Roman Militaria at Khirbet el-Maqatir

12. Militaria Katherine A. Streckert and Boyd V. Seevers The Late Hellenistic and Early Roman military artifacts at Khirbet el-Maqatir fit into seven groups: hobnails, slingstones and ballista balls, sling pellets, arrowheads, javelin heads, blades, and equestrian fittings. Though most of these artifacts came from Early Roman contexts (Stratum 3b), the dating of the whole collection may extend from the Late Hellenistic period to the mid-third century CE. Overall, the Early Roman militaria found at Khirbet el-Maqatir seems to support the excavators’ assertion that the Late Hellenistic and Early Roman settlement was founded in the second century BCE, destroyed by the Romans in 69 CE, and subsequently occupied by Roman soldiers until sometime before the Second Jewish Revolt (132–135 CE) when a Jewish population resettled the site. CE (pp. 377, 385, Groups H–O).1 However, the smaller and possibly later examples were very flat, suggesting significant wear. In contrast, the best-preserved and slightly worn examples were of medium head diameter (11–12 mm originally) and may date to 20–180 CE (Groups D–K). Their date range could even narrow to 40–80 CE, when 11–12 mm diameter heads seemed to be used exclusively (Groups E–F, pp. 360–61, 364–65, 369, 373).2 Three of the four hobnails reported from Gamla which dated securely to the Roman attack in 67 CE also had head diameters of 11 mm with worn head heights of 5–6 cm (Stiebel 2014, 80–81; Volken, Paccalot, and Volken 2011, 336). Thus, the best-preserved nails from Khirbet el-Maqatir could imply Roman activity at the time of both Jewish Revolts, possibly with special emphasis on the First Revolt. The excavators propose that the Romans attacked the settlement in 69 CE and that a small detachment of soldiers could have stayed there until sometime before the Bar Kokhba revolt, when a group of Jewish rebels appear to have reinhabited parts of the site (Peterson and Stripling 2017, 80*; Raviv, Stripling, and Farhi, 2021). Along with the numismatic and ceramic evidence, our interpretation of the site’s hobnails supports this historical reconstruction. Hobnails Similar to many sites displaying an Early Roman presence, Khirbet el-Maqatir produced many hobnails, also called sandal tacks or shoe tacks (fig. 12.1; see table). The excavators recorded 55 hobnails with head diameters of 3–20 mm and total lengths up to 19 mm. Nearly 50 additional hobnails were uncovered but not retained due to poor preservation. Throughout the Roman era, civilians and soldiers both wore hobnailed shoes. Though nailed footwear was strongly associated with the Roman military, the use of hobnailed shoes spread to the general population of Palestine. Jewish civilians wore hobnails until their prohibition, most likely during the Second Jewish Revolt (Roussin 1994, 188, 190; Mishnah Shabbat 6.2 and explanation in Babylonian Talmud 60a–b). The findings from the Qumran area, for example, suggest that Jewish civilians used hobnails at least into the Early Roman period (Stiebel 2003, 223). The rocky terrain of Judea lead to the advanced wear and loss of hobnails from Roman footwear (Stiebel 2015, 432), and thus, many sites yielded parallels. Notable sites from the First Jewish Revolt which record hobnails include Masada (Stiebel 2007, 1:372) and the Roman Camp A below Masada (Stiebel 2007, 2:3.20a/D), Gamla (Stiebel 2014, 80–81), Herodium (Stiebel 2003, 223; 2015, 432–34), and Jotapata (Stiebel 2007, 2:3.2/D.1). Many other sites in Palestine also exhibited hobnails from both before and after this time period. Given the apparently broad time span of Khirbet el-Maqatir’s collection of hobnails, the examples from Khirbet elMaqatir added most significantly to this corpus of First Revolt finds. Swiss archaeologist Marquita Volken has undertaken foundational research on the dating of hobnails based upon widespread consistency in nail measurements from the strata of a Roman road (Volken, Paccalot, and Volken, 2011). Few excavation reports note the exact dimensions of hobnails, and the research required to support Volken’s theories in a Near Eastern context would exceed the scope of this chapter, but it is useful to mention the application of Volken’s typology to the hobnails from Khirbet el-Maqatir. With nail heads ranging in diameter from 3 to 20 mm (not accounting for possible mass loss from wear or corrosion), Khirbet el-Maqatir may have displayed nails dating from 60 BCE (Volken, Paccalot, and Volken 2011, 365, Group A) to 285 1 The typology is divided into chronological groups based on weight and measurements, Group A being the earliest and Group P the latest (Volken, Paccalot, and Volken 2011, 336). Group designations in citations refer to the diagram in Volken, Paccalot, and Volken (336) and typology catalog detailed in Volken Paccalot, and Volken (356– 87). Additional page numbers reference specific examples within the catalog. 2 As with the group delineations, this time span is based on the typology in Volken, Paccalot, and Volken 2011 (diagram on p. 336, catalog on pp. 356–87). The advanced wear of many of the hobnails from Khirbet el-Maqatir makes the proposed original masses and head diameters somewhat subjective. Thus, the resulting dates are inconclusive. 300 12. Militaria Figure. 12.1. Hobnails from Khirbet el-Maqatir, ca. 60 BCE to 260 CE. Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. Khirbet el-Maqatir’s hobnails contributed additional detail to the site’s history. Three excavation areas revealed a concentration of hobnails (fig. 12.2). The area in and around a large first century CE dwelling at the center of town produced the largest concentration (26 hobnails). Added to the quantity of hobnails, a lapis lazuli die (see chap. 11, “Jewelry and Personal Accessories,” regarding Object 1476) was also discovered in this mansion. This suggests that the Romans may have occupied the building as a barracks or headquarters following the siege (Peterson and Stripling 2017, 80*). The northern fortified tower (see chap. 3, “Monumental Tower and Fortification System”) and a modified natural cave, designated Cavern 1 (CAV1), both yielded four hobnails. The findings of Cavern 1 were perhaps the most compelling (see chap. 4, “Subterranean Features”). Khirbet el-Maqatir’s inhabitants used this cave in the first century CE as an olive-oil processing installation, and when excavated, it was found to contain the remains of five or six human skeletons, along with a hiding tunnel in one wall.3 Cavern 1 contained the skeletons of an adolescent male and female, a woman of 20–30 years, a child, and an elderly adult. Unidentifiable remains of a sixth person, likely female, were also present. The connecting cave, Cavern 3, contained the remains of two more women—one 16–20 years old and one elderly (Wood 2018, 32). 3 301 The Excavations at Khirbet el-Maqatir: 1995–2001 and 2009–2016 Figure. 12.2. Map of the Early Roman town with hobnail concentrations circled in red. Drawing by Leen Ritmeyer; concentrations added by authors. Khirbet el-Maqatir was likely in the path of the Roman army as it marched south to Jerusalem in 69 CE. In fact, some researchers have proposed that Khirbet el-Maqatir was the town of Ephraim mentioned by Josephus in his account of Vespasian’s conquest of the Judaean highlands (Peterson and Stripling 2017, 82*– 83*). If this identification is accurate, then according to Josephus, Vespasian’s army would have come from Caesarea into the hill country, decimating the districts of Gophna and Acrabata before taking Bethel and Ephraim (i.e., Khirbet el-Maqatir). Josephus records that Vespasian left garrisons in these towns before continuing to Jerusalem, supporting the idea that Roman soldiers were placed at Khirbet el-Maqatir after 302 12. Militaria its fall (Jewish War 4.550–51). Evidently, some women, children, and elderly citizens of Khirbet el-Maqatir retreated to the underground olive-press-turned-hiding complex during the attack, where the Roman garrison eventually found and killed them. The hobnails found in the cave argue for Roman involvement in the fates of these civilians and their hometown, though the Jewish population may have also worn hobnails at this time. It is plausible that the Roman soldiers occupied Khirbet el-Maqatir until the close of the First Revolt or at least until the fall of Jerusalem in 70 CE when they may have been recalled to Jerusalem. As presented in chapter 4, the subterranean hiding system was also used by rebels in the Bar Kokhba revolt. Figure. 12.3. 6-libra ballista ball. Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. Slingstones and Ballista Balls ballista balls fit libra calibers, this suggests that one of the two legions used in the revolt that had come from the west—V Macedonia and XV Apollinaris (Holley 2014, 46) likely attacked Khirbet el-Maqatir. This smaller stone came to light at the peak of the hill where the Byzantine monastery was later situated, some 120 meters to the northwest. It was discovered with a slingstone and mostly Byzantine pottery. Though this ballista ball may have been out of context, it is quite logical that the Romans camped on this high point and made sling and ballista balls there before attacking the western wall where the larger ballista ball was found. Perhaps the Roman soldiers attacked the city wall on the west and the tower on the north and then converged at the mansion at the center of town. Three hundred rounded stone balls came to light in the Khirbet el-Maqatir excavations, distributed throughout Bronze Age, Iron Age, Late Hellenistic, and Early Roman strata. While the largest was 1.96 kg, most balls weighed 250–350 grams. The majority were flint, though some were made of limestone, including the largest stone. They all exhibit a pecked finish, the result of chipping the stone into shape with a hard implement, such as another stone. Khirbet el-Maqatir’s long history complicates the dating of the stones, as the majority of the slingstones were found in mixed contexts. Due to evidence of a Late Bronze Age battle at the site and the fact that many balls came to light in a Bronze Age stratum (Stratum 7), excavators originally assumed that the majority were slingstones from this era, reused as projectiles or pounders in subsequent periods. However, some of the balls under 655 grams could have also been slingstones from the Roman attack. Likewise, stones weighing over 655 grams could have been Early Roman ballista balls (Seevers, forthcoming; Stiebel 2013a, 299–300). Two stones from Khirbet el-Maqatir weighed-in above this 655-gram cutoff (see table). When besieging small settlements and large fortresses alike, the Romans were known to use war machines, such as ballistae (Stiebel 2005, 100). Khirbet el-Maqatir, with its towers and walls, could well have been such a target. Though the evidence of the few ballista balls and the artifacts found with them was too meager to confirm the plan of attack used by the Romans in the siege of Khirbet el-Maqatir, the corpus does at least suggest that this reconstruction is possible. Other sites reporting 6-libra ballista balls include Gamla (Holley 2014, 39) and Masada (1994, 357), while Herodium (Stiebel 2003, 217), Jotapata (2007, 2:3.2/M), and the fortified settlement of Meroth (2005, 100; 2007, 2:3.1/M.1) yielded ballista balls of other small calibers. The largest rounded stone from Khirbet el-Maqatir, which weighed 1.96 kg and had an 11 cm diameter, was very likely a Roman ballista ball (fig. 12.3). It was recovered along with four slingstones and a hobnail just outside the apparent Late Hellenistic and Early Roman perimeter wall on the western side of the site, at the high point of the wall in that area. The pottery found with this artifact was mostly Middle and Late Bronze, though two Early Roman sherds were present. At 1.96 kg, the ball weighed exactly 6 libra, one of the smaller calibers listed by the Roman military engineer Vitruvius (Ten Books on Architecture 10.11.3, in Marsden 1971, 191), suggesting that the Romans may have made it for the attack on Khirbet el-Maqatir in 69 CE. The other ballista ball, ca. 90 percent complete, weighed 0.92 kg, suggesting an original weight of approximately 1.012 kg, just over 3 libra (0.987 kg). Holley (2014: table 3.5–6) lists 3 libra as a common caliber. Since both Sling Pellets In addition to slingstones, the excavators discovered one lead sling pellet at Khirbet el-Maqatir (fig. 12.4; table). The majority of the pellets of this kind found in Palestine are Hellenistic, though Early Roman examples are also documented, dating at least to the late first century CE. Elsewhere, the Romans used lead sling pellets into the second century CE (Stiebel 2013a, 299; Bishop and Coulston 2006, 135). Because of Khirbet el-Maqatir’s extended occupation and the mixed Late Hellenistic and Early Roman context in which the 303 The Excavations at Khirbet el-Maqatir: 1995–2001 and 2009–2016 Lead sling pellets have been found at many sites in Palestine, including at Gamla (Stiebel 2014, 98), Jerusalem (2007, 2:3.14/K.1–2; Sivan and Solar, 1994, 173), Jotapata (Stiebel 2007, 2:3.2/K), and Jericho (2013b, 293). The pellet from Jericho is the closest parallel for the one from Khirbet el-Maqatir, with its hammered surface and very similar dimensions (3.6 × 2.3 × 1.9 cm). Arrowheads Figure. 12.4. Lead sling pellet; hammer marks and apparent break from mold sprue on upper left. Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. Khirbet el-Maqatir produced five arrowheads from the Late Hellenistic or Early Roman periods, four of which (Objects 1230, 2425, 2429, and 3037) appeared to be of the Roman bodkin-tanged variety—square or triangular in cross section and approximately 4 cm long (Coulston 1985, 265) (fig. 12.6; table). Khirbet el-Maqatir’s most well-preserved examples (Objects 2425 and 2429) measured 4.8 cm and 4.4 cm in length and exhibited a clear square cross-section. Object 2429 appears to have been bent upon impact. Both arrowheads were found in or near the northern Roman tower entrance, with Object 2425 wedged between the stones of the entryway. Object 1230 measured 3.9 cm long and is another clear example of a square bodkin arrowhead, though it is quite corroded and cracked, likely due to differing alloy concentrations within the metal. It was discovered in Cavern 3, a small auxiliary chamber of the Cavern 1 hiding complex, directly across from the exit of the tunnel connecting the two caves. Cavern 3’s context was plainly Early Roman, and the arrow appeared in the same location as a collection of human bones. Its placement suggests that the arrow may have been fired artifact came to light, this pellet could have been either Hellenistic or Roman. While words and symbols were often molded on Greek and Republican Roman sling pellets, the practice ended by the mid-first century CE (Stiebel 1997, 302; Bishop and Coulston 2006, 58; Feugère 2002, 160). No clear sign of inscription appeared on this pellet, perhaps suggesting an Early Roman date (Stiebel 2007, 2:3.14).4 Sling pellets were typically made in a two-part mold with biconical cavities connected by sprues, which formed a tree-like product (fig. 12.5). Lead was poured into the mold and cooled, and then the pellets would be broken off the “branches” for use (Stiebel 1997, 301). Such a telltale scar is present on one end of Khirbet el-Maqatir’s sling pellet. Additionally, its surface was rough, suggesting it was hammered after molding to perfect its biconical shape. Figure. 12.5. Reconstructed sling pellet mold (after Stiebel 1997, fig. 2). Drawing by Katherine A. Streckert. 5 Figure. 12.6. Roman arrowheads: 1, Object 3037 (possible bodkin tanged); 2, Object 2425 (bodkin tanged); 3, Object 2429 (bodkin tanged); 4, Object 1018 (possible flat-bladed); 5, Object 1230 (bodkin tanged). Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. The authors are grateful to Roman military scholar Raffaele D’Amato for his help in the dating and identification of several artifacts discussed in this chapter, including the sling pellet, sica and socketed blades, and possible harness phalerae. 4 304 12. Militaria Javelin Heads into Cavern 3 by an entering Roman soldier, possibly aimed at one of the individuals whose remains were uncovered there. Parallels of square bodkin arrowheads similar to those from Khirbet el-Maqatir come from Gamla (Magness 2014, 28–30; Stiebel 2005, 100), Magdala (Stiebel 2007, 2:3.4/I.1), Meroth (2:3.1/I.4–6), and the City of David (2013a, 297). Another possible bodkin arrowhead from Khirbet el-Maqatir (Object 3037) was half the size of the others at 2 cm, and its square crosssection was not as clear. Though smaller, the head bore a striking resemblance to a bodkin arrowhead found at Ein Feshkha (Stiebel 2007, 2:5.5/1). Both Objects 2425 and 3037 appeared in clearly Early Roman contexts. Though Object 2429 came to light in a combined Late Hellenistic and Early Roman context, its characteristic bodkin form dictates a Roman dating. Object 1018 may be a bent flat-bladed arrowhead, but not enough information was available to say conclusively (Coulston 1985, 265). The artifact measured 4.7 cm but was likely longer originally, as it seemed to be missing a tang or socket. Its context was mixed Late Hellenistic and Early Roman. Similar flat-bladed arrowheads came from a Hasmonean or Herodian fortress at Ein Rachel (Stiebel 2007, 2:4.6/1), Gamla in a First Revolt context (“Type B,” in Magness 2014, 24–25), and Cave 1 of the Second Revolt site Wadi Murabba’at (Stiebel 2007, 2:5.16/1.6). One possible Roman javelin head was uncovered in the floor of the northwest Early Roman tower (fig. 12.7; table). It measured 7 cm long and 1.3 cm wide at its center, weighing 11.54 grams. It lacked a socket or tang. Though the terms javelin and spear are similar, javelins were light and meant for throwing, while spears were most often used for close combat (Stiebel 2007, 1:138). Javelin and spearheads could range from 6–8 cm to 40 cm, and the example from Khirbet el-Maqatir falls on the lower end of this spectrum, even if one adds a presumed tang or socket (Feugère 2002, 132). However, in the Gamla excavation reports, Magness classifies projectile heads over 12 grams as spearheads, increasing the weight by 20 percent to account for corrosion. At an adjusted 13.85 grams, the Khirbet el-Maqatir projectile just fit within this category (Magness 2014, 30). It was found within the northern tower, contending for an identification as a javelin or small spearhead rather than a civilian implement. The types of arrowheads recovered at Khirbet el-Maqatir are intriguing. As the trilobate-tanged arrowhead was the most common in the Roman Empire and was especially associated with the Jewish Revolts, being used by both the Romans and Jews, it was surprising that none appeared at Khirbet el-Maqatir (Stiebel 2003, 216; Coulston 1985, 264). Even more surprising was that bodkin arrowheads were the primary type uncovered at the site. While trilobate-tanged arrowheads were best for unarmored targets, bodkin arrowheads are thought to have been used as armor piercers. This opens the discussion of who shot the bodkin arrowheads at Khirbet el-Maqatir. Though it is possible that the Jewish people of Khirbet el-Maqatir possessed some sort of armor and that the bodkin arrows were aimed at them for this reason, it is not probable. The Jewish rebels during the revolts were generally unarmored (Coulston 1985, 268; Stiebel 2007, 1:219, 2005, 100). On the other hand, while the Jews were skilled in archery, they typically used trilobate, flat bladed, bone, or recycled Persian heads (Magness 2014, 24; Stiebel 2007, 1:219– 20). It is possible, though perhaps not probable, that the bodkin arrows from Khirbet el-Maqatir were used by the Jews against the armored Romans after they had taken possession of the tower, though the arrows must have come from a Roman source. Whether either of these two explanations is correct, it seems that, unless arrowheads were collected after the siege of Khirbet elMaqatir, archery was not a main feature of the attack, taking into account the relatively few clearly Roman arrowheads which were found. Figure. 12.7. Broken javelin-head. Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. Parallel javelin heads of comparable size from the First Revolt came from Cave FQ37 near Khirbet Qumran (Stiebel 2007, 2:3.18/H.1) and Masada (2:3.19/H.2). Inter-revolt and Second Revolt examples of similar dimensions are known from Kurnub (2:4.3), Ein Rachel (2:4.6/H.1–2), and Tel Shalem (2:5.2/H.1–11), though these later artifacts seem to be Nabatean. Blades Khirbet el-Maqatir yielded an impressive number of poorly preserved apparent blade fragments from Late Hellenistic and Early Roman contexts. Lengths ranged from a tip piece 1.1 cm long to a large middle section measuring 16.8 cm (table). Most came out of mixed Late Hellenistic and Early Roman contexts, though four (nos. 1128, 1219, 2841, and 3097) exhibited primarily Early Roman pottery. Especially intriguing pieces included a partial blade with socket (Object 1048) and two partial blades with tangs (Objects 2559 [two pieces] and 2926) 305 The Excavations at Khirbet el-Maqatir: 1995–2001 and 2009–2016 Figure. 12.8. Blade fragments (Objects 1048, 2559, and 2926). Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. (fig. 12.8). Though Object 1048 could have conceivably been a knife blade or spearhead, its opened socket was reminiscent of a catapult bolt. It resembled bolts from Gamla (Magness 2014, 26) and Jericho (Stiebel 2007, 2:1.4/M.1b). However, only a piece of Object 1048 was preserved, making it difficult to determine whether its head had the typical square cross-section of a catapult bolt. The second of the latter blades was straight-backed, while the other was somewhat curved. Jews in the revolts used straight-backed daggers, though perhaps with longer blades than these (Stiebel 2007, 1:110, 112). Interestingly, both knives from Khirbet el-Maqatir were found in or around the first-century CE mansion. Perhaps they were Jewish weapons, though they could have also been domestic tools, or even tools employed as weapons during the Roman siege. It is also possible that these blades could have been brought and used by Roman soldiers after the attack, especially if they later occupied the mansion. Object 1049 may be a rare fragment of a sica sword, a short, curved sword especially associated with the Jewish rebel group, the sicarii, in the First Jewish Revolt (Stiebel 2007, 1:112–13) (fig. 12.9). Only two other sica blades have been discovered in Palestine, one from Khirbet Qumran (Stiebel 2007, 2:3.17/F) and another from Nahal David, Cave 2 (2:1.8/F). The latter is a close Figure. 12.9. Possible sica blade fragment (Object 1049). Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. 306 12. Militaria Figure. 12.10. Possible harness phalerae. Left, top view of Object 1590; right, bottom view of Object 1588. Photograph by Michael C. Luddeni. parallel to Khirbet el-Maqatir’s sica fragment, with the same intact pin (which would have originally attached the handle) as well as remnants of a central rib and a similar, somewhat curved shape. Whether the sica blade from Khirbet el-Maqatir signals the presence of sicarii at the site is not clear, though this would have further motivated the Romans to attack the town. This artifact and the socketed Object 1048 were found with three other iron fragments. It is possible that this was a collection of weapons, as several seemed to be corroded blades. However, due to the proximity of where the group was found (Square S19) to the ruins of the Late Bronze gate, which apparently functioned as a wine press and industrial installation during the Late Hellenistic and Early Roman periods, it may be that the artifacts in this cache were simply tools or were scrap metal consolidated for melting. Equestrian Fittings Figure. 12.11. Likely a girth buckle, Object 866. Photograph by Michael C. Luddeni; drawing by Katherine A. Streckert. Two small metal objects came from Khirbet el-Maqatir, bearing a distinctive eight-petaled flower motif (fig. 12.10; table). The identical pair of artifacts came to light in a silo inside the first-century CE mansion. Both were domed on top and nearly flat on the bottom, with a hole in the center. Though nearly half the size of the usual Roman equestrian phalerae and lacking the typical surrounding flange (Bishop 1988, 95), they may be a type of non-standard phalera, perhaps attached with a rivet to decorate the narrow reins or muzzle. A small harness phalera (3.2 × 1.0 × 1.2 cm) from Samaria had a similar flower pattern (Kenyon 1957, fig. 108:6; Stiebel 2007, 2:1.6/Q.1–2). Roman harness phalerae were generally made of copper alloys, some bearing a white silver coating (Bishop 1988, 94). Taking into account the artifacts’ coloration, the possible phalerae from Khirbet el-Maqatir may be lead or perhaps bronze or brass with remnants of silver.5 A large iron buckle found at the site appeared to be another fitting from equestrian gear, a girth buckle (fig. 12.11; table). Girth buckles tightened the saddle around the horse’s middle, fastening the saddle’s strap under the belly. Unlike harness fittings, which were almost always made of copper alloy, other horse equipment, including the girth buckle, were generally iron (Stiebel 2007, 1:373, 375). The buckle from Khirbet el-Maqatir was the correct width (6.7 cm wide) and general shape to be a Roman girth buckle (Bishop 1988, fig. 36:4 and table 2), though perhaps slightly taller than usual. A D-shaped girth buckle from Khirbet Qumran was remarkably close in size to the Khirbet el-Maqatir example, measuring 6.6 × 5.8 cm (Stiebel 2007, 2:3.16/ Q1). 5 Much of the above information was adapted or excerpted from Hassler, Streckert, and Seevers 2020. 307 The Excavations at Khirbet el-Maqatir: 1995–2001 and 2009–2016 Table. Selected inventory of militaria from Khirbet el-Maqatir Artifact & no. Size (cm) Weight (g) Field Square Locus Pail ... A O22 2 9 1.73 A O21 17 29 1.01 B W22 6 8 2.35 F A24 0.94 B Q25 3 6 ø 0.9 × 0.8 0.73 F AB23 2 10 1.68 B P22 10 75 ø 1.6 × 0.7 2.97 F F25 2 8 1,960.00 E Q8 4 11 920.00 C ZH05 6 4 72.60 G P18 2 25 Hobnails 1356 1444 2561 2577 2683 2722 2835 2870 ø 1.2 × 0.6 ø 1.3 × 1.1 ø 1.4 × 1.0 ø 2.2 × 1.4 ø 1.1 × 1.2 ø 1.6 × 1.2 Surface Surface Ballista balls 443 ø 11.0 698 ø 8.0 Sling pellets 553a 3.9 × 2.1 × 1.7 Arrowheads 1018 1230 2425 2429 3037b 4.7 × ? × 0.8 3.9 × 0.5 × 0.5 4.8 × 0.6 × 0.6 4.4 × 0.7 × 0.7 ... G R19 8 15 ... A CAV3 2 2 2.13 I X23 4 39 2.24 I X22 4 5 2.0 × 0.5 × 0.3 1.06 B Q25 12 30 7.0 × 1.3 × 0.6 11.54 I X23 14 31 4.4 × 1.1 × 0.4 3.60 B P21 3 25 Javelin heads 2342 Knife blades 776 916 1046 1047 1048 1049 1128 1219 1703 1828c 2157 2158 2559 2705 2841 2926 2945 3097 Equestrian fittings 866 1548 1550 6.5 × 3.1 × 0.1 16.40 G R19 3 7 10.0 × 3.0 × ? ... G S19 4 16 ... G S19 4 16 10.0 × 2.0 × ? ... G S19 4 16 10.1 × 3.0 × ? ... G S19 4 16 1.61 A M28 4 4 6.0 × 2.1 × 0.6 5.28 A CAV1 7 19 4.4 × 3.2 × ? 2.7 × 1.0 × 0.3 5.4 × 1.5 × 0.3 8.16 A O23 17 22 16.8 × 3.2 × 0.9 118.72 A CAV1 33 68 10.85 A O24 11 15 7.5 × 2.7 × 1.0 29.58 A O24 12 16 4.48 B P24 4 7 6.0 × 2.2 × 0.3 6.29 B Q22 16 29 10.16 B W22 11 37 13.36 B P22 8 100 0.28 B R24 1 1 6.53 B Q25 2 24 84.50 C ZI10 7 7 12.06 A O22 6 16 12.61 A O22 6 16 3.0 × 3.8 × 0.9 6.0 × 1.7 × 0.7 7.1 × 2.1 × 0.5 3.5 × 1.5 × 0.5 1.1 × 0.6 × ? 4.0 × 1.9 × 0.2 6.7 × 6.2 × ? 1.8 × 1.8 × 0.8 1.8 × 1.8 × 0.8 A040880. b A045303. c Perhaps too thick for a blade. a 308 12. Militaria Appendix: History and Manufacture of Early Roman Militaria the shoes more comfortable for walking long distances. In addition to being fashionable and promoting comfort, hobnail patterns could identify the wearer by his print, and some may have been meant to offer cosmic protection, as seen in the use of arrow symbols, perhaps linked with the god Mercury, and Neptune’s tridents (fig. 12.13) (Driel Murray 1999, 132–34; Bishop and Coulston, 2006, 112; Burandt 2016, 14). Hobnails Caligae, perhaps the most well-known type of Roman military footwear, were in fact boots with heavily hobnailed soles, though they featured an intricately latticed, sandal-like upper. Though when exactly caligae were first worn is unknown, they were common at least by the Germanic campaigns of Augustus and Tiberius. Gaius, Tiberius’s adopted son and successor, was given the nickname Caligula or Little Boot when he accompanied his father on campaign dressed in miniature military attire (Sumner 2009, 193). These distinctive military boots were synonymous with the Roman military from the early empire until the beginning of the second century CE (Goldman 1994, 122; Driel Murray 1987, 33). Caligae were made of three leather parts: the inner sole, the middle sole with latticed upper, and the outer sole—clenched together with hobnails (fig. 12.12). While a perimeter of nails held all three pieces of the shoe together, additional tacks were driven through the outer soles for durability and traction. Thus, the outer sole was studded with nails, and the inner sole provided a barrier between them and the foot. The shoe was replaced when the sole wore down enough for the bumpy nails to become uncomfortable or when the nails themselves had worn down to the point of uselessness (Goldman 1994, 122; Sumner 2009, 198–99; Bishop and Coulston. 2006, 113). A document found in Egypt records two soldiers receiving three pairs of shoes each per year (Sumner 2009, 197) and a pay record from Masada notes the deduction of five denarii from a soldier’s wage for caligae (Stiebel 2003, 223). Though caligae were often discarded rather than repaired, soldiers also replaced hobnails which fell out while marching (Volken, Paccalot, and Volken 2011, 316–17; Sumner 2009, 197; Driel Murray 1999, 137). According to Tacitus, soldiers may have received an allowance to buy replacement nails, called a clavarium (Histories 3.50). The volume and prevalent loss of hobnails explains why they are such a ubiquitous Early Roman small find. Hammered through the desired layers on an anvil, the tip of the nail curled back into the leather, fastening itself in place and giving the shoe nail its identifiable hooked end (Driel Murray 1995, 6; Sumner 2009, 194; Volken, Paccalot, and Volken, 2011, 316). The nonstructural nails of the outer soles were arranged in patterns, the frequency of which in specific time periods suggests something akin to fashion trends (swastikas, circles, and S-shapes seemed to have been popular). Patterns which supported the ball and heel of the foot often appear on soldiers’ caligae and presumably made Much like that of caligae, the manufacturing process of hobnails was highly uniform. Nails sold by piece, though a set number were made per unit of weight (based either on the Roman ounce or a variant of the Greek obal, depending on where in the empire the nail was made). Hammering a predetermined weight of raw iron into a square rod, the end of the rod was sharpened into the nail shaft. Then the piece, weighing a specific fraction of the entire rod, was detached. The shaft was then inserted into a nail header, a circular tool with a square hole which held the nail while the head was fashioned. The surface of the nail header was curved, and the examples of early nails suggest that some nail headers had incised Figure. 12.12. Cutout of Roman caliga with hobnail pattern (after Bishop patterns of dots and lines and Coulston, 2006, fig. 64); Assembled on their faces, which caliga. Based on: https://sutor.jimdo. molded the underside com/1st-century-ad/mainz-caligaof the nail head. These no-9/. With kind permission from markings may have Martin Moser. been the trademarks 309 The Excavations at Khirbet el-Maqatir: 1995–2001 and 2009–2016 Figure. 12.13. a, “S”; b, swastika; c, trident (after Driel Murray 1999, fig. 1); d, pattern for support (after Burandt 2016, fig. 2.1). helmets, body armor, and shields (Feugère 2002, 85). Considering this, it seems that most ballista balls were employed for anti-personal uses, while the largest ones could break through battlements and towers (Holley 2014, 52; Stiebel 2003, 217; Bishop and Coulston, 2006, 89). The use of ballista balls for primarily anti-personal purposes is seen at both Herodium and Masada (Stiebel 2003, 217). Josephus (Jewish War 5.271–72) records that the Romans learned to blacken the ballista balls so that their targets would not see them coming (Stiebel 2003, 221–23; Stiebel 2005, 100; Feugère 2002, 169). of the manufacturing network (Volken, Paccalot, and Volken 2011, 324). After inserted into the header, the head was then formed into a cone or a pyramid, either by hammering it into shape or by using a mold (Volken, Paccalot, and Volken 2011, 322–23, 332–33). As the price of iron went up, so did the number of nails fabricated per unit (Volken, Paccalot, and Volken 2011, 337). Thus, the earliest nails were made six per Roman ounce (weighing 4.54 grams, with heads no less than 25 mm in diameter) and the latest were 60 per ounce (weighing as little as 0.41 grams, with head diameters as small as 5 mm) (Volken 2014, 182; Volken, Paccalot, and Volken 2011, 333). Signature of Roman imperial footwear, the nailing of shoes continued to the end of the fourth century CE. After this, one-piece, sewn shoes became the variety of choice until hobnailed shoes were reintroduced to the West in the late seventeenth and early eighteenth centuries (Driel Murray 1987, 39– 40; Volken, Paccalot, and Volken 2011, 319–20). Soldiers made ballista balls with a chisel and hammer at the site of the siege (Stiebel 2003, 220; 2005, 100; Holley 2014, 37). Ballistae were built based on the weight of the stones they were meant to throw. This weight could be based on Roman libra or Attic-Euboic mina, depending on which system of measurement the soldiers were most comfortable with (Holley 2014, 25; Feugère 2002, 169). While slingstones were a maximum of 655 grams, ballista balls likely weighed from 655 grams to 26 kg (Stiebel 2003, 218; 2013a, 299–300). Vitruvius (Ten Books on Architecture 10.11.3, in Marsden 1971, 191) recorded a list of possible ballistae calibers which range from 2 librae (0.655 kg) to 360 librae (117.900 kg) (Holley 2014, 45). Philon (On Making Missiles 51.21, in Marsden 1971, 109) also developed a list of ballistae calibers in Greek mina, though his only spans from 10 mina (4.366 kg) to 3 talents (78.588 kg) (Holley 2014, 45). For hobnail parallels, see Volkin, Paccalot, and Volken (2011, 336, 356–87). Objects 1356 (see Groups I–O, 120– 285 CE), 1444 (Groups D–K, 20–180 CE), 2561 (Groups D–O, 20–285 CE), 2577 (Groups C–I, 25 BCE–140 CE), 2683 (Groups D–K, 20–180 CE), 2722 (Groups G–M, 80–235 CE), 2835 (Groups C–L, 25 BCE–215 CE), and 2870 (Groups I–O, 120–285 CE). The date ranges for Groups I–O and C–L are expanded, taking into account possible mass loss. Ballista Balls Ballista balls are common finds at sites exhibiting conflict in the Hellenistic and Roman eras. While soldiers likely collected metal arrowheads and catapult bolts for reuse after battle, ballista balls and slingstones were often left, as their large size and weight made them difficult to transport (Holley 2014, 37–38; Stiebel 2005, 101). Vegetius, though writing in the fourth century CE, suggests that each Roman legion had 10 stone-throwing siege machines (Holley 2014, 46; Bishop and Coulston, 2006, 88). The ballistae hurled stones with such force that no armor could withstand them, the stones penetrating 310 12. Militaria References Bethel Hills.” Near East Archaeological Society Bulletin 65, 1–23. Roussin, Lucille, 1994. “Costume in Roman Palestine: Archaeological Remains and the Evidence from the Mishnah.” In The World of Roman Costume, edited by Judith Lynn Sebasta and Larissa Bonfante, 182–90, Madison, WI: University of Wisconsin Press. Seevers, Boyd. Forthcoming. “Small Finds.” In Excavations at Khirbet el-Maqatir, 1995–2001 and 2009– 2016. Vol. 1, The Bronze and Iron Ages, by Bryant Wood and Boyd V. Seevers. Oxford: Archaeopress. Sivan, Renée, and Giora Solar. 1994. “Excavations in the Jerusalem Citadel, 1980–1988.” In Ancient Jerusalem Revealed, edited by Hillel Geva, 168–76. Jerusalem: Israel Exploration Society. Stiebel, Guy D. 1997. “‘. . . You Were the Word of War’: A Sling Shot Testimony from Israel.” In L’équipement militaire et l’armement de la république (IVe-Ier s. avant J.-C.): Proceedings of the Tenth International Roman Military Equipment Conference, Held at Montpellier, France, 26th–28th September, 1996, edited by M. Feugère, 301–7. Journal of Roman Military Equipment Studies 8. Oxford: Oxbow. Stiebel, Guy D. 2003. “The Militaria from Herodium.” In One Land, Many Cultures: Archaeological Studies in Honour of Stanislao Loffreda, OFM, edited by G. Claudio Bottini, Leah Di Segni, and L. Daniel Chrupeala, 215–43. Studium Biblicum Franciscanum, Collectio major 41. Jerusalem: Franciscan Printing Press. Stiebel, Guy D. 2005. “ ‘Dust to Dust, Ashes to Ashes . . .’: Military Equipment from Destruction Layers in Roman Palestine.” In Archäologie der Schlachtfelder: Militaria aus Zerstörungshorizonten; Akten der 14. Internationalen Roman Military Equipment Conference (RoMEC); Wien, 27–31 August 2003, edited by Werner Jobst, 99–108. Carnuntum Jahrbuch 2005. Vienna: Österreichischen Akademie der Wissenchaften. Stiebel, Guy D. 2007. Armis et litteris: The Military Equipment of Early Roman Palestine, in Light of the Archaeological and Historical Sources. 3 vols. PhD diss., University of London. Stiebel, Guy D. 2013a. “The Military Equipment.” In Excavations in the Tyropoeon Valley (Givʿati Parking Lot): Final Report, edited by Doron Ben-Ami, 1:297– 304. IAA Reports 52. Jerusalem: Israel Antiquities Authority. Stiebel, Guy D. 2013b. “The Military Equipment from the Hasmonean and Herodian Palaces at Jericho and Cypros.” In Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations. Vol. 5, The Finds from Jericho and Cypros, by Rachel BarNathan and Judit Gärtner, 290–98 Jerusalem: Israel Exploration Society. Stiebel, Guy D. 2014. “Military Equipment.” In Gamla III: The Shmarya Gutmann Excavations, 1976–1989; Finds and Studies, edited by Danny Syon, pt. 1, 57–107. IAA Reports 56. Jerusalem: Israel Antiquates Authority. Bishop, Mike C. 1988. “Cavalry Equipment of the Roman Army in the First Century A.D.” In Military Equipment and the Identity of Roman Soldiers: Proceedings of the Fourth Roman Military Equipment Conference, edited by Jonathan C. N. Coulston, 67–196. BAR International Series 394. Oxford: B.A.R. Bishop, M. C., and J. C. N. Coulston, 2006. Roman Military Equipment: From the Punic Wars to the Fall of Rome. 2nd ed. Oxford: Oxbow. Coulston, J. C. N. 1985. “Roman Archery Equipment.” In The Production and Distribution of Roman Military Equipment: Proceedings of the Second Roman Military Equipment Research Seminar, edited by M. C. Bishop, 220–336. BAR International Series 275. Oxford: B.A.R. Feugère, Michel. 2002. Weapons of the Romans. Translated by David G. Smith. Charleston, SC: Tempus. Goldman, Norma. 1994. “Roman Footwear.” In The World of Roman Costume, edited by Judith Lynn Sebasta and Larissa Bonfante, 101–32. Madison, WI: University of Wisconsin Press. Hassler, Mark A., Katherine A. Streckert, and Boyd V. Seevers. 2020. “A Monumental Fortification Tower and Militaria: Late Hellenistic and Early Roman Military Architecture and Equipment Discovered at Khirbet el-Maqatir, Israel.” In the Highland’s Depth Depth 10, no. 1 (Spring): 37*–69*. Holley, Andrew E. 1994. “The Ballista Balls from Masada.” In Masada IV: The Yigael Yadin Excavations, 1963–1965; Final Reports, edited by Joseph Aviram, Gideon Foerster, and Ehud Netzer, 349–65. Jerusalem: Israel Exploration Society. Holley, Andrew E. 2014. “Stone Projectiles and the Use of Artillery in the Siege of Gamla.” In Gamla III: The Shmarya Gutmann Excavations, 1976–1989; Finds and Studies, edited by Danny Syon, pt. 1, 35–55. IAA Reports 56. Jerusalem: Israel Antiquities Authority. Kenyon, Kathleen M. 1957. “Miscellaneous Objects in Metal, Bone, and Stone.” In The Objects from Samaria, edited by John W. Crowfoot, Grace M. Crowfoot, and Kathleen M. Kenyon, 439–77. Samaria-Sebaste 3. London: Palestine Exploration Fund. Magness, Jodi, 2014. “Arrowheads and Projectile Points.” In Gamla III: The Shmarya Gutmann Excavations, 1976– 1989; Finds and Studies, edited by Danny Syon, pt. 1, 21–33. IAA Reports 56. Jerusalem: Israel Antiquates Authority. Marsden, E. W. 1971. Greek and Roman Artillery: Technical Treatises. Oxford: University of Oxford Press. Peterson, Brian, and Scott Stripling. 2017. “Kh. elMaqatir: A Fortified Settlement of the Late Second Temple Period on the Benjamin Plateau.” In the Highland’s Depth 7:61*–91*. Raviv, Dvir, Scott Stripling, and Yoav Farhi. 2020. “A Hiding Complex from the two Jewish Revolts against the Romans at Khirbet el-Maqatir in the Eastern 311 The Excavations at Khirbet el-Maqatir: 1995–2001 and 2009–2016 Stiebel, Guy D. 2015. “Military Equipment from the Area of the Mausoleum and the Theater at Herodium.” In Herodium: Final Reports of the 1972–2010 Excavations Directed by Ehud Netzer. Vol. 1, Herod’s Tomb Precinct, by Roi Porat, Rachel Chachy, and Yakov Kalman, 432–53. Jerusalem: Israel Exploration Society. Sumner, Graham, 2009. Roman Military Dress. Gloucestershire: History Press. Van Driel Murray, Carol. 1987. “Roman Footwear: A Mirror of Fashion and Society.” In Recent Research in Archaeological Footwear, edited by D. E. FriendshipTaylor, J. M. Swann, and S. Thomas, 32–42. Technical Paper 8. Association of Archaeological Illustrators & Surveyors. Van Driel Murray, Carol. 1995. “Nailing Roman Shoes.” Archaeological Leather Group Newsletter 1 (February): 6–7. Van Driel Murray, Carol. 1999. “And Did Those Feet in Ancient Time . . . Feet and Shoes as a Material Projection of the Self.” In TRAC 98: Proceedings of the Eighth Annual Theoretical Roman Archaeology Conference, Leicester 1998, edited by Patricia Baker, Colin Forcey, Sophia Jundi, and Robert Witcher, 131–40. Oxford: Oxbow. Volken, Marquita, 2014. Archaeological Footwear: Development of Shoe Patterns and Styles from Prehistory till the 1600’s. Zwolle: SPA Uitgevers. Volken, Marquita, Olivier Paccalot, and Serge Volken, 2011. “Les clous de chaussures du site de Pfyngut: Les bases d’une typo-chronologie.” In Pfyn/Finges: Évolution d’un Terroir de la Plaine du Rhône; Le Site Archéologique de “Pfyngut” (Valais, Suisse), edited by Olivier Paccalot, 315–87. Cahiers D’Archéologie Romande de la Bibliothèque historique vaudoise 121, Archologia Vallesiana 4. Lausanne: Cahiers D’Archéologie Romande. Wood, Bryant. 2018. “The March to Jerusalem: Roman Brutality at Khirbet el-Maqatir.” Bible and Spade, Spring, 29–34. 312