1The Negev region of southern Israel enjoyed a florescence of urbanization in the Roman and Byzantine periods (1^st-7^th century CE)1. The region was controlled by the Nabataeans emanating from Petra between the 4^th century BCE and the Roman annexation of Nabataea in 106 CE. Initially, the Negev served as a buffer between Petra and the coastal region, and important trade routes ran through the region, particularly the «Incense Road» that led to the Nabataean’s main port at Gaza. The Nabataeans transported valuable incense resins, frankincense and myrrh, and other rare aromatics and spices along these roads. From the end of the 1^st century BCE, they began to produce perfumed oils that they packaged in ceramic unguentaria produced in Petra, a product that apparently brought even greater revenue to the small kingdom2. The production of perfumed oils did not diminish with the Roman annexation in 106 CE, and the newly founded Provincia Arabia, with Petra and other cities in the Roman East, entered a new phase of urbanization and trade and an elevated standard of living that continued until the 3^rd century CE.

2The 3^rd century CE was a period of crisis throughout the Roman East and it marked the cessation of international trade along the Incense Road. The re-deployment of Roman legions in the region under Diocletian, such as the Legio X Fretensis, which was transferred from Jerusalem to Aila (the modern Red Sea port of Aqaba), marked a new phase of architectural and economic activity. From the 4^th century until the early 7^th century, agricultural produce, and particularly wine production for export, provided the economic backbone of communities in what was to become Palestina Tertia under Byzantine rule.

Palatial private houses with stables in Mampsis

3The site of Mampsis was built by the Nabataeans in the 1^st century CE as a major way-station at the junction of two important roads: the road leading up from the southern end of the Dead Sea and the road leading up from the copper mines of Faynan. Following the Roman annexation of Nabataea in 106 CE, much of the original village was razed in order to make way for a number of palatial private houses, the investment of which evinces the high standard of living in the region in the post‑annexation period.

4The largest private building of this type was Building XII, a structure that contained ca. 1500 m. of living space on at least two stories (fig. 1). In the south-western corner of Building XII, the excavator, A. Negev, uncovered a large stable (Locus 431), 11.35 x 11.35 m. in size, with two rows of troughs (fig. 2 and 3)3. The troughs were originally covered with vaults that consisted of three courses of ashlar masonry with evidence of holes into which wooden or metal bars could be inserted in order to confine animals, probably horses4. He also found evidence of smaller holes that would have been used to fasten a metal or wooden grille covering the front of the vaulted windows over the troughs. The northern aisle between the troughs and the wall was 2.45 m. wide while that of the southern aisle measured 2.3 m. in width. The southern aisle was accessed through an arched doorway, 0.75 m. in width. The courtyard-like space between the northern and southern rows of troughs, described as a «nave», was 6 m. in width. According to A. Negev, the stable was originally covered by a roof which could be accessed by way of a staircase tower (Locus 414) and suspended balconies between the roofs.

Fig. 1. — Plan of Mampsis, Israel

Fig. 2. — Plan of the stable in Building XII, Mampsis, Israel

Fig. 3. — Stable in Building XII, Mampsis, Israel

5A second, smaller stable (fig. 4) was discovered inside the south-east corner of Building XI (Loci 327, 329 and 330)5. Constructed in the Roman period as a private villa with a stable, in the Byzantine period a church was built inside the west side of the building and the stable went out of use. This stable was found in a particularly good state of preservation due to the fact that the vaulted windows over the troughs were blocked up following structural damage resulting from the 363 earthquake. Each row contained four troughs. The central space between the rows measures 4.75 x 7.5 m. while the doors leading into the aisles in back of the rows measured 0.8 m. in width. Each trough was constructed of two or three stones and the sides of the troughs contained holes for grilles and bolts similar to the stable described above. The aisles behind the troughs measured 2 x 7.25 m. in size. Perforations in the corners of some of the piers were apparently used for tethering animals6.

Fig. 4. — Stable in Building XI, Mampsis, Israel

6The post-annexation date of the Building XII was confirmed by the discovery of a hoard of 10,800 coins, the great majority of which date no later than 222 CE7 and in later excavations conducted by the writer along the eastern side of the building8. Both stables shared the same construction techniques and they both were accessed from the main, ground-floor courtyard. Both were positioned in such a way as to take advantage of north-westerly winds that would blow unpleasant odors away from the rest of the household.

A stable in the khan at Rehovot-in-the-Negev

7A structure described as a khan or caravanserai (Area C) was excavated between 1976 and 1979 by a team led by Y. Tsafrir on behalf of the Hebrew University in Jerusalem9 (fig. 5) at Rehovot. The khan measures 30 x 30 m. and the arch-covered stable is located in its northwest wing10. The stable contains two rows of three troughs, only one trough of which was fully cleared. The stable was accessed by way of a doorway, one meter in width that led into an open space, 3 x 7.3 m. in size, situated between the rows of troughs. The spaces behind the toughs measure 1 x 2.1 m. in size. Holes for bars were found over the troughs. Similar to the stables in Mampsis, finds discovered from under the floor in another part of the khan points to its existence in the period after the Roman annexation in 106 CE.

Fig. 5. — Plan of the stable in the khan at Rehovot‑in‑the‑Negev, Israel

Stables in Byzantine dwellings at Oboda and Shivta

8Stables dated to the Byzantine period have been found in private dwellings in Oboda (Avdat) and Sobota (Shivta). The stable in a partially (unpublished) dwelling in the «Roman Quarter» in Oboda (now known as the Byzantine Quarter) was noted in passing by A. Negev along with a second surveyed on the northern end of western slope of the site11. Two mangers are present in first structure, one of which was blocked by later building, presumably as a result of structural damage caused by an early 5^th century CE earthquake at the site12. At Shivta (fig. 6), A. Segal excavated a stable that originally had five mangers in the south-western corner of a typical dwelling of the Byzantine period, dated «no earlier than the 4^th century CE13» (fig. 7-8). The stable contains five mangers and at least two large hewn vessels present on the floor nearby. The aisles on either side of the mangers measure 2.80 m. wide (west side) and 3.78 m. wide on the east. Two original doorways in the room measure 0.95 m. and 0.91 in width. The stable has some similarities to the Roman palatial houses in Mampsis: it is located on the ground floor of the building off the main courtyard and in a corner that would have benefited from winds from the north-west. The piers between the mangers have large holes for placing wooden or metal bars. In addition to the mangers, a number of large, stone-hewn vessels that may have been used livestock were found nearby. A smaller, Byzantine dwelling on the opposite side of Shivta, has an arched manger in its small courtyard. This structure was partially cleared by members of the Colt Expedition in the 1930s but was never properly excavated. Like the mangers in the private house in Oboda, its presence in a very small dwelling is interesting to note.

Fig. 6. — Plan of Shivta, Negev, Israel

Fig. 7. — Plan of the large stable at Shivta, Negev, Israel

Fig. 8. — Large stable at Shivta, Negev, Israel, 2015

Stables associated with monasteries at Oboda

9Oboda was the site of at least three monastic communities located around the South Church, in the structure adjacent to the Saints’ Cave and in a cave and structure located on the south side of the western slope excavated in 201214 (fig. 9). The interesting feature about the stables associated with these monasteries is the manner in which they utilized space of earlier structures.

Fig. 9. — Map of Oboda, Negev, Israel

10One monastery was situated around the South Church at Oboda, called the Church of St. Theodore, a church that contained a reliquary that apparently drew pilgrims15. The church was built into a pre-existing space of the old Nabataean temenos on the acropolis at Oboda and it incorporated stones and also walls of earlier structures. The stable was not recognized by either the excavator, A. Negev, or earlier explorers such as C. L. Woolley, the latter or which described «curtain walls made up of vaulted cells» and a small irregular courtyard that he assumed was the site of an oil press16. The confusion seems to stem from the fact that the troughs were constructed in a pre-existing room of the acropolis, the walls of which were severely damaged in a local earthquake sometime in the early 5^th century CE17. The original room measures ca. 8 x 9 m. with two rows of troughs, each row broken with an entrance (fig. 10). Altogether, eight troughs are present in this area, which incorporates up to four rooms or spaces. The main entrance leading into the facility is located along the northern wall and it measures 0.80 m. in width. The entrances leading off into rooms to the east and west of the rows of troughs all measure 0.77 m. in width, as does the entrance into a space between the two rows located directly opposite the main entrance. The troughs themselves vary in size but average one square meter and 0.28 m. in depth. This stable may have housed donkeys either used by pilgrims visiting the St. Theodore Church and reliquary or more likely those used to haul up grapes to be pressed in the Byzantine winepress located outside the church18.

Fig. 10. — Acropolis stable facing north-east, Oboda, Negev, Israel, 2015

11The Byzantine structure built in front of the «Saints’ Cave» excavated by A. Negev appears to have belonged to a second monastic community19. Support for its identification as a monastery may be found in the presence of many crosses and red-painted dipinti on the walls of the cave, one of which appears to be a saint soldier, possibly St. Theodore20. Moreover, the excavator uncovered an unusual urinal (Room 11) where urine was apparently collected and possibly sued for whitening clothes by a fuller21. Similar to the South Church monastery, a winepress present below the structure on the plain north of the site was probably used by monks to press grapes and the wine was fermented in the caves, and particularly Room 16 in back of the monastery where a hewn bench with depressions for jars is present22. The small stable that was used by the community was found in a cave on the tier above the main structure. The entrance faces northwards and the cave, which has not been excavated and is in a collapsed condition, contains the remains of at least three or four arched troughs hewn from the chalky bedrock (fig. 11). A cross is carved above the troughs and traces of red dipinti decoration are found throughout the cave and on the exterior walls.

Fig. 11. — Monastery cave stable on the North Slope facing south, Oboda, Negev, Israel, 2015

12Evidence of the use of donkeys and camels as pack-animals to transport grapes at the site was discovered on an inked-inscription present on a camel bone in another part of the site dating to the Early Byzantine period (4^th century CE)23. The presence of these animals could justify the presence of the troughs.

Stables in the Negev in other contexts

13A stable with troughs (as yet unpublished) was uncovered in the Roman/Byzantine village of Mezad Yeruham, a site originally excavated by R. Cohen24. This stable was discovered adjacent to a tower of the post-annexation (Late Roman) or Byzantine period. This small stable contains a row of three troughs. A doorway, 0.74 m. in width, provides access to a space behind the troughs measuring 2.30 m. from the back wall. The doorway leading into the main room of the stable measures 0.98 m. in width.

14Evidence of stables (in a deteriorated state) has been detected by the writer in the Early Byzantine cavalry camp at Mezad Hazeva25 and the so‑called «Governor’s House» at Shivta of the Byzantine period. The cavalry camp was severely damaged by the 363 earthquake and the remains of arches over mangers were built in two rooms of the camp, including the former gate area, in the second phase of use. This later phase appears to have continued as late as the mid-6^th century during the period that mounted patrols passed between Mezad Hazeva and Mampsis by way of the strategic Scorpions Pass and a number of small forts guarding the road. At Shivta, a public building known as the «Governor’s House» shows signs of occupation in the Early Islamic period, when a mud-brick structure was built into the north wing over the traces of the troughs of a Byzantine stable.

15An underground stable is located in close-proximity to the Incense Road, between the ancient Nabataean capital of Elusa (Halutza) and Oboda in the modern settlement of Retamim. The stable and its one row of three troughs were hewn into the bedrock similar to those found at Oboda described above (fig. 12). However, no signs of carved crosses or red dipinti were found that could possibly date the construction or the reuse of the stable to the Byzantine period. The stable is located only a few meters from an early Nabataean cistern of a type found in the Negev Highlands near the Incense Road and also secondary roads used by the Nabataeans in the Hellenistic period26. Both the underground stable and the cistern remain unexcavated. Future investigations may reveal the date of the construction and use of the stable. In the meantime, underground rooms of the Byzantine period that were either carved into bedrock or into loess continue to be discovered throughout the region of the Western Negev and Be’er Sheva basin. This region was heavily populated in that period. Some of the underground rooms appear to have been used for fermenting wine, such as one example at Be’er Shema, located northwest of Elusa27, and possibly a second example also found northwest of Retamim at Tell et‑Tuwaiyil28.

Fig. 12. — Cave stable, Retamim, 2015

16The appearance of stables with troughs and the presence of holes for bars and grills found in Roman and Byzantine contexts in the sites of Mampsis, Rehovot-in-the-Negev and Shivta point to their use for expensive animals such as horses or camels. This seems to be part of a regional trend of constructing stables starting in the Roman period, particularly within the confines of private homes and other buildings, which A. Negev pointed out extended to the Hauran and Northern Syria29. The stables present in Mampsis in large, private villas built in the 2^nd century CE are additional evidence of the high living standards of the inhabitants in an era of international trade that came to an end with the crises of the 3^rd century CE. In the 5^th and first half of the 6^th century CE, settlements in the Negev experienced a renewed degree of prosperity due primarily to the widespread production of wine that was exported abroad30. In this period, towns such as Shivta, Rehovot-in-the-Negev and Nessana, as well as the local capital city at Elusa, expanded in a dramatic manner. The smaller stables and troughs that were found in Byzantine contexts at Shivta and Oboda lack holes for bars or grills, something that suggests that these stables were used for smaller, less expensive animals such as donkeys that were utilized in agricultural work and particularly the role of transporting grapes to winepresses31.