Our sincere gratitude goes to all who have contributed to this study. We particularly appreciate the Department of Antiquities (DoA) of Jordan and EJAP, whose support was vital to our research. We sincerely thank the Bedouins of the Al-Amarin and Al-Jafr tribes, especially Dhalish. His Bedouin wisdom and generosity greatly enriched our desert stay and made our research journey meaningful beyond archaeology. Finally, we thank Zuzana Halgašová for her diligent English editing. However, we accept responsibility for any remaining errors.

Introduction

1The Early PPNB period was enigmatic, even within the extensively researched areas of the southern Levant (Gopher 1996: 151). Its presence in the Mediterranean and Transjordanian Highlands was closely examined (Kuijt and Goring-Morris 2002: 384). The debate revolved around the question of whether the EPPNB in the southern Levant embodied a specific chronological epoch or a distinct cultural phenomenon. However, with the discovery of EPPNB sites such as Aswad (Stordeur 2003; Stordeur et al. 2010), and Tell Qarassa (Ibáñez et al. 2010, 2014) in southern Syria; Kfar Ha-Horesh (Turros and Goring-Morris 2011) and Motza (Khalaily et al. 2007) in Israel; Beidha, Shakrat Mseid, Ail 4 (Purschwitz 2017), Wadi Mushash 163 (Rokkita-Krumnow 2019), Jilat 7 (Garrard et al. 1994a), Hurrat-Juhayra 202 (Fujii et al. 2019), and Jebel Queisa (Henry 1995) in Jordan; along with Jebel-Qatar 101 in northwestern Arabia (Crassard et al. 2013), the southern Levant is now understood to have been a dynamic interaction zone during the EPPNB period, featuring connections with the northern EPPNB expressions and cultural complexes (Gopher this volume).

2Traditionally, scholars now perceive the EPPNB material culture in the southern Levant as a reflection of the assimilation of cultural traditions and novel technological characteristics from the northern Levant (Edwards 2016: 69). This cultural transfer occurred through the Levantine corridor during the first half of the 8^th millennium BC. Further, it radiated eastward to Central Jordan, the Jordanian steppe, and the deserts during the latter half of the same millennium. This dissemination, as demonstrated by Goring-Morris and Belfer-Cohen (2020), was neither linear nor straightforward, nor was it “revolutionary”. It encapsulated various elements of continuity from the southern Levant PPNA and the antecedent Natufian traditions while introducing unique PPNB characteristics from the northern Levant. Among the multitude of innovations like architecture and symbolism, the incorporation of bidirectional blade technology and facets of craft specialisation (Abbès 2003) stood out as a significant influence from the north. Nonetheless, the southern Levant EPPNB, addressed as the Aswadian in this volume, is speculated to have primarily persisted in foraging (Gopher this volume).

3In the southern Levant’s EPPNB context, the case of `Ainab 1 offers a captivating example. Located approximately 75 km SE of al-Jafr in the southeastern Badia of Jordan, not far from the Saudi Arabian border (fig. 1). `Ainab 1 consists of a group of five large structures built with large upright tabular ashlars, forming round and curvilinear rooms that appear to merge. H. G. K. Gebel discovered this site during an inspection tour of the Eastern Jafr Archaeological Project (EJAP); numerous burins and indications of the bidirectional core technique were identified in the surface collection in 2013. In the summer of 2014, structure A of `Ainab 1 was excavated by a small team consisting of B. Kubíková, C. Purschwitz, D. Štefanisko and H. G. K. Gebel as part of the EJAP, with support from ex oriente, Berlin.

4The construction of structure A (figs. 1, 2a) is situated atop an elevated portion of a banked sandstone outcrop. Its deflated stratigraphy denotes both geomorphic and anthropogenic activities above the bedrock. The silt-sand deposits delineating the occupation layers suggest an episodic site utilisation. The building blocks for this structure, in the form of ashlars and stones, were either sourced from the nearby exposed bedrock or collected from eroded surface material. Structure A comprises multi-roomed polygonal or circular rooms buttressed by wall foundations. Notably, scattered around the structure were numerous small structural elements embedded in the Hamad surfaces, hinting at household installations and potential game-processing activities. Despite the scarcity of organic matter in the stratigraphy, it is hypothesised that migratory ungulates provided the impetus for establishing this hunting camp.

Material and method 

5Structure A, as detailed in fig. 2a is the focal point of this investigation, showcasing its significant archaeological features that are crucial for analyses of chipped stone industries. A complex grid system of 10 × 15 squares was established around the structure (fig. 3), facilitating precise documentation of the spatial distribution of each artefact found on the surface, with each 2-meter unit being individually collected and catalogued. The concentration of the excavation process has been towards the eastern section of the structure, focusing mainly on squares C2 and C3, which include both internal and external spaces. In order to ensure the retrieval of microelements, each bucket of excavated sediment has undergone systematic sieving. The structure excavations encompass two distinct internal areas: the eastern area of polygonal room A in grid square C2 and the more confined area B within grid square C3.

6A holistic analytical framework has been employed to gain a thorough understanding of the lithic materials, intertwining spatial and techno-typological examinations. This all-encompassing approach has been applied to both surface-collected and excavated assemblages. Through spatial analysis, we aim to decode any discernible patterns in artefact distribution and organisation within site and reveal chronological disposal or deposition trends. This inquiry has been enhanced further by meticulously evaluating the raw materials leveraged for production. Our research methodology incorporated refitting studies as a crucial component. Through this process, we obtained valuable insights into the sequential approach to bidirectional blade production, strategies employed for core management and rejuvenation, and the spatial arrangement of knapping activities within the site.

Analysis

Surface collection

7Surface material exhibits a low degree of desert varnish development at the surface of chipped industry, typically on one side, characterised by a hue range from orange to brown and aeolian polishes. This pattern shows that the artefacts have not been subjected to significant periods of wind-driven abrasion or the gradual development of a manganese and iron oxide layer, commonly called desert varnish.

8The chipped stone industry, observed within the surface cluster (figs. 3–5), is overwhelmingly represented by debitage, accounting for 43% (n = 562) of the recovered artefacts. Blades emerge as the dominant type within the debitage, forming approximately 59.80% (n = 336) of the total (fig. 4). A detailed examination of the blade products reveals that single-platform unidirectional blades stand out, contributing to 56% (n = 102) of the diagnostic blade debitage. These blades display a roughly even distribution of parallel and non-parallel lateral edges in an estimated 1:1 ratio. On the other hand, products of bidirectional blade production are less common, comprising 44% (n = 78) of the diagnostic blade debitage. The remaining blades (n = 156) exhibit a preservation state that restricts the identification of the operational chain, leading to their categorization as undetermined blades.

9Additionally, flakes are a considerable component of the debitage from the surface, accounting for 40.20% (n = 226). The assemblage further includes what was defined as primary elements—initial flakes with more than 50% of the dorsal surface covered with cortex, usually associated with the initial stages of core preparation. Interestingly, despite their limited frequency per knapping sequence, their proportion in the surface assemblage is considerably high—11.60% (n = 153) of the total surface finds. The high prevalence and some of the tools also suggest that, in some cases, these were intentionally produced as blanks for tool production. The smaller chipped industry, such as flakelets (less than 25 mm) and chips (less than 10 mm), is only scarcely found on the surface, likely due to aeolian processes.

10A significant collection of cores (n = 80, 6.20%) was discovered on the site’s surface, indicating on-site production (fig. 3). The majority of diagnostic cores were identified as blade cores (n = 42, 65.60%), with flake cores being less prevalent (n = 22, 34.10%) in a 2:1 ratio. The rest of the cores (n = 16, 20% of all cores) were so heavily reduced that further identification was impossible. Among the blade cores, a significant majority (n = 29, 69%) showcase a single striking platform, most frequently appearing as (sub) pyramidal blade forms (n = 17) with flaking surfaces extending around the entire core perimeter or cores with a single platform and single face serving as the primary flaking surface (n = 12). On the other hand, bidirectional cores constitute a smaller fraction of the blade cores, accounting for 31% (n = 13) and establishing a rough 1:2 ratio with unidirectional blade cores. Several types of bidirectional cores, in line with Barzilai’s (2010: 170) classification, have been identified, including naviform (n = 4) and bipolar preparation (n = 4), along with bidirectional bladelet cores. However, due to the advanced reduction state, the remaining cores could only be broadly classified as bidirectional blade cores. Flake cores (n = 22) typically present a single unprepared platform (n = 13) with a flaking surface that encompasses the platform’s perimeter or opportunistic discoidal cores that utilise the negative of a previous removal as a platform for the next flake (n = 8).

11A total of 39 tools were identified from the surface assemblage, comprising 3% of the overall collection (fig. 5). The burin represents a majority of the formal tool classes, accounting for 57.90% (n = 22) of the identified tools. Among the burins, several types were noted, including simple burins without a distinct platform (n = 5), burins that utilised the break as a platform (n = 5), and dihedral burins (n =4). Evidence of on-site burin production and resharpening was discovered, albeit infrequently, in the presence of platform spalls (n = 7, accounting for 0.17% of the assemblage). It is worth noting that burins were predominantly fashioned from blades (n = 14) and, to a lesser extent, from flakes (n = 6). The tool assemblage also includes frequent scraping tools (n = 8,21% of the tools), primarily made on flakes (n = 6). Other tools, while present in smaller quantities, still contribute to the diversity of the tool kit. The cutting tools, which are retouched pieces with cutting edge with used wear and truncating or backing retouch, make up 7.90% (n = 3) of the tools, followed by borers at 5.30% (n = 2), and heavy-duty tools accounting for 5.20% of the tool collection. Additionally, a single pick was found (n = 1, representing 2.60% of the tool count), further exemplifying the wide array of activities undertaken at the site.

Spatial patterns of the chipped industry from the surface

12The spatial distribution of artefacts at the site provides intriguing insights into its past usage and the taphonomical processes impacting the site. As figure 3 indicates, the highest concentration of artefacts was found around the central structure. In contrast, the structure’s primary interior space remains almost devoid of items. This pattern results from aeolian taphonomic processes when large flat ashlars appear to have trapped sand blown from the surrounding desert, subsequently burying artefacts between these stones and spreading. This process results in small chips and flakelets being blown away from the surface, as evident in figure 4b.

13Abandoned lithic cores (n = 80), an essential signifier of tool production activities, are predominantly concentrated in specific sections of the `Ainab 1. These cores, originating from flake production, unidirectional blade production, and bidirectional blade production, are notably clustered in the northern region, especially around squares F9 and C8, and the southern part of the surveyed area, specifically around squares J1 and J2. However, the distribution of cores originating from particular technology across the site is markedly uneven, as depicted in figure 3. An exciting pattern emerges when exploring the spatial distribution of specific blank products alongside the cores.

14Evidence of bidirectional blade production is primarily located in the northeast area of the central structure, specifically within square C8. Traces of this technology can also be found east of the structure, notably in squares A2 through A6, albeit in smaller quantities. The lack of bidirectional cores and higher concentration of tools in this area indicates the consumption of bidirectional blades rather than their production. On the other hand, evidence of unidirectional blade production concentrates primarily in the area north of the structure around squares G9 and F9 and east of the structure around squares A7–A8. Consumption of unidirectional blades is more evenly distributed through context, with a slight tendency to accumulate around square E9. The distribution pattern of core trimming elements, specific to a particular technology, generally replicates the spatial distribution of the cores.

15Debitage classified as flakes shows more even distribution around the site overlapping with remains of other technologies, reflecting that a limited quantity of flakes is also produced during blade core preparation. However, flake cores tend to accumulate north and northwest of the structure (squares H10, F11) and in the southeast area in squares J1 and J2, which were likely spaces where flake production occurred. Primary elements, such as flakes and blades with a cortex proportion of more than 50%, generally mirror the distribution patterns observed in blade production areas. However, the southwestern area of the structure, characterised by a relatively sparse presence of blade production, appears to be associated with flake production. Some of these flakes were possibly intended to be produced intentionally as blanks or ad hoc tools.

16The described spatial patterns may indicate the presence of distinct zones and timeframes related to different tool- production strategies, contributing to a deeper understanding of the spatial and temporal organisation of activities at `Ainab 1. Furthermore, the sporadic reoccupation of this currently hyper-arid region during the Early Holocene period highlights its historical significance.

Excavated collection

Primary production

17Three operational chains have been attested at the site: unidirectional blade production, bidirectional blade production, and flake production. This difference also correlates with utilising different raw materials for the particular operational chain.

18In addressing the state of preservation of the excavated material, we noted key aspects that hold implications for our understanding and analysis. As expected, the desert varnish is less frequent on the excavated assemblage than on the surface finds. Its reduced presence on the excavated materials suggests a degree of protection from external weathering processes. Additionally, patination, a chemical weathering process that produces a visible surface alteration on lithic materials, is observed in about 15–20% of the assemblage. It is essential to note that the degree and nature of this patination show significant variation across different raw material groups. Characteristically, the patination exhibits limited shades ranging from blue to white, hinting at varying exposure and weathering processes.

Raw materials

19A comprehensive analysis and characterisation of raw materials were conducted on all excavated chipped lithic samples (figs. 6–7). Chert emerged as the primary raw material, constituting 98% of the chipped industry production. The residual 2% comprised a mix of quartzite (1%), basalt (1%), and limestone (fig. 7a). We distinguished eight macroscopic groups of cherts that exhibited variation in flaking properties, physical features, colours, shapes, and likely origins (fig. 6). Although the identification and survey of chert sources extended beyond the scope of the 2014 season, we could infer certain aspects about the raw materials from the archaeological specimens. Group B chert is the most prevalent category, accounting for 41% (n = 1,425) of the samples. It is defined by its ultra fine-grained texture with sporadic 5–10 mm limestone inclusions, a translucent brown-chocolate hue, and a glassy lustre. Preserved hard limestone cortex suggests this chert is derived from primary outcrops. Group A chert, representing 29% (n = 1,094) of the samples, is characterised by its milky translucency, grey-whitish colour, and medium to fine-grained texture. Most samples with preserved natural properties exhibit a secondary blackish-varnished cortex on old breakages, while a primary cortex with a thin layer of lime is rarely preserved. A refitted sequence of three cortex removal flakes points to this group’s local or regional origin. Less frequently used are the colourful brown-to-pale-black, opaque, medium-grained flint from Group C (12.85%, n = 510) and the locally abundant opaque coarse-grained group E with its sandy colour (4.50%, n = 154), as well as the medium-grained blackish group G (0.70%, n = 31). The remaining assemblage comprises minor groups F, H, and pinkish chert. Each represents less than 1% of the identified flint raw material groups. Furthermore, the preservation of material in 14% (n = 556) of the cases did not allow for further chert identification.

20Upon examining the distribution of primary products among the most commonly encountered flint raw material groups, certain variations become evident, suggesting diverse raw material procurement strategies (fig. 7b–7c). The production of bidirectional blades broadly utilised flint material groups A and B and, to a lesser degree, group C. Group A flint was predominantly used for bidirectional production, with only a modest number of unidirectional blades produced from this type of flint. On the other hand, unidirectional blade production focused on the utilisation of flint group B, with chert groups C and A being a secondary choice. In contrast, flake production displayed more flexibility and opportunism, including the selection of locally available coarser raw materials and finer chert groups A and B. Some of these could be byproducts linked to blade core preparation and maintenance rather than being a part of the primary flake production strategy.

Bidirectional Blade technology

21Bidirectional cores were primarily shaped from nodular or tabular flint from groups A and B. The sporadic presence indication of bidirectional blade production, such as cores and core trimming elements from the chert of group C, implies these elements were possibly manufactured off-site. Moreover, the overall rarity of primary elements or natural chunks from chert groups A and B suitable for bidirectional production further suggests that the initial raw materials testing and rough shaping most likely occurred at the outcrop area, away from the settlement.

22In the majority of cases, the core’s initial shaping into a trapezoidal form involved a series of platform-shaping flakes (figs. 8.22, 9: aggregate I), leading to a typical postero-lateral core (fig. 8.21). In other instances, bifacial reduction resulted in the forming of a typical naviform core (fig. 8.16). This process was subject to the raw material’s shape, the materials’ inherent angles and the skill level of the producer. Specifically, in the case of tabular flint sources, platforms could be established using natural, ad hoc angles without any alteration, resulting in bipolar cores (fig. 8.11).

23Following the core’s initial shaping, the next step entailed the establishment of the striking platforms by removing initial platform spalls from opposite sides. These larger flakes, whose width correlates to the core’s width, bear the negatives of the initial core preparation. In instances of naviform preformation, characteristic crested platform spalls are detached (fig. 8.24). In contrast, in postero-lateral preparation, platform spalls displaying negatives of a series of platform trimming flakes, predominantly deriving from one side are apparent (fig. 8.23). Occasionally, to prepare the characteristic narrow wedge shape of the flaking surface, a series of wide lateral core trimming flakes were removed from the sides of the leading platform (fig. 9: aggregate IV).

24The primary flaking surface is established by a series of initial blades (figs. 8.12–8.13, 9: aggregate VI), detached from both sides of the core. A leading guiding ridge must be present to ensure the removal of these initial blades. This ridge can be established through unifacial or bifacial cresting retouch (fig. 8.13) or without such modifications, in case the raw material’s natural angular or planoconvex edges can be used.

25The targeted products of bidirectional blade technology are standardised central blades with straight profiles and edges. These blade blades are occurring either predetermined (figs. 8.1–8.2, 6–10; n = 92) or unpointed (fig. 8.3–8.4; n = 51). From the sides are removed lateral (débordante) blades (fig. 8.5; n = 69), which often exhibit cortex or scar patterns from initial core shaping. A distinct category in the bidirectional blade production process involves the removal of Upsilon blade flakes (fig. 8.17–8.18; n = 33). These shorter blade-flakes primarily function as core trimming elements within the operational sequence rather than being considered primary products due in part to their small size, which likely explains why no tools have been found that were made from these blades. They serve a specific production purpose: straightening of ridges on the flaking surface after a predetermined blade detachment. 

26Bidirectional blades exhibit a degree of uniformity in dimensions, especially when contrasted with blades produced by other methods (fig. 10a, 10c, 10e). Complete preserved blades are, on average, 50–52 mm long, though exceptional pieces extending up to 100 mm have been found. Their width is typically between 13–15 mm, and they possess a relatively uniform thickness of 3.50–4.50 mm. Central blades predominantly have a quadrilateral cross-section (n = 215), but triangular (n = 70), pentagonal (n = 28), and even hexagonal cross-sections (n = 6) are also present (fig. 10b). Profiles of the bidirectional blades (fig. 10d) are predominantly straight (n = 66), with twisted (n = 21) and proximally curved examples also found. Epsilon blade flakes are notably smaller, with an average length of 32.10 mm, a width of 12.68 mm, and a thickness of 4.20 mm. When combining the average length of predetermined and epsilon blades, the relative length of the flaking surface is around 83.50 mm (based on Barzilai’s formula, Barzilai 2013: 60). Predetermined blades are detached approximately at 61.50% of the length of the main flaking surface.

27The platforms of the blades are typically linear (n = 73), punctiform (n = 47), or lenticular (n = 45) in shape (fig. 10f). Notably, these platforms exhibit a degree of uniformity, with widths falling between 2–3 mm and lengths spanning 2–5 mm (fig. 10g). In most instances (n = 119), platforms are meticulously prepared through a combination of microchipping, which aids in isolation, followed by grinding to improve platform support (fig. 10h). This specific method of preparation is primarily associated with bidirectional blade production. Other frequently employed techniques for platform preparation include grinding (n = 70) or microchipping (n = 23).

28Bidirectional blades from `Ainab 1, typified using the southern Levant classification, were predominantly detached, utilising the “predetermined-upsilon” scheme (Barzilai 2013: 66). This is evidenced by the substantial presence of upsilon blades (figs. 8.17–8.18, 9: aggregates II, VII, VIII). However, a few instances demonstrate a “single-dominant platform scheme”, with an even smaller number exhibiting the “one-on-one” knapping scheme. It is worth noting that this latter technique was also implemented during the later stages of the PPNB in Jordanian highland villages, such as Beidha (Barzilai 2013: 68).

29The sophisticated craftsmanship is evident in the careful maintenance of flaking surfaces and platforms. Mishaps in knapping, such as hinged and stepped negatives on the flaking surface (figs. 8.19, 9: aggregate V), were usually corrected by intentionally producing overshoot blades (fig. 8.15), renovation blades (fig. 8.14, 8.20), or lateral renovation flakes (fig. 8.26). The angle between the platforms and the flaking surface was adjusted as needed by deliberately removing core tablets (fig. 8.25, 8.27), sometimes consecutively (fig. 9: aggregate III). The bidirectional blade production’s blanks were primarily utilised as projectiles, burins, perforators, and cutting tools (fig. 12b).

30Bidirectional production ends when cores are discarded in the stage when the flaking surface’s length is insufficient (ca. 7 cm) to remove desired blades, even though some more blade sequences could still be detached. In other cases, the cores were discarded when a particularly challenging knapping accident occurred.

Unidirectional blade technology

31Compared to bidirectional blade production, the operational chain of unidirectional blade technology is significantly more straightforward. The selection of raw materials tends to be less stringent, with cores being prepared from all medium-to-fine-grained flint categories—A, B and C (figs. 6, 7c). Notably, flint from category B is predominantly used. Unidirectional blades usually exhibit either parallel-sided edges (n = 137; fig. 8.30) or non-parallel-sided edges (n = 142; fig. 8.29), maintaining approximately a 1:1 ratio between the two. Both types likely stem from the same technological process, though the harmonisation of edges typically takes place at later production stages.

32Generally speaking, unidirectional blades are more asymmetrical, less standardised, broader, shorter, and thicker than bidirectional blades (fig. 10a, 10c, 10e). A distinct difference also resides in the blade profiles (fig. 10b). Unidirectional blades commonly display a more accentuated curvature, primarily in the distal part (n = 29), or are entirely curved throughout the profile (n = 26). Straight profiles, though generally preferred, are rare. Cross-sections infrequently have more than five angles, with triangular (n = 83) and quadrilateral cross-sections (fig. 10b) prevailing. Platform shape and preparation methods in unidirectional blade production (fig. 10f–10h) exhibit considerable variation and an absence of standardisation. Lenticular (n = 28) and linear (n = 22) platforms are the most common, followed by less frequent punctiform shapes (n = 13). Trapezoidal (n = 10), rhomboidal (n = 7), and rectangular (n = 7) platforms constitute the minor categories. Platform preparation methods for unidirectional blades primarily include grinding (n = 46), microchipping (n = 29), or even no modification (n = 22). Platform dimensions also show more variability, with 2–15 mm lengths and widths from 2–6 mm. The characteristics of the platforms and the frequent presence of pronounced bulbs with scars strongly suggest that direct hard hammer percussion was the primary technique employed in producing the unidirectional blades.

Secondary production

33In the material unearthed from `Ainab 1, tools comprise a substantial proportion of 4.50% (n = 180) of the chipped stone assemblage (fig. 5). Notably, a significant majority of these tools, approximately 68.3% (n = 123), exhibit signs of intentional retouching, marking their classification as formal tool types. Meanwhile, the remaining 30.60% (n = 55) are non-formal tools, predominantly constituted of blades and flakes with use-related retouches along one of their lateral edges.

34In the subset of deliberately retouched tools, burins (fig. 11.23–11.32; 25.70%, n = 33) and projectile points (fig. 11.1–11.18, 11.21–11.22; 25.20%, n = 31) are most frequently represented. However, this finding should not overshadow the presence of other tool types, which, albeit less common, are equally critical to deciphering the spectrum of human activities undertaken at the site. Noteworthy among these are cutting tools comprising blade and flake morphologies with a cutting edge and evidencing either marginal or proximal/distal retouching in the form of truncation or backing (fig. 8.1–8.2; 13%, n = 16). Additionally, the repertoire includes scrapers (fig. 11.36, 11.38; 10.60%, n = 31), perforators (fig. 11.32–11.35, 11.43–11.44; 7.30%, n = 9), notched tools (fig. 11.39; 5.70%, n = 7). Worthy of note is the founding of two samples of Hagdud truncation (fig. 11.19–11.20), which despite comprising a mere 1.60% of formal tool classes, serves as an influential chronological and cultural marker for Early PPNB as defined by Gopher (1996). Other minor categories within the formal tool class documented at the `Ainab 1 comprise denticulations (fig. 11.42; n = 4), multiple tools (fig. 11.37; n = 3), picks (n = 1), heavy-duty tools (n = 1), splintered pieces (n = 1) and other retouched pieces (n = 3) with either backing or truncating retouch which could not be classified into other categories. Interestingly, non-formal tools, characterised by their irregular retouch patterns on blades and flakes, comprise a substantial proportion of the tool assemblage (30.60%, n = 55). These tools, displayed in figure 12 (fig. 12.40–12.41, 12.45–12.46), suggest they were frequently used as versatile tools in various tasks. Despite their non-formal classification, their substantial presence in the excavation findings underscores their importance in the tool repertoire of the site’s inhabitants.

Projectile points

35As previously highlighted, `Ainab 1 has yielded 31 projectile points, crucial elements for chronological classification. Interestingly, more than two-thirds of these points, precisely 21, conform to the Helwan type (fig. 11.1–11.18), a distinctive characteristic of the Early PPNB period. Identifying the remaining ten specimens, mainly terminal fragments, is somewhat challenging due to their fragmented condition (fig. 11.21–11.22). Merely three pieces (fig. 11.2–11.4) were found intact, indicating an impressive fragmentation rate of 90%. This rate implies that most projectiles reached the site already broken, possibly embedded within the hunted prey, suggesting further processing activities were carried out on-site.

36A strong preference for flint from groups A and B is discernable upon evaluating the lithic raw materials employed in crafting these projectile points (fig. 12a). Despite this trend, flint from raw material group C has been utilised to a lesser extent. Focusing on the technological attributes, all diagnostic Helwan points are meticulously shaped, applying pressure retouch from fine, straight-profiled, bidirectional central predetermined blades. These blades exhibit an array of cross-sections, encompassing triangular, trapezoidal, and at times, hexagonal shapes. Notably, the lateral edges are carefully fashioned through pressure retouch, with a dominant inclination towards the ventral side. Looking at stylistic and hafting features, the Helwan points commonly incorporate pairs of lateral notches located on the basal corners of the blank, supplemented by an additional pair of notches in the medial part (fig. 11.2–11.6, 11.8–11.9). This specific configuration likely facilitated effective hafting. Interestingly, a subset of five examples features a distinct design with double lateral notches at the medial part (figs. 11.1, 11.7–11.9, 12d).

37From a morphometric perspective, the complete Helwan points exhibit an average mass of 2.30 g (fig. 12e), with standard dimensions hovering around 47 mm in length, 13.50 mm in width, and 3.20 mm in thickness. However, these metrics can be skewed due to outliers, such as the unfinished sample EJP14.78.01 (fig. 11.1), which possesses extreme measurements. As a result, the median mass value of 1.70 g is considered a more representative measurement, minimising the impact of such anomalies. The resulting morphological characteristics of the Helwan points, including their length, width, thickness, and mass, make them ideally suited for use as arrow armatures, providing a high degree of accuracy and penetration power (Borrell and Štefanisko 2016). The thoughtful design, precise dimensions, and accurate crafting techniques utilised in producing these Helwan points illustrate the significance of these tools in the hunting practices of the societies that populated the `Ainab 1 during the EPPNB period.

Burins

38Burins constitute the most frequently found formal tools at the `Ainab 1. These tools are exclusively manufactured from flint raw materials groups A, B, and C (fig. 12a). The majority of burins are created using both bidirectional and unidirectional thick blades, with the production of burins on flakes being a less common practice (fig. 12b). Of these burins, over half (54.50%, n = 18) are simple burins characterised by a single burin blow either on a break (fig. 11.26), with a transverse burin blow (fig. 11.25), or without a designated platform The burins are further classified based on the orientation of truncation; there are burins on straight, oblique, and concave truncations (fig. 11.23, 11.30). Burins with multiple blows fall into various categories, including those with resharpening of the same edge using multiple blows (fig. 11.27), dihedral burins (fig. 11.23, 11.28–11.29), and multiple opposed burins (fig. 11.24, 11.31). The observed 1:5 relationship between burins and their corresponding burin spalls confirms the evidence of on-site burin production and maintenance activities. Indeed, the considerable quantity of burins (45.50%) demonstrating more than one detached spall invites fascinating conjecture. The recurrent detachments may indicate these spalls are being produced primarily and subsequently used as piercing or drilling tools. This perspective raised by Köhler-Rollefson’s (1992) propounds the notion of spalls becoming commodified tools within agro-pastoral communities. The planned use-wear analysis on these tools will provide empirical data to contest this assumption, thereby giving us a clearer understanding of whether they result from these tools’ production or usage patterns.

Scrapers

39Scrapers account for 10.60% (n = 13) of the formal tool types discovered at the site. The predominant variants are end-scrapers (n = 6), notable for steep backing retouch. However, in two instances, repurposed bidirectional cores were used as scrapers with minimum further shaping capitalising on the already present edge at the intersection of the platform and flaking surface. A standout amongst the scrapers is a single scraper with a deliberately retouched base for hafting (fig. 11.36). This unique piece was derived from a non-bidirectional blade flake detached from the tabular raw material of group C using a solitary striking platform. The base, featuring a platform from the blank, was deliberately snapped, followed by a direct retouch to form the tang. The lateral side bears shaping retouch for hafting, while the distal working edge has been carefully shaped to oblique at approximately a 45-degree angle.

40The noticeable absence of agricultural tools such as sickles or celts and the prominent presence of projectile points and burins suggests that Ainab 1a functioned as a periodically used hunting camp. This interpretation aligns with a lifestyle characteristic of hunter-gatherer societies, wherein inhabitants would have been adapted to follow migratory herds or to seek out fluctuating resources, adjusting their survival strategies in response to seasonal changes and resource availability. The conspicuous scarcity of groundstone artefacts at the site lends credence to this hypothesis. This lack suggests that plant processing, a principal activity in agricultural societies, may not have been a primary focus at `Ainab 1. However, their absence could also be the outcome of environmental settings or the socio-economic function of the site within the broader network of settlements.

Spatial patterns of the chipped industry from stratigraphy

41To reiterate, for the sake of our analysis, we segmented the studied area into four designated zones: two interior spaces labelled A and B, and two exterior zones, C and D, situated around the structure (fig. 2a). Furthermore, our excavation brought to light three significant concentrations of bidirectional blades coupled with numerous core trimming and maintenance elements (fig. 9). These clusters are located in locus C2:9, south of the main structure; locus C3:12 located east of the structure and locus C2:8, inside the structure (fig. 3a). The blades from these concentrations are commonly broken or exhibit a thick or asymmetrical profile—possibly undesired attributes for their producer. Moreover, typical characteristics of knapping accidents, such as hinged and stepped terminations, are present in higher frequency, pointing to their interpretations as single event “chipping floors” featuring bidirectional unsatisfactory products and by-products. The eight refitted aggregates add weight to this interpretation while offering unique insights into the subtleties of bidirectional technology practised at the site.

42In area A, over 30% of the lithics excavated were located in locus C2:8 (fig. 2b). This significant assembly of bidirectional debitage, comprised of two flint raw material groups, A and B, hints at two separate production sequences during the same event. Interestingly, tools are not common in this sector, with the majority being projectile points (fig. 2c). The relative scarcity of domestic tools such as burins and scrapers implies that domestic activities were likely carried out of the main interior space, although a single event of tool production is seemingly present inside.

43Located directly to the north of area A, we find a compact space termed area B (fig. 2a). Within this sector, the tool assemblage is significantly denser, comprising 6.16% of the findings in sector. That suggests these areas may have been subjected to more intensive tool usage or decommodification processes (fig. 2c). The lithic collection in this area features a notable quantity of burins, scrapers, and non-formal tools. Importantly, this is the only excavation zone where exclusively osteological material has been retained, identified as originating from ungulates.

44A significant majority of lithics, approximately 80%, were unearthed from areas C and D, located on the structure’s periphery. This distribution could suggest that most activities were conducted in open settings. Two distinct “chipping floors” in areas C2:9 and C3:12 host a rich concentration of bidirectional products. These concentrations likely mark unique events in the bidirectional blade production sequence and reflect the site’s dynamic and evolving usage.

Summary and discussion 

Dating of `Ainab 1 chipped industry

45While `Ainab 1 is located a considerable distance away from the Fertile Crescent and the southern Levant mainland, a detailed examination of its chipped stone artefacts presents classical traits indicative of the Early PPNB period in the Levant, especially the northern tradition, which are linked directly to the establishment of a semi-circular structure. This assertion is primarily backed by the substantial presence of Helwan points, recognised as crucial chronological markers for this period (Gopher 1996). Alongside these, there is a contained incidence of Hagdud truncations. Though these are primarily linked to the preceding PPNA era, they also appear in small quantities in the Early PPNB, as demonstrated by research at Jilat 7 (Garrard et al. 1994b: fig. 8.1). Further illustrating the connection to the PPNB period is the extensive employment of bidirectional blade technology. Originating in the Euphrates Valley, this technology was introduced to the southern Levant at the beginning of the PPNB (Abbès 2003; Barzilai 2010). Primary production attested at the site demonstrated a high level of expertise and skill in employing this technology, which would have necessitated years of repeated practice to acquire. However, it is essential to note that certain traits of Early PPNB features, such as chipped celts and sickle blades commonly found at permanent settlements on the Mediterranean coast, are notably absent at `Ainab 1. This absence could reflect the site’s environmental conditions, seasonal habitation, cultural origin and difference in subsistence strategies employed by communities rather than an anomaly in its cultural and chronological affiliation.

Spatial and temporal pattern of lithic production at `Ainab 1

46The analysis of primary production at `Ainab 1 revealed the presence of three distinct core reduction strategies, each with varying degrees of standardisation. Apart from the highly prominent and standardised bidirectional blade production, the site also hosted a less standardised unidirectional blade production and a more opportunistic flake production strategy. The high degree of standardisation in bidirectional blade technology is evident in several aspects: a meticulous selection of raw materials, where only high-quality flint from raw material groups A and B has been used for production on site; the systematic approach to core preparation; the management of knapping accidents; platform preparation, the removal sequence of blades, and, most importantly, the characteristics of the primary product—bidirectional blades. Although all types of bidirectional core preparation (naviform, postero-lateral and bipolar) were attested, the dominant was the postero-lateral preparation with the predetermined-upsilon sequence of blade removal. On the other hand, unidirectional blades were produced from a broader range of fine-grain cherts, although cherts of group B seem to be the most preferred. Higher variability among them could also be seen in the variation of platform preparation and blade properties in terms of size, edges, or profile symmetry. Finally, flake production was very opportunistic, utilising mainly locally available coarse-grained cherts of group E without attention to platform preparation or sequence of removal.

47A unique temporal pattern in primary production emerges when comparing primary lithic production from the surface material and the excavation (fig. 4a). The surface collection reveals that unidirectional blades were produced more often than bidirectional blades in a 5:4 ratio, contrary to unearthed material, where bidirectional blades dominate in a 1:1.8 ratio (fig. 4b). Another notable discrepancy is the contrasting abundance of primary elements, accounting for 11.60% of the surface collection while constituting only 2.40% of the excavated assemblage. Moreover, flakes were also more frequent in the surface collection among the debitage group, in a 2:3 flake-blade ratio, while the excavated material exhibits a ratio of 1:4.

48The spatial analysis of surface material has unveiled distinct patterns linked to primary production activities. Remarkably, these patterns suggest separate spatial clusters for varying types of primary production, each exhibiting a degree of overlap. These stratigraphically unconnected clusters likely represent a mix of seasonal reoccupations by disparate communities over time, each utilising unique production strategies. Interestingly, these precise spatial patterns are not visibly mirrored in the stratigraphy. Instead, as highlighted previously, the stratigraphy presents a more homogeneous, bidirectional blade-oriented production, emphasised by the presence of several chipping floors with corresponding refits.

49The spatial distribution of lithics within the stratigraphy of `Ainab 1 offers profound insights into the site’s unique functional divisions (see fig. 1b–1c). Area A, an inner region, is characterised by a notable lack of domestic tools, suggesting it was maintained as clean. However, a particular event of bidirectional blade production, represented by a single chipping floor, contributes to 30% of all recovered chipped industries from this space. In stark contrast, the confined space of area B boasts a substantial concentration of domestic tools like burins and scrapers, in association with the site’s only preserved osteological material, likely of ungulate origin. On the site’s peripheries, the distribution of lithic classes is more uniform, reflecting a broad spectrum of tool production and game- processing activities. The relative uniformity across stratified layers of structure implies that communities with shared cultural and economic traditions seasonally reoccupied these spaces. Notably, the unidirectional blade production appears mixed in the structure’s exterior, likely due to the influence of aeolian taphonomic factors.

Site function and subsistence strategies of inhabitants as reflected by tool production

50The lithic tool assemblage at `Ainab 1 provides compelling insights into the past inhabitants’ activities and subsistence strategies. The surface assemblage predominantly showcases characteristics typical of burin sites, with burins representing a significant 60% of the total tool class. A considerable presence of burins is also noted in the stratigraphy, albeit at lower frequencies than in the surface findings. This pattern aligns with observations at other Early and Late Pre-Pottery Neolithic B (EPPNB and LPPNB) sites, such as Jilat 7 (Garrad et al. 1994b) and the ephemeral PPNB sites (DAJ 112–125) in Al-Jouwf, Saudi Arabia (Crassard and Hilbert 2020). It is worth noting that burin sites are generally linked to agro-pastoral communities during the LPPNB-FPPNB period (Köhler-Rollefson 1992; Miller et al. 2019), and their significant presence on `Ainab 1 surface could be interpreted in this context.

51In examining the stratified finds at `Ainab 1, it is equally intriguing to note what was absent. Tools such as sickles and bifacial celts, usually indicative of sedentary agricultural societies, are notably missing from the site. Instead, the evidence points towards a hunting-based economy amongst `Ainab 1’s early inhabitants. This hypothesis is suggested by discovering a solitary fragment of an unidentified groundstone object and the high frequency of projectile points, accounting for over 25% of the total assemblage. Furthermore, the high fragmentation rate (90%) of these points implies that most of them were brought to the site already embedded within hunted prey, suggesting the presence of butchering activities at `Ainab 1. Adding weight to this interpretation is the recent discovery of numerous “desert kites” in the surrounding area (Abu-Azizeh and Tarawneh 2015; Crassard et al. 2022; Crassard et al. 2023). These structures, thought to have served as large-scale traps for ungulates during the Late and Final Pre-Pottery Neolithic B (LPPNB and FPPNB) periods, underscore the historical significance of the region and the strategic position of `Ainab 1 within this hunting network. Therefore, `Ainab 1 could have represented an early experimentation with organised hunting strategies in the region, as expressed by the intricate elaboration of Helwan points of northern cultural origins. The analysis of the evidence suggests a complex and nuanced understanding of early Neolithic subsistence strategies in this arid region of the Arabian Peninsula.

52The available evidence posits that `Ainab 1 initially served as a seasonal Early PPNB hunting camp. The site may have also been sporadically reoccupied by other communities whose subsistence strategies increasingly leaned towards agro-pastoral subsistence strategies as reflected by burins and different production strategies. A similar pattern of change in subsistence strategies was already proposed by Fujii (2013a) in Jafr Basin. This intricate sequence of occupation and diverse site utilisation enriches our understanding of the dynamics of the Neolithic revolution and its spread to Arabia. Interestingly, this transition is mirrored in the petroglyphs found at Shiwaymis in central Saudi Arabia (Gaugnin et al. 2015). These ancient rock carvings narrate a story strikingly similar to the narrative captured in the archaeological record, reinforcing the distinctive trajectory of Neolithisation in these regions. The observed progression underscores the dynamic and non-linear nature of cultural evolution, underscoring its inherent adaptability across diverse geographical landscapes. As such, the transition from hunting-foraging strategies to agro-pastoral practices during the Neolithic period illustrates the influence of the environment on human adaptation, with potential implications for our understanding of ancient human migration and cultural diffusion.

`Ainab 1 in the context of Eastwards Early PPNB expansion

53Despite the current absence of radiocarbon data from `Ainab 1 posing some interpretative challenges, necessitating reliance on relative chronologies extracted from the site’s chipped industry, `Ainab 1’s significance in our understanding of Early PPNB expansion is undeniable. Identified as the farthest eastern seasonal settlement showcasing architecture and emblematic early EPPNB lithic tradition, `Ainab 1 offers new insights into the intricacies of the Neolithic transformation. Furthermore, it serves as a crucial archaeological nexus, evidencing the dispersal of the Neolithic tradition from the Levant into the Arabian peninsula.

54The chipped industry of `Ainab 1 bears the closest resemblance to sites approximately 300–500 km to the west: Early Pre-Pottery Neolithic B (EPPNB) sites such as Harrat Juhayra 202 (Fujii et al. 2019) in the western Jafr Basin and the sites of Jilat 7 (Garrard et al. 1994) and Mushash 163 (Rokitta-Krumnow 2019) in the Azraq Basin. All these sites exhibit shared elements in their chipped stone industries, especially in the bidirectional blade production, the prevalence of Helwan points, and architectural commonalities. However, `Ainab 1 lacks a Levantine PPNA component, reflected by el-Khiam points, suggesting a later chronological introduction into the area. This absence could reflect the limitations of current data or the reality of a later settlement timeline. As it stands, these findings lend support to the theory that the expansion of the Early PPNB population originated from the north (Edwards 2014; Finlayson et al. 2014), extending this tradition further east into the Arabian Peninsula. It is plausible that northern migrants, potentially resisting sedentary lifestyles, ventured south and south-east towards new hunting grounds. Through these movements, they disseminated the classical northern Early PPNB traditions, engaging in cultural exchanges with local populations in southern Jordan.

55Situated roughly 40 km north of `Ainab 1, Jibal al-Khashabiyeh and the Ghassanian complex offer an unparalleled look into the latter phases of the Pre-Pottery Neolithic B period, ca. 7000 BC in the Jordanian Desert (Crassard et al. 2022). This complex is closely associated with “desert kites”, expansive installations ingeniously devised as gazelle traps, a testament to the sophisticated hunting methodologies employed by these prehistoric communities. A salient feature of the Ghassanian complex is the prevalence of unidirectional blade production. Interestingly, this method of technology and the utilised materials align intriguingly with those found at `Ainab 1, particularly in terms of the resemblance of cherts of group C. Nevertheless, the Ghassanian complex diverges from `Ainab 1 in unique ways. Unlike `Ainab 1, the Ghassanian site reveals various projectile types from later periods, such as Byblos, Amuq, Ha-Parsa, and Nizzanim, in tandem with the noticeable absence of bidirectional blade production. Furthermore, the Ghassanian complex illustrates an exceptional adeptness in crafting bifaces and foliates, components strikingly absent at `Ainab 1 with only few bifacial thinning flakes. Therefore, `Ainab 1 might indicate an initial stage of hunting experimentation in this locale, introducing a new facet to the rich narrative of cultural transformation during this Neolithic era.

56The compelling early data for the Early Pre-Pottery Neolithic B from Harrat Juhayra 202 (Fujii et al. 2019: 188), particularly from the first half of the 9^th millennium, corresponds to the so-called Late Pre-Pottery Neolithic A (LPPNA) phase concurrently present in southern Jordan (Finlayson et al. 2014). This parallel cultural tradition is demonstrated at sites such as WF16 (Wick et al. 2016), el-Hemmeh (Smith et al. 2014), and ZAD2 (Edwards et al. 2004), thereby contributing to a broader understanding of Neolithic cultural diversity. LPPNA is distinguished by the lack of projectile elements, initial trials with bidirectional blade technology, and signs of preliminary domestic plant cultivation. Moreover, these sites share a unique architectural tradition of communal structures, indicative of deep-seated local customs (Makarewicz and Finlayson 2018). The evidence suggests as above mentioned authors point out, that LPPNA and EPPNB could represent different cultures that coexisted during the same timeframe. LPPNA might embody a continuation of local practices with origins in predomestic cultivation, whereas EPPNB could signify cultural traditions migrating from the north. In this discourse, `Ainab 1 closely aligns with the classical EPPNB traditions.

57The discovery of El-Khiam and Helwan points at Jebel Qatar 101, the easternmost known location of these artefacts in central Saudi Arabia, provides indications for a cultural diffusion between the Levantine population and local communities (Crassard et al. 2013). Interestingly, Jebel Qatar 101 does not exhibit evidence of bidirectional blade production. Consequently, `Ainab 1 could present critical evidence linking the northern Levantine influences to the northern Arabian Peninsula. Thus, the distinct parallels are surface sites DAJ-112 and DAJ-125 in northern Saudi Arabia (Crassard and Hilbert 2020) that further substantiate this claim, hinting at a degree of interconnectedness during the Neolithic era. Despite the absence of projectile points, the similarities extend beyond bidirectional blade production, as both `Ainab 1 and the Saudi Arabian sites significantly emphasise the production of burins. Moreover, the sporadic detection of Levantine Late/Final PPNB style projectiles and instances of bidirectional blade production at various sites, including Al-Aynah (Al-Asmari 2019), Wadi Sharma 1 (Fujii 2013b), and Kilwa in northwestern Arabia, and even as far as Qatar (Inizan 1980), further strengthens the connections between these culturally diverse regions of Arabia during the Neolithic period. 

Conclusion

58The comprehensive study of `Ainab 1 provides a unique glimpse into human life during a significant societal transition. The relative chronology of the projectile points, characterised mainly by an abundance of Helwan points, sporadic Hagdud truncations, and prevalent use of bidirectional technology, allows us to confidently attribute the site to the Levantine Early PPNB cultural tradition, as per Gopher’s definitions (Gopher 1994, 1996).

59Our data suggest that the structure primarily functioned as a seasonal hunting camp during the Early PPNB, a conclusion supported by the high number of projectile points and the absence of celts and sickle blades. Furthermore, only a single grinding stone was found in the stratigraphy, further solidifying our hypothesis. The projectile point morphology underscores the inhabitants” reliance on bow and arrow, with high fragmentation rates suggesting the site served as a hub for butchering activities. Upon scrutinising the spatial and temporal patterns in primary and secondary production, we uncover the complex interplay of environmental, subsistence, and cultural factors. These influenced the spatial distribution of the chipped industry throughout the site’s occupation history. Remarkably, the distribution of flake and unidirectional blade production differs from that of bidirectional production. Such non-random patterns imply diverse site occupants over time, with different social groups imprinting their footprints in the chipped industry in distinct site areas.

60The data from `Ainab 1, when viewed in conjunction with evidence from other Early PPNB settlements in Eastern Jordan, suggest a distinct trajectory compared to sites along the Mediterranean coast and the Jordanian Highland, which evolved into sedentary villages focused on agriculture. In contrast, eastern Jordanian desert populations may have maintained hunting and foraging as their primary subsistence strategy. Basically, four main different scenarios can be assumed for the establishment of EPPNB lithic traditions in the `Ainab region: either indigenous (local) hunter/gatherers have adopted lithic traditions from the settlements of the western favourable regions, or immigrant hunter/gatherers from these favourable areas exploiting the eastern hunting grounds brought these traditions with them. Neither should an autochthonous modification of western EPPNB lithic technologies and styles by regional hunting-foraging people be excluded. In the end, however, it might have been a regionally dependent and variable combination of all three of these scenarios (discussions with H. G. K. Gebel). In the later stages of the PPNB, this lifestyle transitioned towards agro-­pastoralism as hypothesised by S. Fujii (2013a), or developed into an autochthonous pastoral-venatorial culture during the L-FPPNB (Gebel 2020, 2024: fig. 8).

61The presented data significantly contribute to bridging the research gap concerning the Neolithic revolution’s spread from the Fertile Crescent to the Arabian Peninsula’s western province. However, it is vital to address a significant data bias: the geographical remoteness and arid conditions of many Eastern Jordan and Arabian regions have led to these areas being under-explored, thereby creating an imbalance in our understanding of the Early PPNB progression. These landscapes, though arid, may conceal valuable insights into our shared past. `Ainab 1 has undoubtedly provided significant insights into the migration of Early PPNB culture from the Levant to Arabia. However, these insights represent only a fragment of a larger narrative. Addressing the existing bias is essential for unravelling the various facets of the Neolithic revolution and necessitates a comprehensive archaeological exploration of these remote areas. By adopting a holistic approach that encompassing these under-studied regions, we inch towards a more textured, detailed portrayal of our collective past.