Archaeometallurgical Analysis of Lead Weights and Sling Bullets from Seleucid Tell Iẓṭabba: More on Lead Origin in Seleucid Palestine

Abstract

Lead isotope and portable energy-dispersive X-ray fluorescence spectrometry were applied to a small group of Hellenistic lead objects from Tell Iztabba (Beth Shean, Israel). The market weights and sling bullets from this short-lived Seleucid-founded site were analysed in order to understand their production process and the provenance of the raw materials. The sling bullets are of pure lead, and its lead isotopic signature suggests that the lead was produced from Lavrion ores. The weights are heterogeneous in composition and point to the use of mixed or reused scrap metal. Anatolian or Aegaean lead ores are the raw material sources for the weights. Whether the sling bullets belonged to the defenders of the site or to the attackers cannot be determined from the analysis, but their Greek provenance together with the fact that they were recovered at the site may suggest their use by its defenders.

Keywords:

• Lead weights
• Sling bullets
• Hellenistic Levant
• Lead isotope
• Multicollector-Mass Spectrometry
• pXRF

Introduction

Beth Shean, in the northern Jordan Valley, has an urban settlement history stretching from the Bronze Age to the Early Islamic period. Located where the Jezreel Valley meets the Jordan Valley, the area is water-rich and fertile. The settlement history in the Hellenistic period is attested by a Ptolemaic foundation, a Sel eucid refoundation, a Hasmonaean destruction and a subsequent recovery of the settlement under Roman rule (Lichtenberger 2003: 128–134; Ebeling et al. 2020; 2021). Of particular interest is the Seleucid phase of the town, because it can be pinpointed to a specific area and a clear time frame. While the Ptolemaic foundation, called Scythopolis, was mostly situated on Tell el-Ḥuṣn, the location of the Middle Bronze–Iron Age settlements, the Seleucid refoundation in the 2nd century BCE was located on Tell Iẓṭabba, where a new settlement was established and named Nysa (Fig. 1). The ancient settlements of the Beth Shean area are enclosed by a chain of hills north of Naḥal Ḥarod—the watercourse that crosses it. To the south is Tell el-Ḥuṣn, to the east is Tell Ḥammam, where one of the main cemeteries of the Roman and Byzantine city is located, and to the west lies Tell Iẓṭabba, comprising three hills that slope steeply towards Naḥal Ḥarod and more moderately towards the plain north of greater Beth Shean.

Fig. 1: Aerial photograph of the Beth Shean area and location map

When the Seleucids founded the settlement at Tell Iẓṭabba in the 160s BCE, the mound seems to have been unoccupied since the Bronze Age. The foundation took place as part of an urban development programme in the area when the Seleucids settled the region that they had newly conquered from the Ptolemies (Lichtenberger 2008; Tal 2011). The settlements were occupied by Seleucid settlers and probably received civic rights of a Hellenistic polis. It was the Greek character of the settlements that made them targets for the Jewish Hasmonaean state in Judaea; hence, Nysa was attacked and destroyed by the Hasmonaean high priest John Hyrcanus I in 107 BCE (Finkielsztejn 2018; Lichtenberger et al. 2022). Tell Iẓṭabba was largely abandoned until the Byzantine period. This implies that the site provides archaeologists with well-defined Seleucid occupation layers. Excavations in the Hellenistic settlement of Tell Iẓṭabba were carried out in 1962, 1977 and 1991–1994 (Mazor 2008; Mazor and Atrash 2018), and since 2019 a new archaeological cooperation project between the Universities of Münster and Tel Aviv has systematically explored the Hellenistic settlement (Ebeling et al. 2020; 2021).

The material culture and economic policies of the Seleucid site were well connected to the Hellenistic world, although it is also obvious that these were well integrated into local networks and traditional cultures. Thus, for example, the agricultural products and the mudbrick architecture of the town largely relied on local traditions (Orendi, Lichtenberger and Tal 2021; Lorenzon and Cutillas-Victoria, forthcoming), whereas consumption habits indicated by imported Aegean amphorae and the faunal evidence may underline the Hellenistic orientation of some of the settlers (Ariel 2006; Finkielsztejn 2018; Pines et al. 2022). In the following we present the results of portable energy-dispersive X-ray fluorescence spectrometry (pXRF) and Multicollector-Mass Spectrometry for lead isotope measurements of lead objects from the site; we discuss the implications of the production of such objects and supply additional evidence on the economic networks of the Seleucid site.

The lead weights and sling bullets and their archaeological context

In total 14 lead objects, eight weights and six sling bullets were analysed (Fig. 2). With the exception of one weight, which came from our excavations, all the objects were surface finds collected by residents of Kibbutz Bet Alpha (especially Misha Reshef). The 14 objects constitute the entire assemblage of lead weights and sling bullets available for analysis from Tell Iẓṭabba stored in Kibbutz Bet Alpha or discovered during our excavations.¹ Their typological characteristics and the settlement history of the site suggest that their production and use occurred during the site’s Seleucid occupation, between the 160s and 107 BCE.

Fig. 2: Weights and sling bullets from Tell Iztabba

Fig. 2: Weights and sling bullets from Tell Iztabba

Only one of the weights, the disc-shaped lead weight (No. 1), stems from a controlled excavation, the 2020 campaign of our German–Israeli Tell Iẓṭabba Excavation Project. This weight came from Locus 451, which represents top soil in Area D, Square KC33, an area with Hellenistic domestic structures on the western slopes of Tell Iẓṭabba (East) (Ebeling et al. 2021: Pl. 3C). Many of the lead weights found in the country did not state either their value or a municipal authority, although weights have been uncovered in archaeological excavations at various sites (Lifshitz 1976). Our unearthed example is comparable to many other such mina-based lead weights from Palestine (cf., in general, Finkielsztejn 1999; 2001: 16*–17*) that are known to have entered into practice in the Southern Levant based on inscribed examples in the days of Antiochus IV and that continued to be used until the Hasmonaean period. Another weight that came from the site is of a rather less common design: half-moon in shape (No. 8).

Most of the analysed lead weights belong to a specific Hellenistic type: a seashell-shaped (Acanthocardia tuberculata) weight with a knob handle on top (Nos. 2–7). Such weights are known from the site’s previous excavations (as yet unpublished) and have also been found in moderate numbers in the Southern Levant (cf., e.g., Manns 1984: 22, Nos. 37–38 [Pl. 6]; Hendin 2007: 184, 190–191, Nos. 233–234, 271–276; Holland 2009: Nos. 242–243 [made of bronze]; Dar and Nagar-Hillman 2009: 120–123).² They were probably cast in moulds that were imprints of physical seashells, with a knob handle added on top in order to facilitate their use, likely made by pouring a drop of liquid lead into water. It seems that the calibration of the lead weights was done during the production process. To calibrate underweight weights, small strips of lead were sometimes added to the body or to knob handles (Fig. 3), but weight correction may also have taken place at a later stage for recalibration purposes, following the wear and tear of the weights. All these lead weights attest to a developed commercial life at the site during the 2nd century BCE. The weight of our disc-shaped lead weight (230.75 g) suggests the use of a mina unit of weight of some 462 g in the second half of the 2nd century at the site of Tell Iẓṭabba (cf., e.g., Hendin 2007: 88–94). Our seashell-shaped weights (35–88 g in weight) may suggest a local sheqel-based standard of some 12–13 g (cf., e.g., Tal 2007), whereas our moon-shaped weight (158.12 g) may align with a libra-based standard (cf., e.g., Hendin 2007: 95–99). The metrology of such weights usually followed local civic standards and attests to a developed market economy. Greek cities as well as their eastern versions used official civic weight standards and the weights were usually produced in bronze or lead, with bronze being more rarely preserved, probably because it was more often recycled (cf. Killen 2020).

Fig. 3: Top (from left to right): attached strip of lead on bottoms of lead weights of Nos. 3, 4 and 5; bottom (from left to right): details of lead weight knob handles of Nos. 3, 4 and 7

The lead sling bullets (Nos. 9–14) were used as a throwing weapon by stone throwers and slingers (whether attackers or defenders of a site) and are typical of Hellenistic-period warfare (Rihll 2009; Weiss and Draskowski 2010: 123, n. 1; see, in general, Prichett 1991: 1–67 for their history and archaeology). These elliptical/biconical-shaped bullets normally exhibit pointed ends and are either plain or decorated with motifs, slogans, or legends. They were apparently cast in two-part moulds and their motif and/or inscriptions were incised (in either regular or mirror writing) on the mould’s cavities (see reconstruction in Stiebel 1997: 301–303, Fig. 2). The analysed Tell Iẓṭabba material is plain. The lead bullets range from 34 to 51 g in weight. These supplement several other such bullets (not yet fully published) from the site’s previous excavations, the weight of which ranges from 19 to 41 g (cf. Bar-Nathan and Mazor 1991: 53; W. Atrash and G. Mazor, personal communication).

Lead-made sling bullets are among the more characteristic Hellenistic weapon types dating from the mid-2nd/early 1st century BCE to be found in the Southern Levant. For example, one such plain sling bullet is apparently attested at Maresha (Bliss and Macalister 1902: 149, Pl. 79:13z), yet nothing was written about it. At Gezer, Macalister (1912: 370, Fig. 470) mentions sling bullets of lead (decorated with the motifs of a serpent or an anchor) or of clay. He adds that most of the bullets found were of hard compact clay or simply pebbles (attesting to the diversity of materials) and confined to the site’s two later strata. Earlier documentation of such finds is also known from Samaria (Reisner, Fisher and Lyon 1924: 326, Pl. 83:c2). At Akko, the dozen lead sling bullets found were attributed to the end of the 2nd century BCE; some of them bear a motif of a scorpion (Dothan 1976: 41, Fig. 48). At Dor, only a few sling bullets are reported from the excavations on the tell and they bear motifs such as an anchor (Stern 2000: 211, not illustrated), supplementing four other accidental finds (ranging from 27 to 60 g) from the site’s vicinity. These are either plain or decorated by motifs (a winged thunderbolt) or by inscriptions and motifs that are likely related to Tryphon (Schlesinger 1982), the Seleucid king who was besieged in the city by Antiochus VII (1 Macc. 15:10–14,25–26; Josephus, War I.50; Ant. 13.223–224). Several readings have been proposed for this unclear inscription (Schlesinger 1982; 1984; Gera 1995 ; Fischer 1992; see also Supplementum Epigraphicum Graecum 32: 1500, 35: 1535; Ameling et al. 2011: 2137). It is likely that it reads ΤΡΥΦΩΝΟ[C] ΝΙΚΗ, ‘the victory of Tryphon’, on the obverse, as supported by the fact that it was found outside the city walls and depicts a dove—the symbol of the city—above another, less clear, inscription on the reverse, which is beyond the scope of this synopsis (Gera 1995; see, however, Fischer 1992 for a different reading). Five more lead sling bullets were reportedly found on the western side of Tel Tanninim, exhibiting different motifs (winged thunderbolt, augur’s staff [lituus], anchor, cornucopia, jar) or legends, some in abbreviated form (among them a monogram of Tyre, above which appear the letters BA; see Schlesinger 1984; Ameling et al. 2011: 2092). Another group of sling bullets is stored at Bet Miriam, the Sea Museum (Marine Archaeology & Ecology) of Kibbutz Palmaḥim (Schlesinger 1991: Fig. 10.2; Ameling et al. 2014: 2274–2278). The group consists of 23 sling bullets which differ from one another in weight (ranging from 25 to 59 g). Half are plain and half decorated, mostly with either a winged thunderbolt or a scorpion and a cornucopia. Some slogans appear in the imperative, e.g., ΓΕΥCΑΙ (Nos. 12 and 20) ‘taste!’, ΔΕΞΑΙ (No. 21) ‘take!’, or with a monogram of Tyre and the letters BA above it with the relief of a reversed cornucopia (No. 14), similar to one from Tel Tanninim. The source of these sling bullets, although unclear, is likely to be local to the area of Dor/Tel Tanninim or even >Atlit (see, in this respect, Galili et al. 2016: 27) (contrary to what was argued in the publication).

Magen mentions 22 plain and decorated sling bullets from Mount Gerizim (with motifs of varied animals: a spider, a crab and a scorpion, as well as a trident or winged thunderbolts); he divided them into two groups, small (3–6 cm in length) and large (6–7 cm in length) and ascribed them to the Hasmonaean army occupying the site (Magen uncovered in excavations at the Citadel of Jerusalem, together with many other bullets made of clay. They bear the relief of a winged thunderbolt. On one, an arrow appears with a partial inscription in Greek (Sivan and Solar 1994: 173–174). Other lead sling bullets from Jerusalem, mostly plain, are known from the Armenian Garden (Tushingham 1985: 64, Fig. 70:34, Pl. 117 [with thunderbolt/trident]), the Jewish Quarter (Zitronblat and Geva 2003: 358, Pl. 14.3: M51), and the City of David (Yuzefovsky 2021: 640–641). Another six are reported from Maresha (Korzakova 2010), ranging in weight from 40 to 51 g. One is an inscribed projectile that likely reads ΗΡΑΚΛΕΟΣ on one side and ΝΙΚ[H] on the other. The other five might have borne symbols or legends that did not survive. Another sling bullet with a winged thunderbolt came from Sasa (West) (Smithline 1997: 22, Fig. 23). The finds from Transjordan are mostly confined to the site of Tulul adh-Dhahab, which is identified with Amathous (Josephus, Ant. 13.356). Several inscribed sling bullets are attested here (Dreyer 2016; Weiss 2017), including one that reads ΖΗΝΩΝΟC ΝΙΚΗ, ‘the victory of Zenon’, and could provide some historical context (i.e., Josephus, War I.88; Zenon’s struggle against Alexander Jannaeus), such as the above example of Tryphon. Others are either plain, inscribed (with slogans), or bearing motifs (e.g., winged thunderbolts).

It may be added that the bullets mentioned above were, in some cases, attributed to documented struggles. For example, in Dor and in Jerusalem, the excavators attributed the sling bullets to the armies of Antiochus VII who besieged these cities in 138/139 BCE and 132/133 BCE respectively, whereas at Tulul adh-Dhahab they are assigned to Zenon, the local Amonite ruler, who fought Alexander Jannaeus and the Hasmonaeans in 101/100 BCE. The motifs that appear on them—for example, the scorpion, the winged thunderbolt, the staff and the anchor—may be interpreted as the symbols of the military units of the slingers who used them and may also have manufactured them. At Mount Gerizim, on the other hand, projectiles with pagan motifs were assigned to the conquering Hasmonaean army, which may not necessarily be the case as their recovery within the walled settlement does not preclude their use by the defenders of the site—either a Seleucid garrison or local warriors. It is not unlikely that those recovered from Tell Iẓṭabba were also produced in the context of the site’s destruction in 107 BCE, although it remains unknown whether they were produced by the attackers or the defenders of the site.

Sampling and analysis methods and laboratory

Samples were taken by drilling with drill bits of 1 mm diameter. One sling bullet, No. 11 (IAA 91-6209/A [corpus], B [waste remain]) and three weights, Nos. 4, 6, 7 (IAA 91-6362/3A [corpus], B [handle]; IAA 91-6362/5A [corpus], B [handle]; IAA 91-6362/6A [corpus], B [handle]), were sampled twice for the different pieces of the object. From another one of the objects (No. 3, IAA 91-6362-2) with attached lead strips, three samples were drawn (A [corpus], B [handle] and C [lead strip]).

Analyses were performed in September 2021 at the Deutsches Bergbau-Museum Bochum (operator M. Jansen), using a portable X-ray fluorescence spectrometer (portable XRF) for elemental screening (Niton Xl3t GOLDD) and Multicollector-Mass Spectrometry for lead isotope measurements (Neptune Plus high-resolution MC-ICP-MS, ThermoFischer Scientific). The samples are of lead metal so that there was no need for a column separation prior to lead isotope analysis. The samples were dissolved in ultraclean HNO +HCl, then dried, and finally diluted to a 2% concentration of HNO .

Results

Portable energy-dispersive X-ray fluorescence spectrometry (portable ED-XRF)

A portable ED-XRF instrument was used to examine the lead composition of the samples (see below, Table 1). The method is suitable for affordable and rapid screening of the material to check the metal or alloy for major and minor element composition, while the analysis of trace elements requires more sensitive methods. Some conditions significantly limit the quality of analysis and quantification: only small sample spots (typically up to 3–8 mm) are analysed with the instrument, the video system limits the visual information of the sample spot, the geometry does not include a sample vacuum chamber, and the analyses are performed on surfaces that are without prior cleaning when used as a non-destructive archaeometric method. Weathered or altered surfaces (patinas) of metals may have certain elements either enriched or depleted, so that the analysis of the surface no longer reflects the original metal composition. For all these reasons, the method is severely limited to a semi-quantitative, if not semi-qualitative, analysis. In order to improve the method as much as possible, the uncorroded metal obtained as drill curls was measured here, so that at least compositional errors due to altered surfaces were avoided.

Table 1: Elemental composition of the lead samples and standard reference material by portable XRF^I

^I Samples ID are either Israel Antiquities Authority (IAA) Nos. (prefixed 91) or IAA excavation licence No. (prefixed G-17). <LOD = below detection limit = <3s error. When the detected and calculated concentration of the individual element is below 3s, it is noted as <LOD; note that the certified As content of PG standard reference material is < 0.02%.

As an energy dispersive method, portable XRF has another serious limitation, namely the overlapping of peaks in the X-ray spectrum. In the analyses performed (Table 1), this unfortunately has a particular effect on the detection of the major element lead with arsenic, since the arsenic Kα peak overlap almost completely. Therefore, the arsenic concentration determined here is certainly overestimated, if not completely incorrect. Test measurements of standard reference material with certified arsenic concentrations of less than 0.02% produced data with overestimated values of up to 20% As (Table 1). The results suggest that it is less the sample amount measured and more the sample geometry that has an influence, because larger metal curls led to a lower overestimation of arsenic, whereas smaller crumbs of metal led to a larger overestimation, thus feigning a higher arsenic content. For this reason, arsenic is not considered in the following discussion of the elements. It should not be omitted that in samples with substantial Sn concentration (n=3; 7–12 wt.% Sn), As contents are below detection limit. However, checking the complete set of analyses and considering the methodological problems of portable ED-XRF, this observation cannot result in a statistically proven correlation between As and Sn.

Although only a limited number of detected elements remain for discussion, some features can be quickly identified:

1. Detectable elements below the percentage range are Cu, Fe and Bi. Below the detection limit of the method are Ni, Zn and S.

2. The sling bullets are made of pure lead metal.

3. This is in contrast to the weights at which tin is detectable. Tin varies from major to (corpus)/12.3 wt% Sn (knob handle); moon-shaped weight No. 8 (IAA 91-6340) = 7 wt% Sn; seashell-shaped weight No. 4 (IAA 91-6362-3) = 3.4 wt% Sn (corpus)/4.0 wt% Sn (knob handle).

4. The attached lead strip to the bottom of seashell-shaped weight No. 3 (IAA 91-6362-2, sample C) is pure lead metal. Tin is below the detection limit.

Lead isotopes

Lead isotopes have already been used in different contexts to identify the metal sources of sling bullets (e.g., Vlad et al. 2011; Müller et al. 2015; Gomes et al. 2017; Forouzan et al. 2012; Cuchí-Oterino et al. 2021). They have proven to be a useful tool for identifying the raw metal sources of lead. The isotopic ratios of the examined sling bullets and weights from Tell Iẓṭabba (see below, Table 2) range from 18.637–18.846 ²⁰⁶Pb/²⁰⁴Pb; 15.657–15.678 ²⁰⁷Pb/²⁰⁴Pb; 38.723–38.866 ²⁰⁸Pb/²⁰⁴Pb. Weight No. 5 (IAA 91-6362-4) has lead isotope ratios of 18.057 ²⁰⁶Pb/²⁰⁴Pb; 15.6457 ²⁰⁷Pb/²⁰⁴Pb; 38.154 ²⁰⁸Pb/²⁰⁴Pb. Some observations can be made, which characterise isotopic differences of the lead metal (Fig. 4):

1. The sling bullets are made of isotopically homogeneous lead and form a very tight cluster. Identical isotope composition has the waste remain, which is attached to sling bullet.

2. The isotopic ratios of the weights are more scattered than those of the sling bullets, regardless of their particular shape.

3. Seashell-shaped weight No. 5 (IAA 91-6362-4) has a lead isotope signature significantly distant to the other weights.

4. The lead strip attached to the seashell-shaped weight No. 3 (IAA 91-6362-2) has a similar lead isotopic signature to the sling bullets and does not match the lead of the weights in composition.

Fig. 4: Lead isotope diagrams of objects; a trendline is drawn for the weights (triangles) (analytical errors are smaller than symbols)

Table 2: Lead isotope results of the analysed samples^I

^I Samples ID are either Israel Antiquities Authority (IAA) Nos. (prefixed 91) or IAA excavation licence No. (prefixed G-17).

Comparatively sampled and analysed were the corpus and knob handle of selected seashell-shaped weights, a lead strip, which is attached to the bottom of one of the weights (No. 3, IAA 91-6362-2C), and a sling bullet with attached waste remain, which is spilled lead on the surface. Lead isotope analysis shows that the corpora and knob handles were made of identical lead metal (Fig. 5), whereas the isotopic composition of the attached lead strip is different. It agrees with the lead isotope signature of the sling bullets. It was hence made of a lead metal different to that of the weight.

Fig. 5: Lead isotope diagrams for comparison of corpus bottom (A, triangles), knob handles (B, circles) and lead strip attached to the bottom of No. 3 (C, ) (analytical errors are smaller than symbols)

For the resource provenance of the weights and sling bullets their lead isotope ratios are compared with reference data of lead ores (Fig. 6). For a local perspective, the Arabah is intensely researched for geology, geochemistry and archaeology. This research dimension is in stark contrast to other ancient metal sources e.g., in Saudi Arabia. The Arabah is very rich in copper mineralisation (e.g., Hauptmann 2007), but is not a source of lead. In Mount Hermon, lead-zinc is mineralised in dolomite (Magaritz 1975 ), and a lead mine is reported, which is dated to the late periods (probably including the Hellenistic period, cf. Dar 1994). Available lead isotope reference data are almost exclusively from Timna and Feynan, so that lead isotope data from lead ores of the surrounding region are not available. The use of lead metal of a local source must therefore remain unanswered.

When considering lead imports based on knowledge of contemporary ore suppliers of the supra-region, contemporary archaeology, as well as political and trade relations, the focus should lie on Iran, Egypt, the Aegean (with a special focus on Laurion) and Anatolia. Reference data on lead ores from these regions were withdrawn from the recently published GlobaLID database (Westner et al. 2021 and original references herein), the OXALID database (Stos-Gale and Gale 2009 and original references herein) and our own compilation of published data from Turkey/Anatolia (see separate list in the references). We considered it useful to filter for lead ores, so that reference data are available to the following extent: Turkey/Anatolia (n = 268); Aegean (n = 1032; including 145 Lavrion data); Iran (n = 88); Egypt (n = 66).

When comparing objects and reference data in the lead isotope diagrams (Figs. 6–7), the isotopically homogeneous sling bullets agree with the Aegean lead ores and especially with galena from Lavrion (on Hellenistic lead exploitation in Lavrion, cf. Nomicos 2021: 112–113, 116). The weights scattering more widely agree with Aegean and Anatolian lead ores, which overlap and cannot be resolved even trying statistical methods (e.g., density distribution). The outlier, weight IAA 91-6362-4, is in close proximity to a few Anatolian data.

Fig. 6: Lead isotope diagrams of objects and ore minerals; for sources for reference data, see below, References section; all reference data are filtered for lead ores (the majority is reported as galena) (analytical errors of the objects are smaller than symbols, analytical errors for the reference data vary with the original publications)

Fig. 7: Lead isotope diagrams of objects and ore minerals: close-up (analytical errors of the objects are smaller than symbols, analytical errors for the reference data vary with the original publications)

Interpretation

The sling bullets were made of homogeneous and pure lead metal, which is different from the lead metal used to make the weights. The lead metal for the sling bullets is of Aegean origin. More detailed, their homogeneity agrees with the ore isotope field of Lavrion. In contrast, the weights are of a heterogeneous lead metal by containing variable tin contents. The elemental and isotopic heterogeneity of the weights could further point to either mixed or reused lead. The knob handles were potentially attached to the cast corpora to calibrate the weight. Since corpora and knob handles are respectively similar in their elemental and isotopic composition, it can be concluded that the calibration was performed in the same workshop and timely related to the casting process and that the knob handles were made from the same batch of metal as the corpora. The same applies to the lead ‘waste’ on the top of the sling bullet No. 11 (IAA 91-6209); it is similar in composition to the corpus of the sling bullet. Only in one case must the (re-)calibration of a weight have taken place at a later time: weight No. 3 (IAA 91-6362-2, Sample C). The metal used for the attached lead strip was from similar pure lead material such as the sling bullets. The weights, which isotopically scatter more widely, are consistent with Anatolian and Aegean ores. Overlap of the Aegean and Anatolian ore fields make further approval of this thesis impossible. Since lead ore deposits do not extend beyond single minor occurrences (e.g., Mount Hermon), local origin cannot be discussed here.

Of all the weights analysed, one seashell-shaped weight (No. 5, IAA 91-6362-4), differs significantly in isotopic composition but without significant elemental difference with portable ED-XRF. Therefore, this weight is considered to be another representative of a broader compositional spectrum of all weights produced or used in Tell Iẓṭabba, of which only a small window of objects could be analysed. The isotopic composition agrees with a small number of reference data from Anatolia. This might represent the endmember of a mixing line along which the weights plot (Fig. 4).

Conclusions

The portable ED-XRF and lead isotope analysis of 14 lead objects from Tell Iẓṭabba provided important results for a characterisation of the Seleucid settlement site. Whether both or one of the two groups of objects were produced in the short-lived settlement or imported cannot be clarified by the analyses. However, the lead metal of the objects came from different lead sources and was chosen deliberately for the various object groups. The sling bullets were deliberately made from pure metal, most probably with lead metal from Lavrion in Greece. The remains of cast channels on both short sides reveal that the sling bullets were not individually cast but rather produced in larger quantities en chapelet, in two-part closed moulds with the single hollows in a row being connected by channels. The purity and homogeneity of the material suggest an organised effort to produce them, perhaps even on one occasion in the context of military conflict. In 107 BCE the Seleucid settlement was destroyed by the Hasmonaeans in a fierce military conflict, and it is likely that the sling bullets should be attributed to this event. It remains unknown, however, whether they were prepared by the Hasmonaean attackers or by the Seleucid defenders, although due to the compositional homogeneity, it seems more likely to attribute them to only one of the two groups. Their Aegean origin suggests the Seleucid defenders, but this does not need to be taken for granted, since the composition differs from the civic weights and the Hasmonaeans too could have had access to lead from Aegean sources. The weights were made from mixed or reused lead metal from Aegean and/or Anatolian sources. The composition suggests that they were produced more on an ad hoc basis whenever they were needed and with the raw material available. They were calibrated in the workshop where they were cast (on site or external), but weight corrections became necessary at a later time. Since the lead strip attached to one of the weights has the same elemental and isotopic composition as the sling bullets, it can be assumed that weight corrections were made at the settlement, where the flows of weights and sling bullets converged.

Additional information

Notes on contributors

Sabine Klein

Sabine Klein: Deutsches Bergbau-Museum Bochum, Leibniz-Research Museum for Georessources, Bochum, Germany

Moritz Jansen

Moritz Jansen: Deutsches Bergbau-Museum Bochum, Leibniz-Research Museum for Georessources, Bochum, Germany

Achim Lichtenberger

Achim Lichtenberger: Institute for Classical Archaeology and Christian Archaeology, University of Münster/ Archaeological Museum, Münster, Germany

Oren Tal

Oren Tal: The Jakob M. Alkow Department of Archaeology and Ancient Near Eastern Cultures, Tel Aviv University, Israel

Notes

1 Although the artefacts were illicitly collected, their registration by the Israel Antiquities Authority made them ‘provenanced’ artefacts. It should be added in this context that 23 additional plain ‘slingshots’ are recorded in a yet unpublished report of the Israel Antiquities Authority’s 1991–1994 campaign at Tell Iẓṭabba; 21 of these came from Area W and the other two from Area Z. They are 19–41 g in weight, 30–36 mm long, 17–20 mm wide and 12–14 mm thick on average. Unfortunately, they were all lost after excavations (G. Mazor and W. Atrash, personal communication; see also Mazor and Atrash 2018).

2 See also https://pondera.uclouvain.be/artifact/11197/ and https://pondera.uclouvain.be/artifact/14015/; https://pondera.uclouvain.be/artifact/2711/.