A Hiding Complex and Ritual Bath (Miqveh) at Ḥorvat Maskana and the Question of Their Use in the Byzantine Period

ABSTRACT

Parts of a subterranean hiding complex and a miqveh (ritual bath) were excavated at Ḥorvat Maskana in 2020. The Byzantine pottery from the excavation, along with the dating of the plaster, suggests that the complex was used in the Byzantine period and may originally have been hewn in the Roman period. Additional hiding complexes with not only Roman but also Byzantine pottery exist elsewhere in Galilee. The presence of Byzantine pottery inside some of the hiding complexes points to the possibility that their use should not be limited to the Roman period, and that there may have been limited use in the Byzantine period by the local Jewish Galilean.

KEYWORDS:

• hiding complex
• Galilee
• miqveh
• ritual bath
• Roman
• Byzantine

Introduction

Currently, seventy-five ancient hiding complexes have been identified at fifty-nine sites in the Galilee (Figure 1), including a few that are uncertain (Shivtiel 2019). Hiding complexes are artificially hewn subterranean caves near or inside Jewish settlements. The hewn chambers are connected by narrow, low burrows that require crawling. In general, the burrows were added as difficult passageways between pre-existing hewn chambers and installations. The chambers and burrows create a complex that could have been used in times of danger. It is the burrows that are the main feature in identifying these as hiding complexes. Because of the physical difficulties and the harsh conditions in the hiding complexes and burrows, archaeological excavations have only been partially conducted in twenty-one of them. Archaeological excavations have been conducted at less than a third, for example, at Ḥorvat Ḥuqoq, ‘Enot Sho‘im, Ḥorvat ‘Amudim, I‘billin, and Ḥorvat Ruma, (Shivtiel 2016; Leibner, Shivtiel and Distelfeld 2015; Shivtiel and Osband 2019; Rochman 1985, 35; Muqari 1999; Leibner, Shivtiel and Distelfeld 2015, respectively). All of them have similar characteristics to the hiding complexes discovered many years ago in Judea. The hiding complexes were used by the local Jewish population at a time of danger and stress mainly in the Roman period. The archaeological evidence suggests that the hiding complexes were prepared in the Galilee for the Great Revolt, and possibly also for the Bar Kokhba Revolt (Shivtiel 2021), despite the fact that there is no concrete evidence as yet for the Galilee’s participation in the Bar Kokhba Revolt.

Figure 1. Map of Ḥorvat Maskana and sites with hiding complexes.

The site of Ḥorvat Maskana (henceforth, Maskana) is located in the Eastern Galilee and lies about 1.2 km northeast of the Golani Junction (Figure 1).¹ The Arabic name for the site preserves the ancient name ‘Mishkena’ (משכנה), which is mentioned in rabbinical literature as being located between Tiberias and Sepphoris (Y Berakhot 9, 5, 14d; Y Sanhedrin 3, 1, 21a). The site is described in the Survey of Western Palestine as having ‘Extensive ruins of much-worn stones, foundations of walls, and rock-cut cisterns. There are several sarcophagi and a large birkeh near; also some caves and wine-presses cut in the rock. Probably an ancient place.’ (Conder and Kitchener 1881, 403). Safrai (1985, 167–72) surveyed the site and identified structures including an inn. The site is located c. 300 m north of the Roman road linking Tiberias and Sepphoris and according to Safrai, contains remains of a local road that linked the site to the Roman road.

Leibner’s survey of the site focused on the Hellenistic, Roman and Byzantine periods. He estimated that the ancient village covered 3.5 ha (35 dunams) (Leibner 2009, 280–84) and found pottery that dated the settlement to the Early Roman–Early Islamic periods. Leibner notes the relatively abundant amount of Byzantine pottery (Leibner 2009, 284). He also suggests that the remains of architectural fragments from a monumental structure may have been from an ancient synagogue. Cisterns, agricultural installations, burial complexes and caves are found throughout the site and in the surrounding agricultural area. In 2011 (Berger 2013), two salvage excavations took place north of the ancient site beside the modern road. Remains of a modern field wall, a limekiln, a large stone quarry from the Roman and Byzantine periods and burial complexes from the late Second Temple period were uncovered as well as a round rock-hewn pit that might have been a limekiln and a nari quarry to the west of the road.

Subterranean Hiding Complexes at Maskana

Two hiding complexes have been identified at Maskana (marked by Shivtiel 2019, 201–05 as Maskana 1 and Maskana 2). These systems were surveyed in 2017 (IAA License No. S-769/2017). Maskana 2 is located on the eastern edge of the ancient settlement, whereas Maskana 1 is less than 100 metres to the north of the ancient settlement, apparently located outside of the village, in its agricultural hinterland. Additional caves are visible around Maskana 1. Above Maskana 1 are the collapsed remains of stones that seem to have been from a dwelling, but it is difficult to ascertain the period or the nature of the surface structure. The area surrounding Maskana 1, especially to the north and east, has been disturbed by modern olive groves (Figure 2). Our survey outside the cave (L104) collected pottery dating mainly from the Roman and Byzantine periods. However, since this area has been used for modern agriculture, it is difficult to know the extent to which these finds reflect its use in antiquity.

Figure 2. Plan of the Hiding Complex.

The Subterranean Complex of Maskana 1 (Figure 3)

The complex contains three main cavities. Cavity A is a large space that was hewn into a rocky conglomerate of the Ahuzam Formation composed of rocks and various kinds of pebbles. The cavity can be divided into two areas that were probably originally separated by part of the cave wall. The larger space measures c. 8 × 15 m. The dividing wall, which no longer exists (although traces of it are still visible), separated the main cavity from a small plastered area with stairs and a pool that has been identified as a miqveh. Cavity B (L201) is connected to Cavity A by a tunnel L200). Cavity C is connected to Cavity B by a short-hewn tunnel (L202). The tunnel (Figure 4), Cavity 2 (Figure 5) and Cavity C are identified as a hiding complex, similar in plan to hiding complexes discovered in the Galilee (Shivtiel 2019). Unusually for hiding complexes, the main passage was also quarried into the conglomerate, thus creating natural shelves and depressions that served as oil-lamp niches.

Figure 3. The area around Ḥorvat Maskana. (Photography M. Osband).

Figure 4. The hiding complex tunnel. (Photography Y. Shivtiel).

Figure 5. Hiding complex cavity B. (Photography Y. Shivtiel).

There are five openings to the complex today, four of which lead into Cavity A. The current southern entrance, which is the main way of entering the cave today and where there is a fig tree, is apparently not part of an original opening and was probably created by collapse or by robbers. The original entrance via the steps leading into the ritual bath has collapsed and no longer remains, but the upper steps are preserved. Two openings in Cavity A that may be part of a cistern are blocked by stones from above. The northern opening in Cavity A is a crudely hewn vertical shaft. Slightly to the southwest is an additional round vertical opening that is well-hewn. We did not find plaster remains in these openings. The large area with the two openings in the ceiling may have been a cistern, as the two openings on the northern ceiling suggest. However, no plaster has been positively identified here. A small opening on the northern side of the complex in Cavity C was also blocked by stones from above. There is no entrance to Cavity B from above.

Two areas were excavated over a three-day period: 1) A plastered water installation identified as a ritual bath in the southwestern part of Cavity A. 2) A hiding-complex tunnel, a small section of Cavity B and part of the entrance into Cavity C.

The Ritual Bath (Figures 6 and 7)

The southwestern part of Cavity A contains remains of a stepped water installation coated with hydraulic plaster that leads from the outside down into a rectangular pool. The original entrance is no longer preserved but was probably at the top of the steps, where there is a large tree. Prior to excavation, only a single step was visible from above. Some of the steps are missing and the north-western wall has collapsed or been removed. There were probably originally seven steps, of which parts of five are preserved (riser height 47 cm, tread depth 40 cm, tread width 87 cm). A pool at the bottom (L102) was cleared of fallen debris in the excavation (L100; fill depth 38 cm, underlain by a few cm of fill above the pool floor [L101]). The area surrounding the pool where the steps were apparently damaged was cleared of a fill (L103; depth c. 8 cm) down to a plastered surface. After removing the soil at the top, two steps hewn into the rock near the entrance were clearly visible, as well as a broken part of a third step. There were probably more steps leading into the pool, but they had either collapsed or were deliberately removed. Both steps and the walls were plastered. The pool at the bottom is rectangular (2.43 × 0.80 m, depth 0.50 m). The height of the cave from the floor to the current ceiling is 2.10 m. There is a small groove (10 × 18 cm, depth 15 cm) approximately in the centre of the bath’s western edge. The fill in the pool and surrounding area yielded a small amount of pottery (Figure 8), rock debris, bones² and many fragments of hydraulic plaster.

Figure 6. Plastered stairs of the miqveh. (Photography Y. Tor).

Figure 7. Plastered wall and pool at the bottom of the miqveh. (Photography Y. Shivtiel).

Figure 8. Selected pottery from Ḥorvat Maskana 1 hiding complex and miqveh³. (Drawn by Alexander Iermolin).

Table 1. Pottery Table for Figure 8.

Pottery Figure	Location	Parallel and Dating
1 Storage Jar	Miqveh, Locus 100, B.1000a	Loffreda 2008, 133–34, ANF 20; Early and Middle Byzantine
2 Storage Jar	Miqveh, Locus 100, B.1000b	Loffreda 2008, 133–34, ANF 20; Early and Middle Byzantine
3 Bottom of Cypriot Red Slip Bowl	Miqveh, Locus 100, B.1000e	Possibly form LRD (CRS) 2, Hayes 1972, 373–76, mid-5th– early 6th centuries ce
4 Casserole	Hiding Complex, Locus 201, B.2002a	Leibner 2018, 310, KH Casserole Form 3a/b (rim fragment too small to attribute to Form KH3a or KH3b); Roman period, likely 2nd–3rd century ce
5 Cooking Pot	Hiding Complex, Locus 201, B.2002b	Loffreda 2008, 191, Pent 15; Middle to Late Byzantine

The hydraulic plaster was intact beside the stairs and in the pool and back wall of the cave, and some traces remained on the sides. All of the plaster has a similar composition, and two layers were applied, possibly in two different periods. The inner layer (thickness 12–17 mm) is of a greyish colour and the outer layer (thickness 3–8 mm) is a light reddish colour with some charcoal remains. Much of the plaster on the eastern wall is preserved down into the bath. Plaster remains were also found on the cave’s northern wall, to the west of the bath, extending slightly along the southern wall of the bath. These remains suggest the original outline of the miqveh’s walls, which were once plastered and have either collapsed or were deliberately removed. The plaster remains suggest that the miqveh was originally in a separate room divided by cave walls from the large chamber in Cavity A.

The greyish plaster (Sample No. RTD 11121) was measured for radiocarbon dating by Elisabetta Boaretto of the Weizmann Institute. A date of 410–540 ce was determined with a 95.4% probability. No material for sampling was found in the outer reddish plaster. The analysis provides a clear Byzantine date for the miqveh’s inner plaster and is evidence of its use at that time. It may of course have been used both earlier and later than this date. The fact that the plaster traces are only found in the miqveh suggests that at this time, this area was not connected to the main part of Cavity A. It is not clear if the outer light reddish plaster is contemporary with the inner greyish plaster, or from a later date.

Discussion

The Use of the Hiding Complex and the Miqveh

The style of the hiding complex at Maskana 1 has parallels in the Roman period. The plaster in the miqveh dates from the Byzantine period. The pottery (Figure 8) from the Maskana 1 hiding complex was found deep in the passages and cavities and could not possibly have been swept down from the surface. We suggest two possible reconstructions for the use of Maskana 1: 1) The miqveh was built in the Byzantine period, as dated by the plaster, and the wall was later breached to provide access to the cavity containing a hiding complex that was also hewn in this period. 2) Alternatively, the hiding complex was hewn in the Roman period and later, in the Byzantine period, a miqveh was built in a separate cavity in close proximity. The wall of this miqveh was later breached to provide access to the hiding complex, which was reused in the Byzantine period.

The hiding complex at Maskana 1 reflects a well-known phenomenon, namely the breaching and incorporation of an existing miqveh into the complex, thus rendering them obsolete. However, in this instance there are two differences. One is the distance between the miqveh and the hiding complex. The other is the fact that the miqveh was evidently installed in the Byzantine period and was later breached toward the large cavity and the crawl way that predates the miqveh, despite the fact that this clearly rendered it useless for its primary purpose of purifying the body and the mind. In the Galilee, at Meroth (Furstenberg 2016, 50–56; Shivtiel 2019, 109) Ḥuqoq (Shivtiel 2019, 133), Ṣippori (Shivtiel 2019, 177); in the Judean foothills, at Ḥorvat Naqiq (Tepper and Shahar 1987, 169) Ḥorvat Rafia, (Kloner 1987, 251); in southern Samaria at Ḥorvat Abirjan (Raviv 2018, 16) Ḥorvat Jaba’it (Raviv 2018, 155) Ḥorvat Beit Radduf (Raviv 2018, 49) Ḥorvat es-Shuna (Raviv 2018, 230) and Beit ‘Iweis (Raviv 2018, 58). At these sites and elsewhere, in three separate regions, miqvaoth were breached in order to use them as part of a hiding complex. Such an act, in which the Jewish community abandons its means of ensuring ritual purity, is a vivid reminder that the principal purpose of the hiding complexes was to save lives. Such an important religious practice may only be forfeited in the face of imminent life-threatening danger, as evidenced by the nearly 850 miqvaoth found throughout Jewish settlements in the country (Adler 2011, 321–43). The phenomenon is certainly worthy of further research and discussion and requires special consideration.

The Question of post-Roman use of Galilean Hiding Complexes

Ḥorvat Maskana contains a hiding complex with pottery dating from the Roman and Byzantine periods (and some from the Mamluk period). The Byzantine pottery from the excavation, along with the dating of the plaster, suggests that the complex was used in the Byzantine period. The Byzantine pottery found deep within the hiding complex at Maskana 1 shows that it was likely used for some emergency in this period. Such proof is important because in some cases, pottery washed down from the surface has contaminated the archaeological dating evidence.

Maskana 1 is not the only hiding complex in Upper and Lower Eastern Galilee that has yielded Byzantine finds. They have also been recovered from deep inside hiding complexes at other Jewish settlements. The most prominent sites that have yielded Byzantine pottery from deep inside tunnels are Meroth (Ilan and Damati 1987, 91–92) Huqoq Synagogue Complex (Shivtiel, 2016, 200), I‘billin (Muqari 1999, 19), Ruma (Rochman 1985, 32), Ṣippori (Strange and Longstaff 1987), Migdal Ha‘emeq (Shalem 1996) and Kafr Kanna (Aviam 2004, 125). These complexes may be interpreted as reflecting one or more events in the Byzantine period that forced Jews to reuse existing hiding complexes or prepare new ones.

The Jewish settlement in the Galilee, in the Roman period, continued into the Byzantine period despite the catastrophic event of the Great Revolt. This is in stark contrast with the pattern of Jewish settlements in Judea, which ceased almost entirely in the second century ce, following the Bar Kochba Revolt (Ben David and Raviv 2021). If new settlements were later established in Judea in the Byzantine period, we have no evidence of the reuse there of first–second-century ce hiding complexes. The hiding complexes in the Galilee were located at Jewish settlements that continued from the Roman into the Byzantine period. Byzantine pottery in some of the complexes suggests their use at that time by Jews. Currently, we cannot associate them with a particular event that affected Jewish life in the Byzantine period and that prompted the preparation of the complexes, for lack of historical sources attesting to a Jewish rebellion or military conflict after the Gallus Revolt (For a discussion of hiding complexes in the Byzantine period, see also Shivtiel, 2016, 200–01).

Based on all its features, the Maskana 1 complex was intended as a hiding complex from the outset. It has a main, winding passage that was evidently hewn specifically as a crawl way, and another side passage, only the entrance to which was excavated. The main passage leads to three rock-hewn cavities (see plan) where refuge could be taken. The complex yielded various pottery vessels attesting to habitation in the cavities accessed by the passage. As stated above, a similar hiding complex was found in the southern part of the settlement (Maskana 2) that also contains all the identifying features of a hiding complex (Shivtiel 2019, 203–05). One of the definitive features of hiding complexes, in contrast with underground storerooms, is their access via a narrow, very low twisting passage that was deliberately planned as a crawl way. This eliminates the possibility that it was used for daily habitation or as an underground storeroom, as might be suggested by the size of the natural cave in which the hiding complex was made, since the deliberate formation of a crawl way attests to its intended use. It is clear that both in antiquity and in subsequent periods, underground cavities were adapted for use as dwellings, storerooms and installations that were later transformed into hiding complexes, as at ‘Iyei Ma‘arot (Shivtiel and Stepansky, 2013, 249–53), Ṣippori (Strange and Longstaff 1985, 298), Usha (Shivtiel 2019, 208) and elsewhere. At these sites, there was originally easy access to the underground storeroom or installation, either upright or slightly bent over, and it was only later that a crawl way was hewn.

The discovery of hiding complexes containing impenetrable passages and dated to the Byzantine period reinforces the conclusion that the complexes were used as defensive refuges in times of distress, rather than for launching offensive actions, as some have proposed (Dar 2015; Safrai and Ortner 2021, 229–33). The preparation of hiding complexes is similar to that of emergency shelters and the complexes were obviously used as such in the Byzantine period, in response to some period of unrest in the country. The Byzantine ceramics found in the complex did not provide a precise dating within this period, making it difficult to attribute its use to a single event at this time. Hopefully, future excavations and studies will add more sites and evidence in order to address this issue.

Summary

The results from the excavation at Maskana 1 suggest that this hiding complex was used at some point in the Byzantine period by the Jewish residents of Maskana. This interpretation is supported by the fact that unlike their Judean counterparts, many of the hiding complexes of Galilee are found at sites where there was continuous Jewish settlement from the Roman into the Byzantine period. Since hiding complexes would only have been used in times of danger, they may have been used or prepared for use in the Byzantine period in response to some pressing need. We suggest that the possibility of post-Roman period use of Galilean hiding complexes should be considered, and that the Byzantine pottery recovered from some of these complexes provides evidence of their use at this time. The reasons for the distress that motivated the Jews in this part of Galilee in general and the Jews of Maskana in particular to use the hiding complexes are still unknown and await further examination and research.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 This report deals with the northern complex (Ḥorvat Maskana 1), which was excavated over a three-day period by M. Osband and Y. Shivtiel in 2020 (May 2–5, 2020), Israel Antiquities Authority License G-9/2020. We are grateful to Yitzhak Tor of the Israel Antiquities Authority for drawing our attention to the site of Maskana and its underground cavities, and for participating in the excavation of the hiding complex. We thank Aaron Greener for his help in surveying the site. With thanks to the students and volunteers from Zefat Academic College and also to the volunteer Shmuel Grassiani for their assistance in excavating the hiding complex.

2 We thank Dr Ram Bouchnik for his study of the faunal remains. The analysis of tens of animal skeleton remains from contexts reflecting ancient use (and not from open areas or material that was washed into the complex) revealed a large proportion of livestock bones, mainly caprine (sheep and goats) and a smaller proportion of cattle and chicken remains. Cat remains were also discovered. The mortality profile of the herd shows a high rate of mature and a minority that are young. In addition, foetal remains were identified. The remains correspond to the animals’ availability in the rural economic hinterland, and the limited resources of the cave inhabitants who were satisfied mainly with mature livestock that had left the circle of agricultural benefit in the surrounding villages (Bouchnick 2015; Bouchnick 2021).

3 Inside the hiding complex around 70 body sherds were recovered (mainly from 3–8 cm in length) of storage and cooking vessels from the Roman and Byzantine periods. The majority of the body sherds came from the excavation of L201. In addition to Figure 8 the pottery from the hiding complex included cooking pot Kefar Hananya form 4A (Leibner 2018: 308, Early Roman period), and Late Roman/Byzantine period pottery, cooking bowl Kefar Hananya form 1E (Leibner: 2018: 314), a casserole with horizontal handles (Leibner 2018: Byzantine Open Cooking Pot C3a, 315), and two basins (Loffreda 2008: Piat 61 and Piat 63, 241–243). Similar quantities of body sherds were also found in the Miqveh and in addition to the pottery in Figure 8 a storage jar of Late Roman/Byzantine period date. Pottery outside of the cave was mostly of Late Roman and Byzantine period date including three Kefar Hananya form 1E cooking bowls, two casseroles with horizontal handles, four African Red Slip body sherds and two storage jars.