Stratum IV (Roman period)
Remains of an industrial installation were exposed in Area B (L201; Fig. 2). A plastered circular basin (diam. 2 m, depth 1.5 m) whose walls were lined with small and large basalt stones (max. diam. 0.6 m) was excavated. Three layers of plaster (each 2 cm thick) were meticulously applied to the inside of the basin and a plastered depression (diam. 0.3 m, depth 0.3 m) was cut in its floor. Fragments of pottery and white tesserae were recovered from the fill in the basin and on the surface. The tesserae probably belonged to and came from parts of the installation that remained beyond the boundaries of the excavation. The capacity of the basin (in excess of 3 cu m) implies it was used to produce a high out-put product, but the current data is insufficient to determine its purpose.
The pottery in the fill in and around the pit included a Kefar Hananya Type B1bowl (Fig. 3:1) and a jar (Fig. 3:2) from the Roman period (second–fourth centuries CE). Since no finds postdating the Roman period were discovered, it is very likely that the installation was used during this period.
 
Stratum III (Byzantine period)
Architectural remains from this stratum were discovered in Area A, Sq A1, in a probe beneath the floor levels of the Stratum II building (Figs. 4, 5: Section 1-1). The remains included a wall (W132; min. length 3 m; width 0.6 m) built of different size fieldstones that incorporated a fragment of a doorjamb or threshold stone. The wall extended beyond the boundaries of the excavation.
Due to the limits of the excavation area, it was not possible to date the Stratum III remains; however, as pottery fragments from the Byzantine period were found in the accumulations beneath the Stratum II remains, it is assumed that Stratum III should be ascribed to this period. Potsherds from the Byzantine period were mostly retrieved from surface layers above the building and included imported Late Roman C and African Red Slip bowls (Fig. 3:3–5), a casserole (Fig. 3:6), a krater (Fig. 3:8) and a lamp (Fig. 3:11). In the levels of collapse and accumulation layers above a Stratum II floor level (L136), a cooking pot (Fig. 3:7), jars (Fig. 3:9, 10) and a lamp (Fig. 3:12) were found.
 
Stratum II (Early Islamic period)
Most of the remains in Area A belong to this stratum. The tops of walls (width 0.6–0.7 m; min. preserved height 0.9 m) were exposed at a depth of 0.3 m below the surface. Some of the walls were founded above the Stratum III remains; they were usually built of two rows of large basalt fieldstones with a core of small stones. The architectural remains are divided into a complex of rooms in the east (Building A, Sqs 1, 2) and two paved spaces in the west (Building B, Sq 4). Walls (W111, W130) were built next to each other at the interface between the two structures, and were probably used simultaneously (Fig. 6). It is assumed that Building B was originally used as an olive press, and Building A, which contained storerooms, stood alongside it. The structures served as a dwelling for a long period after they were no longer used in their original capacity. Given the ceramic finds, it is presumed that the buildings were constructed in the Early Islamic period and used for an extended period of time, at least until the Mamluk period.
Building A. All that remained of the eastern building was the western wall (W111), the eastern wall (W121), the wall adjacent to it (W109), and several interior walls (W110, W114, W122, W140). The building, founded on the remains of Stratum III, was sealed by floor remains and a habitation level of Stratum I. The walls were preserved to a maximum of eight courses high (W111). Two sections of stone pavement were exposed: a floor (L126; not in the plan) that abutted Walls 110 and 114, and Floor 136, above which was an accumulation (L115) between Walls 111 and 122.
The building consisted of three rectangular rooms that were no wider than one meter; they might have served as storerooms in a larger building. Even if these are constructive spaces of an upper building, they could still be conveniently used for storage. Parts of other units, probably larger and wider, were exposed south and east of them. A wall (W109) n the easternmost unit was constructed differently than the rest of the walls in the building. The wall was built of a row of flat stones to which roughly hewn stones were attached, placed on their narrow side (Fig. 7); this might be part of an installation, or a shelf of sorts that was used for storage.
The ceramic artifacts from the building’s floor levels included a krater (Fig. 3:13) and a jar (Fig. 3:14), both found on the floor of the eastern room (L116) and two other jars (Fig. 3:15, 16) that were discovered on Floor 136 in the adjacent rectangular room; all of them date to the Early Islamic period.
Building B. The eastern (W130) and western (W129) walls and a paved hall between them (L134) were preserved from the building. Another paved surface (L138) that might have served as a courtyard was exposed west of W129. The area in the north of the excavation was disturbed. The floor of Hall 134 was entirely paved (length 6 m, min. width 3 m) and a pit (L137) was located in its center. Pillars (W112, W123) built on the foundation level of the stone pavement adjoined Walls 129 and 130. A jar (Fig. 3:17) dating to the Early Islamic period was found on the floor level, next Pillar 112. A pillar or another column was located in a straight line between these pillars, but at a different angle; therefore, it is unlikely it served as a base for a double arch. This pillar may have been part of an earlier stratum or part of the building’s foundation or installations. A stone used as a press bed (Fig. 8) was shifted slightly when trial trenches were dug prior to the excavation; in all likelihood, it was originally located between the pillars so that it could drain the oil into Pit 137 in the center of Hall 134. The press is identified as a direct pressure screw type. Stone Pavement 138, in the western courtyard, was only partially excavated. In all probability, it was originally part of the olive press that operated there and the adjacent service rooms. Given the multitude of jars on the floor levels below later collapse, it is assumed that they were used in the original habitation level of the building, and it might constitute further evidence regarding the intended function of the building for the production and storage of oil.
The ceramic finds from Stratum II are not homogenous. Along with the pottery from the Early Islamic period, to which the stratum is dated, Crusader and Mamluk pottery was found in the later fill and collapse on the floors of the buildings. A fragment of a glazed jar dating to the twelfth–early thirteenth centuries CE, which originated in Beirut (Fig. 9:1), was found in a mixed level above the floor of the olive press in Building B. The pottery found beneath the floor of Stratum I in Building A included a bowl (Fig. 9:2) and a jug (Fig. 9:3) that were locally produced and painted and two Crusader cooking pots (twelfth–early thirteenth centuries CE) that come from Beirut, one of them has traces of brown glaze (Fig. 9:4, 5). In the fill sealed beneath a Stratum I floor level (L108; Fig. 5: Section 2-2) glazed sherds dating to the twelfth century CE were found; based on these finds, an identical date is also suggested for the group of locally produced red-painted vessels. These are usually difficult to accurately date because they were in use in the twelfth century CE and later; therefore, it is customary to date them more generally to the Crusader–Mamluk periods (twelfth–fifteenth centuries CE).  
                   
Stratum I (Mamluk period)
The meager building remains from this stratum were concentrated in Sq A2 and included a section of a wall (W124; Figs. 5: Section 2-2; 10, 11) and remains of a stone pavement (L108). Wall 124, exposed right below the surface, survived a single course high and was built of large basalt stones. Stone Pavement 108 consisted of large stones (length 0.3–0.4 m) and was founded above Walls 110, 111 and 130 of Stratum II. The stones of W130 were bonded with those of Pavement 108.
Many fragments of pottery vessels from the Mamluk period, which probably postdated the thirteenth century CE, were found on the surface and down to Floor 134 in Building B and Floor 108 in Building A. The finds included two green-glazed bowls (Fig. 9:6, 7), a Cypriot cooking bowl with brown glaze (Fig. 9:8), two bowls with yellow on brown glaze (Fig. 9:9, 10) and a base of a frit bowl (Fig. 9:11), as well as red painted handmade bowls (Fig. 9:12–14); cooking pots (Fig. 12:1, 2) and handmade cooking pots (Fig. 12:3); on the rim of one was an incised decoration (Fig. 12:4). Other handmade vessels included basins (Fig. 12:5–7), one of which has imprints of a mat on its base (Fig. 12:6) and a basin with a colored stripe on its inside (Fig 12:7). Other finds included a jar handle decorated with painted stripes (Fig. 12:8), a red-painted handmade jug/jar (Fig. 12:9), a jug with a circular impression stamped on its body (Fig. 12:10) and a cup of indigenous clay (Fig. 12:11).
 
Additional Artifacts
Three coins were found. Two were discovered in Area A, one of which was identified—a coin from a disturbed level in the north of the excavation, which was dated to the years 364–375 CE (IAA 106080). A single coin (IAA 106081) was found on the surface in Area B and dates to the years 1382–1384 CE—the first reign of Sultan Barquq. The metal artifacts included several iron nails dating to the Crusader period that were found in the collapse of Building A. The glass finds are meager and date to the Late Roman–Byzantine periods, no later than the seventh century CE. 
 
The exposed remains have cardinal implications for reconstructing the history of the site and the settlement model in its surroundings. The finds in Area B include remains of an industrial installation and ceramics from the Roman period. Other installations presumably existed on the eastern side of the channel, close to the agricultural area. It seems that a special find—a bronze incense shovel—discovered not far from the fortress should be dated to the Roman period (below).
Three architectural phases were identified in Area A, alongside ceramic finds from four periods. Having analyzed the architectural complex it is proposed that the main phase belongs to an olive-press compound. This structure was only partially excavated; a room where the press installation was located and an adjacent courtyard (Building B) were exposed. It seems that the western building (A) was used for storerooms, possibly for keeping the product from the olive press.
The finds dating to the Byzantine period were found scattered in the fill and later layers of soil and therefore it seems they neither indicate a construction phase of the building nor its original purpose. It is possible they derived from an adjacent building or an earlier level. In excavations that were conducted near the spring (Permit No. A-4622), not far from the current excavation, pools dating to the Byzantine period were exposed. This is further evidence of the extent of the settlement and its economy in this period. Given the ceramic finds on the floor levels and in the floor beddings of the olive press and the adjacent rooms, it is suggested that the building was constructed in the Early Islamic period. The olive press shows that the residents of the site supported themselves by growing olives and producing oil. The many jars revealed on the floors are likely to corroborate this conclusion. Weights from another olive press installation that operated at Taiyiba were documented in a survey (HA-ESI 124), but do not belong to the type of installation that was exposed in the excavation.
The pottery fragments below the Stratum I floor level were dated to the Crusader period (twelfth century CE). The paucity of these potsherds and their appearance in fill above the Stratum II floor levels, alongside potsherds of the Early Islamic period, underline our assessment that the olive press building and the buildings next to it, perhaps due to their construction, were utilized in secondary use during this period and afterward.
Most of the ceramic finds dated to the Mamluk period (twelfth–fourteenth centuries CE). Potsherds from this period were found in each excavation section from the surface level, on the Stratum I floor and beneath it to the Stratum II floor levels. This is the last phase of the building’s use after it had already ceased to operate as an olive press.
 
Buildings and extensive settlement remains were exposed in the excavation north of the fortress that is identified as a structure from the Crusader and Mamluk periods. Assuming this identification is correct, it seems that the excavation finds indicate the settlement in this part of the site continued uninterrupted throughout the Crusader and Mamluk periods and its inhabited area extended to at least the region of the excavation, north of the modern settlement.
The excavation yielded no remains that dated to the Ottoman period, like those documented in the vicinity of the fortress (HA-ESI 123). Aerial photographs of the site from the mid-twentieth century also show that the settlement from the Ottoman period at Et-Taiyiba was smaller than the surrounding areas with antiquities and was mostly concentrated near the fortress (HA-ESI 124).
 
 
An Incense Shovel from Taiyiba
David Amit 
 
When mechanical equipment, working on behalf of the Southern Jordan Drainage Authority, was preparing the flow channel in the stream near Horbat Haddad in the summer of 2005, Mashor Zou‘abi, an inspector with the Antiquities Authority, discovered a decorated bronze incense shovel (Figs. 13, 14). The incense shovel was found without any context to a nearby building or tomb. Tzach Horovitz, the Lower Galilee and Valleys district archaeologist, turned the incense shovel over to me for publication.
This incense shovel is similar in detail to four incense shovels (Nos. 3–6) that were found in the Cave of Letters in Nahal Hever (Y. Yadin 1963. The Finds from the Bar-Kokhba Period in the Cave of Letters. Jerusalem. Pp. 48–60). The incense shovel from Taiyiba is not identical to any of the aforementioned shovels, as they differ from each other in their size, decorations and other components; nevertheless, it is evident they belong to a common type.
 
The artifact consists of a shovel or pan-like body (and will be referred to as such below) and an elongated tubular handle, with a protrusion attached to the tube at its end, which was broken. This protrusion, based on parallels from the Cave of Letters and other sites in the Roman world, included a leaf-like decoration on its upper part and a support in the form of an animal’s paw below it.
The Pan is trapezoidal and very close to being rectangular; in that respect the incense shovel from Taiyiba is unique compared to those from the Cave of Letters, which have a rectangular shape. The pan (length 14.5 cm, inner width 10 cm, outer width 11 cm) is closed on three sides by a tall rim and only the outer fourth side is open. The rim (height 1 cm) is slightly everted and has horizontal margins that form a ledge of sorts (width 1 cm), adorned with two grooves that extend around the three sides. The pan’s surface is decorated with nine concentric circles that were engraved in three rows that are almost symmetric, both widthwise and lengthwise. Each circle (diam. 1.25 cm) has six smaller circles and a deeper hole is engraved in its center. The coals, which the nine concentric circles apparently symbolize, were placed on the surface of the pan.
Animal ears of sorts (length 1 cm) protrude outward from the two corners nearest the handle. These were brazed to the bottom part on the underside of the pan and the upper part (the ear itself) is concave with a conical depression drilled in it (upper diam. 5 mm, depth c. 2 mm). According to Incense Shovel No. 3 from the Cave of Letters, which survived intact, we know that the stem of a cast goblet was soldered in each of the two conical depressions. The stem of the goblet bore the body of the goblet, a kind of tiny vessel with vertical sides and a flat bottom. Different kinds of scents were placed in these goblets, which were sprinkled from time to time over the coals, thereby creating the fragrant incense smoke.
Four rounded feet in the shape of short nails (length c. 4 mm) were affixed to the bottom part of the pan, near the corners.
The underside of the pan was reinforced on the exterior by two thickened strips (width 0.5–0.6 cm, height c. 1 mm), one running lengthwise in the middle, and the other widthwise, next to its open edge. Gray clay (?) filler, only a small amount of which has survived, was smeared on the step near the handle at the bottom of the pan, between the two ears, as well as beneath the ears themselves (see below).
The Handle. The cylindrical handle is a bronze tube (length 10.5 cm, without the decorative protrusion; diam. 2.4 cm), hammered from a thin strip of metal which has been halved down its length leaving a slot (width c. 1 cm) along the bottom. To strengthen the contact between the pan and the handle, a kind of clasp, recessed in the gray fill and not visible, was set at the point where the two join on the bottom. The upper exposed part of the clasp is split in two like a V and is soldered to the bottom the pan. 
The tube is decorated with three rings (width 1 mm, thickness 1 mm), one near the pan, the second ring on the third closest to the end of the tube and the third ring at the end of the tube, close to a trapezoidal plate (2.3 × 3.0–3.2 cm), to which the ornate protrusion and the foot rest, were brazed, and as previously mentioned, had broken off.
The inside of the tube, and the step at the bottom of the pan next to it, were filled with gray clay (?) that served as insulation, probably to allow a person to hold the handle and use the incense shovel. The length of the handle determined how much effort the user needed to exert (on the shovels from the Cave of Letters the handle lengths range from 11.4 cm to 18. 5 cm). Because of this, we find in the rabbinic literature that to facilitate the work of the high priest on Yom Kippur, a time when he had much work and was fasting, the handle was made longer that day: “on every day the handle (=of the incense shovel) is short and today (i.e., Yom Kippur) it is long” (Yerusalmi Yoma 4, 4 [41, pp. 3–4]).
 
In his report on the incense shovels from the Cave of Letters, Yadin presented analogies from Israel and from sites in the Roman world, including Pompeii and Herculaneum. Mainly based on finds from these two cities, which were destroyed in 79 CE, and the Cave of Letters where the finds in it date to the time of the Bar Kokhba revolt, we are able to date the majority of the incense shovels of this type, which were apparently called vatillum or batillum, to the first–second centuries CE. Yadin argued that although the incense shovel is included among the commonplace decorations in Jewish art, together with the menorah, lulav and etrog, there is no doubt that the incense shovels of the type discussed here are not of Jewish origin and belong to Greek, Roman and later Christian rituals.
In the fifty years that have elapsed since Yadin drew his conclusions, several bronze and iron incense shovels have been discovered in archaeological excavations at Jewish sites from the Early Roman period. Especially noteworthy is a bronze incense shovel of this type from Horbat Petora, near modern day Qiryat Gat, and an iron incense shovel of another type from Horbat Burnat east of Lod—two rural settlements that existed from the Second Temple period until the Bar Kokhba revolt (these two incense shovels will eventually be published by the author). These finds cast doubt on Yadin’s conclusions and raise the possibility that requires further discussion that Jews also used incense shovels during this period, whether for ritual purposes or for more practical uses. Since the incense shovel from Taiyiba was not found in a clear archaeological context, it is impossible to determine if it was used by Jews or pagans.