 2004: 381–387). It is difficult, however, to gauge the impact that these somewhat distant (≈ 70 km) “state” religious centers had on the northeastern Negev. The relation between household ritual and larger state sanctuaries has been argued to have been somewhat mutually influential (Reference Albertz and SchmittAlbertz and Schmitt 2012: 55). However, the state and its sanctioned religion can also function as “top-down” movement that can push for particular patterns of religious practice, as seen for example in the promotion of the Deuteronomistic ideals (Deuteronomy 12; 2 Kings 22–23). Thus, through conformity and centralization, such promoted religion can benefit the elite through increased sociopolitical control and wealth creation (Reference Meyers and OlyanMeyers 2012: 166–168).

A further mode by which to track the popularity of the state-promoted deities is through naming practices and the custom of adding part of a deity name – a theophoric element – to one’s own. For Yahweh, this involves the addition of יהו (yhw), and for Qos, קוס (qws). Previous scholarship has established the popularity of Yahwistic elements in Judah, and Qos elements with Edom, which are well reflected in the naming practices of their kings in the late Iron Age (e.g., ḥzqyhw, ṣdqyhw of Judah vs. qwsmlk and qwsgbr of Edom; Reference BartlettBartlett 1989: 200–207; Reference GolubGolub 2014, Reference Golub2017; Reference SandersSanders 2015; Reference DanielsonDanielson 2020b, Reference Danielson, Hensel, Zvi and Edelman2022: 137–140). Such names cannot immediately be taken to indicate the religious preferences or practices of individuals, but they do identify trends, and when clustered together spatially and chronologically, can yield insights into the ideologies of different communities (Reference Zadok and MeyersZadok 1997; Reference Aldrin and HoughAldrin 2016; Reference Nyström and HoughNyström 2016). Owing to the rich inscriptional record of the northeastern Negev, it is possible to examine onomastic trends by site and diachronically, while noting that this dataset is both androcentric and reflective of the Judahite administration.

Several trends become apparent when compiling the onomastic data from the northeastern Negev and Edom (Figure 21A).Footnote ¹⁷ First, while Yahwistic and Qos names are dominant among the Judahite and Edomite kings respectively, they are not ubiquitous. Nevertheless, the general prominence of these deities in these regions is evident. In fact, Qos names are not at all well-attested at Judahite administrative sites (e.g., Horvat ʿUza; Tel Arad, Tel ʿIra). Nor are Yahwistic names well attested at southern Jordanian or Edomite-administered sites (e.g., Tall al-Khalayfi). Moreover, little overlap between Yahwistic and Qos names is seen at any site, with the only reference to Qos names at Tel Arad referencing persons from outside Arad.Footnote ¹⁸ Among the general corpus, numerous references to El (אל) are also present, although this deity name could be used in its generic sense (“god”) and thus relate to either Yahweh or Qos. With regard to the major “towns” in the region, however, this pattern is not emulated. Intriguingly, Tel Malhata, the site with perhaps the greatest diversity seen in the previous discussion, bears no evidence for Yahwistic or Qos names, though the dataset is limited (n=12). Similarly, at Tel ʿAroer, there is evidence for a Qos name though the dataset is even more limited.

Figure 21 (A) Percentage of onomastic types by site in the northeastern Negev and Edom; and (B) multigenerational naming practices at Horvat ʿUza and Tel Arad

While the overall dataset is skewed toward the military administrative forts of Horvat ʿUza and Tel Arad, these two sites nonetheless evidence a strong association with Yahwistic names. Moreover, through this richer dataset it is possible to examine patterns in naming practices over multiple generations as many of the administrative lists excavated there preserve the patronyms of individuals. Thus, focusing on the names that identify two generations, we can detect a relative degree of stability in Yahwistic names at Arad, while at Horvat ʿUza there is a rapid increase in Yahwistic names in the second generation (Figure 21B). It is possible that the rapid increase in Yahwistic names at Horvat ʿUza relates to the fort’s history. While Tel Arad had existed in the northeastern Negev for centuries, Horvat ʿUza was not established until the seventh century BCE (Table 1; Reference HerzogHerzog 2002: 14; Reference Beit-AriehBeit-Arieh 2007b). As such, it would have required a large population influx in order to build and garrison the fort. It is possible that this influx of persons held ties to the Jerusalem region, where naming practices may have been patterned after Yahweh more rigidly than in the northeastern Negev.

To summarize, the preceding discussion has highlighted the complexities of describing religious practices that can be evidenced at a variety of social scales and through different datasets. While there are indeed differences that determine certain communities to have been more closely aligned with Yahweh and others with Qos, as suggested by the onomastic data, this is only part of the story. On the basis of household data, despite slight differences in the stylistic or technological manufacture of figurines, the practices and use of these objects in domestic contexts appear to have been shared throughout the region. It is not possible to identify specific deities of focus in such a setting, other than the preponderance of female figurines, which, if indeed linked with a female deity through their apotropaic function, could be aligned with either Yahweh or Qos as an association with a consort is likely for both. This is not to say that domestic religion was necessarily identical throughout southern Jordan and the northeastern Negev, but rather that the rituals and practices would likely have been recognizable and cross-culturally understood. In fact, if one is to delve into textual and inscriptional data, it appears likely that numerous ritual customs were shared throughout Judah and Edom, seemingly including male circumcision (Jeremiah 9:25–26) and the offering of unleavened bread (Reference AhituvAhituv 2008: 354).

The commonalities of practice amidst the prominence of different deities in different regions likely relate to the rootedness of the gods in the landscape. In other words, certain gods were more closely linked with particular landscapes, and perhaps variant manifestations of the same deity could be identified at various locales in that landscape. Thus, transitioning from a focus on one god to another between different geographic or social settings, may not have been particularly remarkable to ancient persons migrating to a new context or marrying into a new social group (Reference McCarter, Miller, Hanson and McBrideMcCarter 1987: 140–141; Reference SmithSmith 2016: 91–92; Reference MillerMiller 2021: 88). Such examples are known in migration contexts in the ancient world, where syncretism will occur between a traditional god and those dominant in a new context (Reference WinnickiWinnicki 2009: 300, 303; Reference DanielsonDanielson 2020b: 162–170). The question remains then, as to what degree this household practice reflected, and/or influenced larger patterns of collective religious activity (Reference Albertz and SchmittAlbertz and Schmitt 2012: 55; Reference SandersSanders 2015)

Starker differences do appear at broader social scales. In the northeastern Negev during the seventh and sixth centuries BCE, Horvat Qitmit was dominant, in part due to its central location, but also due to the fact that there were no other regional shrines or temples with which to compete. Its focus on the deity Qos, and at least one other unnamed female deity, would have been common knowledge in the region. The culinary and other material culture evidence from Horvat Qitmit suggests that the shrine was not frequented by all social groups of the northeastern Negev. And yet, its existence does not appear to have been overly offensive to others, as, despite serving a minority of the population, the shrine was not damaged or destroyed during the approximate century of its use. This may very likely indicate that Qos was at home in this landscape. Even if one’s family ideology was more aligned with the Deuteronomistic ideals that promoted Yahweh in Jerusalem, it is unlikely that many of the individuals or families in the region frequented or even saw the temple in Jerusalem. Inadvertently, they would be far more visually familiar with the site of Horvat Qitmit upon its visible hill than the temple in Jerusalem. In this way, Horvat Qitmit and Qos may not have been viewed as foreign or intrusive, but merely different.

4.3 Foreign Accents and Scribal Power

Language and dialect differences form a significant and readily identifiable feature by which humans identify others as belonging to a specific community or as distinct from it. By all accounts, the language or dialect spoken in Edom was mutually intelligible with its immediate surrounding regions: Moab to the north, and Judah to the northwest. Isolating and determining linguistic variances between these regions is notoriously difficult. This is due in part to the methodical challenges of solely using written inscriptions since we cannot access speakers from these ancient speech communities, but also because we have so few excavated inscriptions from Edom. Nonetheless, certain features from the limited inscriptional corpus do allow for a view into the speech practices in Edom and their relation to Judah, revealing a close similarity between them.Footnote ¹⁹

For example, inscriptions associated with Edom make use of the definite article hê- (ה), which together with the use of the lexeme bn (בן), places this dialect as distinct from Aramaic and in the family tree of Canaanite. Further, the use of the relative pronoun ʾšr (אשר), and the accusative marker ʾt (את) highlights a similarity with the neighboring languages spoken in Judah and Moab (Reference GarrGarr 1985: 228–230; Reference Vanderhooft and EdelmanVanderhooft 1995; Reference Rollston, Levy, Najjar and Ben-YosefRollston 2014). Significantly, the present dataset from Edom does not preserve any identifiable phonological features that would mark speech in Edom as overly distinctive from neighboring regions. Thus, on the basis of current data, it is possible to state that the languages spoken in both ancient Judah and ancient Edom were mutually intelligible, and are likely best described as variant dialects of a broader “southern Levantine” language (Reference GarrGarr 1985: 228–230; Reference Segert and HetzronSegert 1997).

Identifying broad similarities in the spoken language, however, is not meant to indicate that differences would have been entirely absent, nor were they unnoticed. Humans possess the unique ability to identify even minute variances in speech or writing, using this to determine closeness with or distance from others (Reference Jaffe, Androutsopoulos, Sebba and JohnsonJaffe et al. 2012; Reference Sebba, Mahootian and JonssonSebba et al. 2012). As the Negev and southern Jordan relate to distinct topographical zones, whose inhabitants were predominantly engaged in subsistence practices, it is to be expected that regional pronunciations and phonological stresses, not to mention slang or slurs, would develop and perpetuate. This has been well documented in modern historical contexts (Reference NielsenNielsen 2005; Reference Weldon, Reaser, Wilbanks, Wojcik and WolframWeldon 2018), and is also identifiable for the ancient world (Reference GarrGarr 1985; Reference Colvin and BakkerColvin 2010). Such features would not necessarily be captured in writing, but would very much form a significant component in signaling distinctions between groups. Associating this principle with the southern Levant then, an individual from Tel ʿIra or Horvat Tov would likely be readily able to recognize differences in the speech of someone from Ghrareh or Tawilan, notwithstanding other highly indexical visual cues like dress or hairstyle, which the statuary from Horvat Qitmit certainly suggests.

The oft-referenced shibboleth narrative from the Hebrew Bible well illustrates this principle of identifying differences on the basis of speech:

The text, in a gruesome manner, highlights the ease with which people can identify variances in speech, and notably the differences recounted here would not be preserved in the Iron Age writing system. Moreover, as this text derives from the first millennium BCE southern Levant, it not only preserves the idea of the ability to identify others on the basis of speech, but through its writing/recitation in its ancient context would have served to reinforce such stereotypes. In this instance, a rather minute and superficial marker of difference was, through narrative, emphasized as the means by which to identify and then enact violence on a neighboring group.

While it can be difficult to detect variances in speech in ancient written sources, certain inscriptions from the northeastern Negev do allow for such insight. For example, in the administrative texts of the Judahite garrison at Tel Arad, and attested elsewhere in Judahite administrative contexts, the phrase: “I bless you by Yhwh” (ברכתך ליהוה) is common (Reference AharoniAharoni 1981: nos. 16, 21, 40). In contrast, from the corpus at Horvat ʿUza, in an inscription identified as relating to Edom, the greeting/blessing formula is slightly different: “I bless you by Qws” (הברכתך לקוס; Reference Beit-Arieh and Beit-AriehBeit-Arieh 2007a: No. 7). While at a certain level contrasts may be drawn between the deities that are the focus of the blessing, deeper differences exist at the linguistic level. English translations are unable to capture the nuance, but in the former, the verb in rendered in the D stem (piʿēl), whereas the latter uses the causative H stem (hip‘îl). Differences here then, are noted not only in the spelling and nuance of the verb, but also in the pronunciation. Thus, in these mutually intelligible greeting formulae, while the variant deities draw a contrast, these differences are accentuated in the variant pronunciations and subtle nuances in the meaning of the greeting/blessing verbs. It is worth emphasizing that this inscription was excavated at the site of Horvat ʿUza, in the midst of these cross-cultural interactions.

The emphasis on regionalisms and their preservation of nuance can also be extended to script. Just as “Edomite” and “Judahite” are problematic descriptions for a language that was mutually comprehensible, so too understandings of script differences as relating to “national” scripts likely overstate the situation. Certain differences in the script do exist. Inscriptions associated with the region of Edom show strong similarities to Moab and Ammon, bearing sufficient similarity to the contemporary Aramaic script to suggest a lack of early scribal influence from Judah (Reference Vanderhooft and EdelmanVanderhooft 1995; Reference Rollston, Levy, Najjar and Ben-YosefRollston 2014).Footnote ²⁰ In particular, we can highlight the paleographic features of the letters bêt, wāw, sāmek, ʿayin, qôp, rêš, and tāw that show an affinity with Aramaic, and the dālet, hê, and alternate sāmek that appear to be unique to scribal practices in Edom (Reference Vanderhooft and EdelmanVanderhooft 1995; Reference Rollston, Levy, Najjar and Ben-YosefRollston 2014).Footnote ²¹

While such regional variances are identifiable, the extremely low literacy rates of the ancient world, combined with the challenges of understanding these kingdoms as “nations,” indicates the labeling of these differences as “national scripts” to be imprecise. Rather, as the overwhelming majority of writing took place under the aegis of political and/or religious institutions by a limited number of scribes, scribal differences instead seem to be the result of different training that was perpetuated through generations of scribes (Reference SchniedewindSchniedewind 2013: 100–115, 117–120, Reference Schniedewind2019). While scribal variations were indeed identifiable, they would have been recognized most readily only by a limited number of literate persons, namely those associated with the political administration.

In summary, the differences in speech between Judah and Edom that intersected in the northeastern Negev appear to have been largely superficial with the language mutually intelligible. And yet, differences would have been readily identifiable: in pronunciations, phonological stresses, and to a degree, formalized in standard greetings and blessings. Whether the outcome of determining these variances in speech was benevolence, ambiguity, or hostility would certainly vary by context, but the identification of such differences melds well with the overall established tapestry of cross-cultural interactions and social entanglements in the settlements of the northeastern Negev.

5.1 Legendary Patriarchs

The fraternal language used to describe Edom in the biblical text may relate in part to similar cultural practices, but more so as a result of the etiological narratives in Genesis. Here, while Judah’s other neighbors, Ammon and Moab, are described as distantly related (first cousins once removed), with crude, incestuous false etymologies provided (Genesis 19:30–38), Edom is described as much more closely related – as a brother (e.g., Genesis 25:19–34; 27; 35:29; Numbers 20:14; Deuteronomy 2:4, 8; 23:7; Amos 1:11; Obadiah 1:10, 12; Malachi 1:2–4; Reference AssisAssis 2006: 9). These kinship terms are exemplified in the patriarchal narratives where the legendary figures of Jacob and Esau – twin brothers – are cast as the ancestors of Israel/Judah and Edom respectively (Genesis 25, 27, 32–33; Reference Blum, Evans, Lohr and PetersenBlum 2012). In this portrayal, as often with kin, it is the relationship that takes precedence over grievances and wrongdoing.

The general trajectory of the Jacob-Esau narrative begins with the younger twin, Jacob, duping his elder brother Esau into selling his “birthright” for a pot of stew (as described previously in Section 4.1). With the help of his mother Rebekah, Jacob then deceives his father Isaac into giving him said birthright (Genesis 25, 27), seeming to begin to fulfill the prophesy surrounding his and Esau’s birth, that Jacob would have dominance over his elder brother (Genesis 25:23). After fleeing Esau’s fury, Jacob, after a period of time and yet in fear, returns with his family to make amends with Esau. Ultimately, Esau greets him warmly and their fraternal bonds are re-established (Genesis 32–33; Reference Blum, Evans, Lohr and PetersenBlum 2012). Throughout the narrative, both explicit and implicit references work to identify Jacob as the ancestor of Israel/Judah, and Esau as the ancestor of Edom. These references include the folk etymology of Esau’s name as “hairy,” or relating to the “hairy” or forested landscape of Seir in Edom, and numerous insinuations of Esau as “red” and eating “red stew.” Here, the emphasis on red corresponds to the meaning of Edom as “red,” and associates Esau with the iconic red sandstone formations of southwestern Jordan (Reference Bartlett and FreedmanBartlett 1992; Reference Knauf and FreedmanKnauf 1992a). While the folk etymology of Esau as “hairy” (Genesis 25:24–26) is not linguistically correct (Reference HendelHendel 1987: 111), it nonetheless serves to situate him in a landscape that would be moderately familiar to the reader. Such etymological associations follow a similar Hebrew Bible and broader ancient Near Eastern practice of connecting epic characters to social groupings, tribes, or polities (Reference HendelHendel 1987: 113–115).

The narrative trajectory of the broader Jacob Cycle, however, likely already represents the weaving together of several traditions. Namely, the Jacob-Esau narratives (Genesis 25–27) likely stood as an independent tradition from the Jacob-Laban narratives (Genesis 29–31), one oriented toward the south, and the other to the north (Reference Hensel and HenselHensel 2021b). These differences and a complex redaction history are evidenced by their variant structure, geographic locations, and contradictions between the texts (Reference Schmid and HenselSchmid 2021; Reference Wöhrle and HenselWöhrle 2021). On the basis of this complexity, a fruitful way to consider these narratives, and of the ways by which meaning may be drawn from them, is through the lens of cultural memory, following Reference Halbwachs and CoserHalbwach’s (1992) ideas of collective memory (Reference Hendel and HendelHendel 2010). By highlighting the ideas of cultural memory, we can reconcile the deep past setting of the patriarchal narratives with their composition and redactions in the mid first millennium BCE (Reference Finkelstein and RömerFinkelstein and Römer 2014; Reference HenselHensel 2021c). In doing so, we can engage with the complex relation between history and cultural memory, and as Hendel argues, consider these texts as a “palimpsest of past and present” (Reference HendelHendel 2005: 45–47).

In this way, knowledge of, and interactions with a neighboring polity and people are cast in a deep time perspective (Reference Hensel and HenselHensel 2021a). As such, the association of Jacob and Esau with the polities of Israel/Judah and Edom is likely a later phenomenon in the development of the tradition, and unsurprisingly, ought not to be used to influence a history between them (Reference HendelHendel 1987: 114–115; Reference CrowellCrowell 2021: 177–264).Footnote ²³ To illustrate, numerous characteristics of Esau stand contrary to other characteristics associated with Edom. Whereas Esau is depicted as lacking in intelligence, contrasted with the cunning of Jacob (Genesis 25, 27), other traditions associate Edom with wisdom (Jeremiah 49:7; Obadiah 1:8; Baruch 3:22–23; Job; Reference HendelHendel 1987: 114–115). Similar, later emendations to the text may perhaps be seen in Isaac’s proclamation of Esau/Edom’s subservience to Jacob/Israel/Judah (Genesis 27:40), which contradicts the remainder of the narrative in which Jacob shows willing submission to Esau, and in the intimacy of their reunion (Genesis 32–33; Reference FlemingFleming 2012: 84; Reference CrowellCrowell 2021: 224–228).

Despite being cast as a coarse and somewhat abrasive figure (Reference ArnoldArnold 2009: 232–233), Esau is portrayed in sympathetic terms, a feature that further belies interpreting Esau as Jacob’s enemy. Likewise, it is not Esau who engages in trickery and theft, but Jacob, creating something of a moral issue for later biblical commentaries (Reference KrauseKrause 2008). Even with the humiliating events of Genesis 27, a simple contrast between Esau and the figure of Laban, representing Aram (Genesis 29–31), together with Esau’s reacceptance of Jacob, places Esau in far higher regard (Reference FlemingFleming 2012: 83–85). In this fashion, an alternative way of looking at this text would be through the lens of the narrative “hero” (Jacob) and the uncultured “other” (Esau), a portrayal frequent in Near Eastern literature that emphasizes one figure as a literary foil to the other (Reference HendelHendel 1987: 101–131; Reference HamoriHamori 2011: 633–636). The contrast between the two, with the later equation between Esau and Edom through various physical (red, hairy, hunter) and personality (coarse, abrasive, intellectually dim) characteristics, would serve to evoke in a Judahite audience a certain image of their eastern regional neighbor (Reference ArnoldArnold 2009: 232–233).

These associations may be described as a pattern of “othering,” defined as the “process which serves to mark and name those thought to be different from onself” – a key feature of identity formation processes (Reference WeisWeis 1995: 18). Consequently, these contrasts of the physical and intellectual characteristics of Jacob and Esau would serve to reify belongingness to communities who identified as Jacob’s descendants in contrast with those associated with the figure of Esau, who at least in the surviving version of the narrative stand as metaphors for Israel/Judah and Edom. In all, the narratives portray a highly entangled relationship between Jacob/Judah/Israel and Esau/Edom, one of competition and conflict, but also closeness and reconciliation, standing as a unique metaphor to exemplify the multifaceted pattern of cross-cultural interaction in the northeastern Negev (Reference TebesTebes 2006b; Reference Hensel and HenselHensel 2021a; Reference Danielson, Hensel, Zvi and EdelmanDanielson 2022).

5.2 Entangled and Forgotten Gods

In light of such complexities of cultural memory interwoven in layers of etiology, it is highly intriguing that the Hebrew Bible does not once mention Qos, the foremost deity of Edom. In fact, regarding any specific deities of Edom, the biblical text is silent.Footnote ²⁴ The only possible intimations of the existence of the neighboring deity Qos are hidden as a theophoric element in the name barqôs (ברקוס), found amidst a list of Persian period returned exiles decades after the kingdom of Edom had faded from the Levantine political stage (Ezra 2:53; Nehemiah 7:55). This silence regarding Qos is all the more striking as numerous passages characterize neighboring peoples in relation to major deities worshipped in their regions. For example, in the curses in Jeremiah 48–49, the god Kemosh is identified with Moab, and immediately afterward, Milkom with Ammon. Yet, in the same chapter, amidst a lengthy and hostile curse against Edom, no gods are mentioned. The absence of Qos stands amidst the numerous narratives and references to Edom in the biblical text, as well as the extensive cross-cultural interaction identified in the northeastern Negev.

A simple lack of knowledge of Qos on the part of the biblical writers is an insufficient explanation (Reference BartlettBartlett 1989: 195), as the shrine at Horvat Qitmit with its focus on this deity was located among settlements and forts claimed and administered by Judah. In fact, even in the epistolary of the Judahite administrators, two references to Qos are made (Reference AharoniAharoni 1981: nos. 12, 26). These inscriptions are poorly preserved, and these references appear to be solely theophoric, but the fact remains that this god was known to the people of Judah, and is attested in the administrative writings of Judahite elites. Even if a very late date for the composition of the Hebrew Bible is presumed, the inscriptional finds from Persian and Hellenistic period Idumea only indicate that the popularity and prominence of Qos had grown. Therefore, the exclusion of Qos from the text appears to have been intentional.

To further complicate the picture, scholars have long-noted that in many of the oldest texts in the Hebrew Bible (Reference SchniedewindSchniedewind 2013: 51–72), it is in fact Yahweh who is most closely associated with the region of Edom and the surrounding deserts. For example, Judges 5:4 states regarding Yahweh: “when you went out from Seir, when you marched from the region of Edom.” Similarly, Deuteronomy 33:2, describes Yahweh as having “dawned from Seir” while also connecting him with the Sinai and Paran, other neighboring desert regions. Further, from the desert site of Kuntillet ʿAjrud in northeastern Sinai, inscriptions were excavated that mention a “Yahweh of Teman and his Asherah,” identifying Yahweh with Teman (Reference Ahituv, Eshel, Meshel and MeshelAhituv et al. 2012: 95–98, 105), a location in the environs of Edom (Reference Knauf and FreedmanKnauf 1992b). This affiliation echoes the likely archaic text of Habakkuk 3:3 that states “God came from Teman.” Beyond associations with Jerusalem and locales to the north, these references, and particularly the Kuntillet ʿAjrud inscriptions, indicate that traditions of Yahweh, or at least certain manifestations of Yahweh, were understood as relating to desert locales to the south and east of Judah.Footnote ²⁵

Outside of the biblical text, there are numerous inscriptional references that link Qos to the landscape of Edom and the surrounding deserts (Reference BartlettBartlett 1989; Reference Dearman and EdelmanDearman 1995; Reference Knauf, van der Toorn, Becking and van der HorstKnauf 1999; Reference KelleyKelley 2009; Reference DanielsonDanielson 2020b). While there is little direct iconographic evidence to reveal the nature of Qos, on the basis of associated imagery at sites such as Horvat Qitmit, Qos appears to have functioned as a type of Iron Age weather deity and/or divine warrior (Reference BartlettBartlett 1989: 200–204; Reference Beck and Beit-AriehBeck 1995: 187–190, Reference Beck1996: 107–109; Reference Dearman and EdelmanDearman 1995; Reference Knauf, van der Toorn, Becking and van der HorstKnauf 1999: 377; Reference KelleyKelley 2009: 260–261; Reference DanielsonDanielson 2020b: 115–116). Etymologically, the name Qos (קוס) appears to be synonymous to the Arabic term for bow – qaws (قوس) – perhaps sharing an earlier proto-Semitic root, with the West Semitic derivation of bow as qšt (קשׁת; Reference Knauf, van der Toorn, Becking and van der HorstKnauf 1999: 676).Footnote ²⁶ This meaning of “bow” in association with Qos, has been interpreted as further substantiating the idea of Qos as a weather deity/divine warrior, who are frequently depicted with bows.

Combined with the potential origins of Yahweh in Edom, the conspicuous omission of Qos from the biblical text has led some scholars to propose the silence to be an inadvertent admission of a shared religious past between Judah and Edom. In such interpretations, Yahweh and Qos are suggested to have been derived from an earlier El deity, with the resulting differences understood as alternate understandings promoted by different communities (Reference Dearman and EdelmanDearman 1995: 126–127). Alternative theories have Qos functioning as an epithet of Yahweh (Reference AmzallagAmzallag 2009: 392; Reference KelleyKelley 2009: 265), or even as the divinized bow (קשׁת) of Yahweh (Reference KelleyKelley 2009: 266; Habakkuk 3:9; Psalm 7:12–13; Genesis 9:13–16).

While this relationship and questions of origins remain unclear, with the dataset heavily biased toward a Yahwistic perspective, the entangled relations between these peoples and their gods exemplify the multifaceted, fraternal, and competitive nature of the biblical text’s allusions to, and etiologies associated with this region. Whatever the motives the biblical writers had toward silencing Qos, the archaeology of Edom and the Negev has placed Qos in the center of the stage, necessitating more nuanced interpretations of the religious heritage of the northeastern Negev.