This article provides the first comprehensive synthesis of one of the least studied styles of column capitals—the Ionic order—in the countryside of the central Palestinian hill country in Late Antiquity. In this study a suggested typology and chronology of the documented finds of Ionic capitals—all are made of local limestone—is presented, while taking into account the geographical distribution of the capital types, and their archaeological/architectural and cultural contexts. Ionic capitals in late antique central Palestine were almost exclusively used in rural settlements of various forms, mostly villages (usually Christian but also Jewish and possibly Samaritan) and monasteries, and were virtually absent from the region's urban centers. While only one capital type represents the (quasi-)canonical or classical style of this order, the other five types clearly demonstrate provincial trends, especially an increasing heterogeneity in the development of the Ionic capital in the region, and the genuine contribution of rural artisans in this respect. The typological diversity and widespread distribution of Ionic-style capitals in the countryside of the late antique central Palestinian hill country is interpreted within local and pan-Levantine historical and socio-cultural contexts. This research inter alia suggests that the specific socio-political history of the central Palestinian hill country during the Early Roman period can partly explain the marked distinction between this area and the regions to its north regarding the evolution of the Ionic style in late antique times. By also using the code-switching model, this study further hypothesizes that the unique Ionic capitals used in the central Palestinian hill country contributed to the development of the cultural identity of the local rural population.
Various media of architectural decoration from the late antique era (fourth to seventh centuries) in historical Palestine (equivalent, for the sake of this study, to modern-day Israel and the Palestinian Authority), notably mosaic floors, liturgical furnishing, and column capitals, have been the target of much scholarly attention since the first half of the 20th century. This is especially true for local mosaic art, which has been extensively studied with relation to both urban and rural contexts, often linked to the investigation of churches, synagogues, and other public and religious structures.1
However, the study of contemporary column capitals has largely focused on their appearance in selected urban centers, such as Jerusalem, Caesarea Maritima, Ascalon, and Eleutheropolis/Beth Govrin.2 As for the countryside, to date no systematic study of column capitals in villages, monasteries, and other rural contexts has been carried out, apart from isolated and often limited discussions on relevant finds from a few excavated or surveyed public buildings. The latter are mostly represented by village synagogues in the Galilee and the Golan3 and churches in some of the large villages of the Negev.4
The present article aims to fill this lacuna by discussing certain column capital finds from rural sites (villages, farms, and countryside monasteries) in the central regions of ancient Palestine in Late Antiquity—the central hill country and its western foothills, and the coastal plain as far north as the Carmel ridge. The Carmel region proper will be also included in the discussion, as it forms a geographical—and in many ways also cultural—transition area between central and northern Palestine. However, since such an extensive theme cannot be exhaustively discussed in a single article, the present study first provides a general review of column capital typology, materials, and distribution in the countryside of central Palestine,5 followed by a detailed presentation and interpretation of a selected style of capital carving—the Ionic order. The reason for my concentration on this style is the fact that it is one of the least studied, specifically with regard to late antique Palestine (and the southern Levant in general), especially when compared to the Corinthian style.
The discussion below interprets the region's geopolitical history starting in the Early Roman period, while suggesting two related agents that left their imprint on the evolution of the Ionic capital in the central Palestinian countryside: The first is a dearth of new monumental architecture between roughly the second and fourth centuries, followed by an architectural renaissance that formed a background for local artisans to develop original decorative styles. The second agent is the development of an independent or semi-independent cultural identity of the rural population in parts of the central Palestinian hill country. This identity also included aspects of architectural and artistic uniqueness, aimed toward urban and certain rural populations within the discussed region and beyond.
The present article further demonstrates how the model of code-switching can be applied to the field of archaeology and material culture studies. Code-switching is a term borrowed from the field of linguistics, originally defined as bilingual speech and writing.6 Recently, the notion of code-switching has been applied to archaeological studies, including those dealing with identity construction among classical- and medieval-period populations. Inter alia, code-switching is now used to explore the fusion of diverse contemporary cultures or identities as reflected in material remains, including architecture.7 Hence, the subject under discussion, namely the fusion of classical traditions and provincial trends in the evolution of Ionic capitals in late antique central Palestine, and its contribution to the shaping of people's identity, can be treated as a case study in this regard.
COLUMN CAPITALS IN THE CENTRAL PALESTINIAN COUNTRYSIDE: AN OVERVIEW
The area under discussion, namely the country's central regions (fig. 1), formed the heart of the province of Palaestina, which, in or slightly before 409 C.E. was divided into Palaestina Prima, Palaestina Secunda, Palaestina Tertia, and Phoenice, with their capitals located in Caesarea Maritima, Scythopolis, Petra, and Tyre, respectively. The great majority of the country's urban centers, including most of its harbor towns, were located in Palaestina Prima. By the sixth century, this province had a predominantly Christian (formerly polytheist/pagan) population, with large minorities of Jews and Samaritans. After the Muslim conquest of 634–40 C.E., the area of Palaestina Prima (as well as that of Palaestina Tertia) was incorporated within the new military district of jund Filasṭīn, whose capital was first located inland at Lydda/Diospolis and starting in ca. 715 C.E. in the nearby, newly established city of Ramla. From the chronological viewpoint, the present article will therefore concentrate first and foremost on the proper Byzantine period, with a slight deviation into the Late Roman and beginning of Early Islamic times (roughly 300–700 C.E.).
When comparing the various geographical regions of central Palestine from the perspective of architectural decoration, especially that of column capitals, some clear differences become clear regarding the materials and styles used. The main identifiable distinction, starting in the Roman Imperial period (second century) and well into Late Antiquity, is between coastal and inland regions. As a general rule, typically decorative (or functional-decorative) architectural elements—notably column capitals and often other column parts—used in public and on occasion in private building projects in settlements located along the Palestinian coastal plain were made of imported marble, usually of Proconnesian origin (column shafts in the Roman period were also made of grey and red granite imported from Troad and Aswan). The marble capitals were imported to the major harbor towns, such as Caesarea and Ascalon, either ready-made or partially-worked, with the latter being completed by local artisans (including marmorarii who specialized in marble).8 Consequently, the dominance of marble resulted in a rather limited repertoire of capital types used in coastal settlements, especially rural ones, during Byzantine times.
During this period, the use of local coastal kurkar (focalized sand) stone was designated mainly for simpler architectural elements and, of course, for creating regular building stones. The lower foothills bordering the coastal plain from the east are characterized by limestone formations. Interestingly, it is not uncommon to find—especially in sites located in relatively inner areas of the coast—an integrated use of kurkar, limestone, and marble, though only the latter was in the form of column capitals.9
The largest harbor towns were responsible for the distribution of imported marble capitals to other urban and rural settlements across the coast and inland (with the urban centers, in their turn, functioning as intermediate distributors). However, archaeological surveys and excavations have clearly shown that the quantity and variety of marble architectural elements was much more extensive in the country's western regions—the coast and the lower foothills—than in the inland regions to the east and south.10 This regionalism seems to have derived from several factors, which partially complemented one another. The first and most obvious reason was the geographic proximity of the coastal and lower foothill settlements to the harbor towns and the convenient communications between them, which was based on the local road system and the moderate topography. Secondly, the brittleness and rough texture of the local stones in these regions (kurkar and a limestone crust locally known as nari) argued against using them for the production of typically decorative elements. Furthermore, the remoteness of the high foothills and the central hill country and their steeper topography made transporting heavy architectural elements from the coastal towns much more difficult and costly. On the other hand, the nature of the various hard limestone formations of the higher inland regions of central Palestine allowed the manufacture of a variety of decorative architectural elements—including column capitals—which often constituted a substitute, either qualitatively equal or inferior, for the marble-made counterparts. Nevertheless, the fact that not an insignificant number of marble items, some extremely large and heavy, were transported to inland cities such as Jerusalem and Eleutheropolis indicates that a similar distribution could also have been possible in the rural hinterland. Yet, this was not the case; presumably, the wish to save time, expense and labor, and the existence of local artisans who specialized in the production of (most likely) cheaper limestone items, were the main motives behind the limited appearance of marble architectural elements in the countryside of inland central Palestine.11
Regarding capital typology, the Corinthian style was the predominant one in central Palestine (as in other parts of the country) since Early Roman/Herodian times.12 In Late Antiquity (but actually since the second century onwards), this was especially true for the coastal regions, where nearly all of the documented capitals from urban and rural sites alike are marble Corinthian capitals. Many of these were imported, apparently ready-cut, from Asia Minor, representing some of the types which are based on the “regular” Corinthian capital/Normalkapitell and date to between the late fourth and sixth centuries (or were otherwise reworked locally from other imported marble items).13 The rather monotonous repertoire of the coastal capitals fits the pan-Mediterranean koiné of the period under discussion, as a representative of “a common architectural vocabulary” based on imported decorative elements.14
In contrast, marble Corinthian capitals are extremely rare in countryside sites in inland central Palestine (though they are somewhat more common in the local urban centers), where they were replaced by locally-produced, hard limestone capitals. Some of these are more or less exact copies of the imported canonical prototypes and seem to be the product of urban (mostly Jerusalemite?) workshops, although the production of such capitals by rural artisans, either based in village workshops or itinerant, should not be dismissed altogether. Other capitals—at least half of the documented examples if not the great majority—represent various provincial derivatives of the “regular” Corinthian style, while reflecting different relations between the basic components of the latter, emergence of new motifs, or disappearance/schematization of some of the basic components and motifs; a few examples may actually be defined as composite capitals, as they combine Corinthian- and Ionic-style (usually atrophied) characteristics. These capitals are barely represented in the hill country cities, and therefore were most likely an original rural development. Alongside the Corinthian style, its derivatives, and the Ionic style, other late antique capital types were also used in the central Palestinian countryside; these were almost exclusively local limestone products whose distribution did not exceed beyond the western fringes of the hill country foothills. They include a limited number of two-zone (“basket”) capitals, most probably originating in Jerusalemite workshops, and a varied array of impost capitals and heterodox capitals, perhaps the products of village craftsmanship.15
IONIC CAPITALS: TYPOLOGY AND CHRONOLOGY, DISTRIBUTION, AND ARCHITECTURAL AND CULTURAL CONTEXT
Ionic capitals formed the second-most common group of capitals in the central Palestinian hill country in Late Antiquity. Ionic capitals are also widely documented in this period throughout northern Palestine, though they do not appear in the central and southern Palestinian coastal region, as well as along the western Galilee coast. Until the beginning of the Byzantine period two main groups of Ionic capitals dominated in Palestine (and the Levant in general). The first group stylistically followed, either reliably or freely, the canonical prototypes of this order, the basics of which had been developed and crystalized in Greece and Asia Minor during the Classical, Hellenistic, and Early Roman periods (between ca. the fifth/fourth century B.C.E. to the first century C.E.). More specifically, the Ionic capitals which prevailed across the Mediterranean in the late Hellenistic and throughout the Roman period were the closest to the canonical model that Vitruvius defined in his famous description of the Ionic order and its optimal proportions.16 These capitals are characterized by a decorated or plain abacus; narrow echinus decorated with ovolo and darts; pronounced volutes with a canal which winds two or three times around the central eye and passes under the abacus; a pulvinus, which can be either plain or added by balteus at its center (fasciae also decorate both sides of the balteus, though they only seldom appear on late antique capitals); a plain or decorated astragal under the echinus; and a very short, round base/shaft. The second group—like some of the Corinthian and Corinthian-derived capitals of inland central Palestine (and elsewhere)—reflects provincial evolutionary trends which are expressed in varying relations between the traditional components of the style, some of which disappear or become schematized, or in the emergence of new motifs, as explained below.17
According to Avi-Yonah, sometime in the early fifth century a significant turning point occurred in the development of the local Ionic capitals, with the provincial variants gradually becoming predominant, a change which brought some scholars to designate these capitals as “pseudo-Ionic.”18 However, it seems that in neighboring regions, such as Syria,19 and possibly in certain areas of Palestine as well, this “turning point” of the main stylistic trends which characterized it had already occurred in the Hellenistic and Roman periods. All of these capitals (both groups), as well as the great majority of their Hellenistic and Roman predecessors, were made of local stone formations, notably limestone (and basalt in parts of northern Palestine).20 These trends were quite markedly different from the above-mentioned case of the majority of the locally-documented Corinthian capitals, which reflect canonical, pan-Byzantine/late antique styles in either imported or local stone.
A close examination of the distribution of Ionic capitals in late antique central Palestine shows that they were almost exclusively used in rural settlements of various forms, mostly villages and monasteries. This fact indicates that Ionic capitals in this region were produced by rural artisans, either village-based or itinerant, a result similar to the local derivatives of Corinthian-style capitals whose distribution was also restricted to the countryside proper. The rarity of Ionic capitals in the cities of central Palestine can possibly be explained by a perception among late antique urban architects and stone artisans (and probably urbanite consumers) that Ionic capitals represented an inferior rustic style, unsuitable for cities’ colonnaded streets and public edifices.21 This artistic/cultural distinction is especially telling in Jerusalem and Eleutheropolis, in which the (canonical) Corinthian style/s dominated, while in many rural settlements in their immediate hinterlands, Ionic capitals (primarily “provincial” ones) were commonly used. In this respect, a different picture is reflected in cities in northern Palestine, such as Hippos/Sussita, Scythopolis and Sepphoris, where during the Byzantine period there was rather widespread use of Ionic capitals. These limestone or basalt capitals (some of which may represent reused Roman-period capitals) were largely fashioned in a style that can be termed as quasi-canonical or simply classical, as they maintained the basic components of the Vitruvian model, though their quality and proportions differ, sometimes considerably, from the latter.22 As such, they were similar to their equivalents from the surrounding countryside, as well as from urban and rural settlements in northern Transjordan and southern Syria (the Ḥauran).23
As noted by Avi-Yonah, the changes in the development of the Ionic capital in Palestine occurred in three directions: the volutes design, the echinus decoration, and the shaft decoration.24 The following is the first, albeit tentative, typo-chronology of the Ionic capitals documented in late antique rural contexts in central Palestine, which altogether includes six types. It takes into account the geographical distribution of these types, and—when possible—the archaeological/architectural and cultural contexts.
This type represents the quasi-canonical or classical style. It is characterized by a shaft which is clearly separated from the echinus, pronounced and deeply-grooved volutes with a pulvinus and a balteus (on occasion, fasciae appear as well), and echinus that has rather low-relief ovolo and strongly-emphasized darts. Occasionally, there might be additional motifs; the abacus does not always exist.
As mentioned, these characteristics appear on many of the late antique basalt-carved Ionic capitals from northern Palestine and southern Syria. However, limestone specimens of Type 1 are rare in rural contexts in central Palestine. Published examples include single finds from Tell Fari‘a (fig. 2: 1)25 and ‘Abud (St. Mary/Sitti Mariam Church),26 and two capitals from Deir Dosi/Monastery of Theodosius (fig. 2: 2).27 The ‘Abud capital is the simplest, with the basic ovolo and dart motif on the echinus and a plain astragal and shaft. One of the items from Deir Dosi is actually an impost capital, with a high, plain trapezoid impost block (with no abacus) and a Greek cross within a medallion engraved at the center of the echinus; this latter detail is rather unusual, as early Byzantine Ionic impost capitals normally bear a cross/monogram at the center of the impost block.28 The details of the second Ionic capital from Deir Dosi are less clear (based only on the available published photograph), though this capital seems to be somewhat larger than the former.29
The most richly-decorated capital of this type is the one from Tell Fari‘a/al-Far‘ah, whose shaft is densely covered with three registers—the upper one depicts a scroll of leaves and tendrils, the middle one is in the form of a continuous branch (“laurel wreath,” according to Avi-Yonah), and the lower one is composed of “alternating rosettes and concentric circles, separated by lotus flowers.” It also has an astragal decorated with a beads-and-reels motif. Avi-Yonah termed the overall ornamental pattern of this capital as an “Oriental profusion of decorative motifs.”30 In other words, this capital reflects a mixture of the classical and provincial (in this case Levantine) styles. Also noteworthy is the high shaft of this capital, a characteristic which, generally speaking, became the norm among late antique Ionic capitals (see below) and is clearly different from the very short shaft/base of the canonical prototypes.
The exact dating and architectural context of the Tell Fari‘a/al-Far‘ah capital are unclear, since it constitutes a surface find; still, since northern Samaria was a region inhabited in Late Antiquity by a predominantly Samaritan population, it is likely that the capital originated in some contemporaneous prominent structure, such as a Samaritan synagogue. The capital from ‘Abud was indeed incorporated in secondary use in the later (Early Islamic and medieval) St. Mary Church, together with other capitals, though an excavation carried out at the building's foundations indicates that it was built over a Byzantine-period church from the fifth century, to which the reused Ionic and other capitals were attributed.31 The Deir Dosi capitals were documented (together with other Byzantine-period architectural elements) in the courtyard of the present-day monastery, though they can be reasonably attributed to the earlier coenobium which was founded at the spot in the late fifth century.32
This type, like the four which follow, clearly represents the provincial trends, especially the increasing heterogeneity, in the development of the Ionic capital in the region, and the genuine contribution of rural artisans with this respect. It is characterized by a straight shaft which forms more or less one piece with the echinus, and round corner bosses (or quasi-bosses, which protrude slightly, if at all, from the echinus surface). These bosses are either plain or decorated with rosettes or crosses. The pulvinus is either plain or bound by a schematic balteus, and the echinus is sometimes decorated with a cross within a medallion or with another motif. The abacus was not always added.
Examples of this type were found at a number of sites, including Khirbet Umm el-Ḥammam (plain [?] bosses; shaft/echinus decorated with a conch-like [?] motif),33 Beit Thul (disproportionally-large bosses, decorated with Greek crosses; fig. 3: 1),34 Beit Shemesh (two capitals with bosses decorated with rosettes; one has an echinus decorated with a cross within a medallion; fig. 3: 2, 3),35 and Eshtemo‘a (plain bosses; the item is classified as “unfinished capital”).36
Most of these capitals (from Khirbet Umm el-Ḥammam, Beit Thul, and one of the Beit Shemesh specimens) were surface finds; therefore, their exact date and architectural context are unclear. The second capital from Beit Shemesh was retrieved from the excavations of the so-called “Byzantine monastery” there, though no datable finds from this structure have been published; still, the crosses which adorn this capital—as well as the one from Beit Thul—allow dating these items to the fifth century at the earliest.37
The capital from Eshtemo‘a was found out of context, though it was attributed to the first construction phase of the local synagogue, dated by the excavator to the late third/early fourth century. This dating was based on the latest pottery sherds found under the foundations, and on the similarity between this edifice and the synagogue excavated at nearby Ḥorvat Susiya.38 However, as Yeivin himself noted, some of the ceramic types represented in the Eshtemo‘a assemblage continued until the beginning of the Byzantine period, while the more acceptable dating of the Ḥorvat Susiya synagogue is the late fourth/early fifth century.39 Therefore, it seems likely that the Eshtemo‘a synagogue was also built at about that time, which determines the dating of this capital.
This type has a conical shaft, which—like Type 2—forms more or less one piece with the echinus with round corner bosses (or quasi-bosses) that are either plain or decorated with volutes. The echinus is either plain or decorated with a cross within a medallion or with another motif. The abacus was not always added, and there was no pulvinus, with the exception of one specimen.
Published examples include Khirbet Abu Rish (four pairs of plain, atrophied bosses that protrude from the four corners; the echinus is decorated on one side with a Greek cross within a medallion and on the opposite side with a concentric disc/sun circle; fig. 4: 1)40 and Khirbet Istabul (two capitals: one has four pairs of bosses decorated with atrophied volutes with long caules that emerge from the base of the shaft [fig. 4: 2]; the second has bosses decorated with deeply-grooved volutes with long caules that emerge from the base. The grooved pulvinus has a triple balteus, which continues as a triple strip and integrates with a rope-motif band/astragal that encircles the base [fig. 4: 3]).41 Another possible example comes from the Northern Church/Jamia‘ al-Yatim mosque at Shiloh (the single, unclear illustration of this capital may indicate that it has plain atrophied bosses).42
The capital from Khirbet Abu Rish was attributed to the building's second construction phase, from the sixth century, when it functioned as a monastery. The first capital from Khirbet Istabul was a surface find.43 The second capital from the same site was excavated in a church that was apparently founded in 700/1 C.E., according to a dated Greek inscription embedded in the original mosaic floor of the nave. This makes this Ionic capital the latest published example of its kind. Still, we should not preclude the possibility that this capital (and others which were found in the building) represents a reused element which originated from an older, Byzantine-period structure (church?) that existed at the site or in its vicinity. At any rate, the similar design of the long-caules volutes of this capital and the design of the surface-find capital may be an indication that they should be dated contemporarily and perhaps be assigned a common architectural context. However, their base diameters are not identical, namely 0.53 m for the surface-find capital and 0.45 m for the excavated one. As for the Shiloh capital, it is indeed embedded in secondary use in a medieval mosque; yet the latter was built over the remains of a late antique church which functioned between the fifth and eighth centuries;44 therefore, a date within the fifth or sixth century is reasonable.
Another site which yielded a group of four capitals that can be related to Type 3 is the Monastery of Martyrius. All of these capitals were likely produced by the same workshop and even by the same craftsman. Still, with the exceptions of their general similar shape—which resembles the Corinthian-style calathus, the atrophied volutes carved on their abacus (not on defined bosses projecting from the latter), and the existence of a pulvinus bound by a schematic balteus, each capital presents a unique decoration pattern. Two of the specimens (fig. 5) have a shaft decorated with simple geometric and vegetal motifs, and an echinus decorated with either a horseshoe-shaped motif, a cross with an ivy leaf, or cross with down-curved horizontal arms (termed by the excavator as a “schematic volute”). The other two capitals (fig. 6) each have four pairs of large, schematic leaves which emerge from the base of the shaft and end beneath the voluted bosses. The leaves have a central vein and either a straight or concave surface. This decoration—especially in the case of the capital with the concave leaves—creates an illusion of a rather defined echinus. The area between the leaves and the capital tops contains four different motifs. In one capital, the volutes and echinus sides bear an amphora on top of a cross with down-curved horizontal arms, and a small cross on top of an ivy leaf (divided by quasi-canalis ivy tendrils which emerge from the volutes bottom), and the pulvinus sides are decorated with a simple cross within a medallion and with a stylized cross with oval horizontal arms. In the second capital, the volutes and echinus sides have a semi-circular motif on top of an ivy leaf and a small cross on top of an ivy leaf (divided by canalis which emerge from the volutes top), and the pulvinus sides bear a stylized cross with oval horizontal arms and a simple cross. All four of the capitals from the Monastery of Martyrius were found in the refectory of the complex and were attributed to its renovation phase that was dated to the mid-sixth century.45 The Corinthian-style calathus decorated with stylized leaves (atrophied acanthus leaves?) allows defining these capitals as pseudo-composite, though based on their general resemblance to the above-mentioned Type 3 capitals I classified them as a sub-group of the latter.
This type has either a short or relatively tall, straight shaft, a defined echinus that can sometimes be rather wide (namely, it starts below the bosses line) and which is either plain or decorated with a cross (within a medallion or freestanding) or with another motif. This type has round bosses (or quasi-bosses) that are either plain or decorated with volutes, rosettes, crosses or other motifs, and a pulvinus that is either plain or bound by a schematic balteus. On occasion the pulvinus is missing altogether, and the abacus does not always exist in this type.
Examples are known from Zamarin (bosses decorated with deeply-grooved [?] volutes connected by a large Greek cross at the center of the echinus; fig. 7: 1)46 and Ḥorbat Damon on the Carmel (disproportionally-large, adjacent bosses, decorated with crosses and leaves between the arms of the crosses),47 ‘Abud (St. Mary Church; four pairs of bosses that protrude from the four corners of the capital and are either decorated with atrophied volutes or plain),48 Ḥorvat Zikhrin (four pairs of plain bosses that protrude from the four corners of the capital [and therefore no pulvinus], and plain echinus; fig. 7: 2),49 Khirbet Jabaris (bosses and echinus decorated with wide circular depressions)50, Sheikh Gharbawi (plain bosses and a protruding pyramid-shaped [?] boss at the center of the echinus),51 Khirbet Meiyita (two capitals; plain [?] bosses),52 Ramet el-‘Amle (bosses decorated with narrow circular depressions, possibly atrophied sun circles; fig. 7: 3),53 Umm er-Rus (bosses decorated with atrophied volutes, and echinus decorated with a Greek cross within a medallion; fig. 7: 4),54 Khirbet Istabul (four pairs of atrophied bosses that protrude from the four corners of the capital [and therefore no pulvinus] and decorated with atrophied volutes which emerge from the base of the shaft, and echinus decorated with an amphora/vase on one face; fig. 7: 5),55 and Ḥorbat Burgin (two capitals; plain [?] bosses and echinus decorated with a Greek cross within a medallion; fig. 8).56 Additional possible examples of Type 4 capitals include Deir el-‘Asal (one plain or defaced boss and another one decorated with a schematic rosette),57 Khirbet el-Jiljil (atrophied round bosses decorated with Greek crosses on the front side, and square plain bosses on the rear side),58 and Yatta (bosses decorated with rosettes, and echinus decorated with a freestanding Greek cross).59 Interestingly, the excavations at the monastic complex of Mount Nebo in Jordan revealed a limestone item identified as a “basin-lavabo,” which is shaped like a capital very similar to Type 4, namely it has a plain “abacus,” disproportionally-large bosses decorated with Greek crosses, and a plain “pulvinus” with a schematic “balteus.” According to Acconci, this “pseudo-capital … clearly appears to have been inspired by the schematized typology of the Ionic capital re-elaborated in the Palestinian region between the turn of the 5th century and the entire 6th century.”60
The capitals from Zamarin, Khirbet Jabaris, Ramet el-‘Amle, Deir el-‘Asal, Khirbet Istabul, and Yatta were found out of context, sometimes in association with church buildings, though the latter were not excavated and their date is therefore unclear.61 Still, the crosses depicted on the Zamarin and Yatta capitals likely point to a date not before the fifth century. The capital from ‘Abud was reused, as mentioned, in a medieval church that was built over a Byzantine-period church. The capital from Ḥorvat Zikhrin was found in the excavations of an affluent village peristyle dwelling which was built in the mid-fifth century. The capitals from Ḥorbat Damon and Khirbet Meiyita were also found in excavated churches whose construction was dated to the fifth-sixth centuries and to the second half of the fifth century, respectively. The capital from Sheikh Gharbawi was retrieved from the excavated remains of a monumental building which was built in the Byzantine period—apparently not before the fifth century—over the remains of sacred tombs and became a Christian pilgrimage site. The Umm er-Rus capital was apparently found in association with the church excavated at the site which was dated to the sixth or seventh century. The capital from Khirbet el-Jiljil was retrieved from a partially-excavated farmhouse whose foundation was dated to the fifth century; it most likely adorned one of the building's column-equipped units, such as a portico or arched room (the same is true also for the other, inscribed Ionic capital from the site). Finally, the capitals from Ḥorbat Burgin belong to two monolithic columns which adorn a rock-cut burial cave (of the arcosolia type). Although the cave was found looted, it can be safely dated to the Byzantine period—apparently to not before the fifth century—based on the crosses that appear on the capitals.
This type differs from Type 4 mainly by the protruding band/astragal under the echinus, namely at the joint between it and the shaft. On occasion, there are two or three stepped bands instead of one, which give the shaft a somewhat conical layout. These bands are usually plain, excluding a single published example where the band is decorated with a rope motif.
A few such capitals were found at sites located on the Carmel—at Khirbet Nasus (one capital, with broken bosses, the echinus decorated with an unidentified motif),62 at Ḥorvat Zikhrin in southern Samaria (five capitals; each has four pairs of plain bosses that protrude from the four corners of the capital [and therefore there is no pulvinus], usually with a stylized leaf under each boss, and echinus which is either plain or decorated with a freestanding Greek cross or a cross within a medallion; figs. 9, 10: 1)63 and at Mazor in the Samaria foothills (one capital with four pairs of plain, semi-circular bosses that protrude from the four corners; hence, here too, no pulvinus exists).64
However, a much larger number of capitals comes from sites in the southern part of the hill country. These include Jifna (two capitals: one with plain bosses and a protruding square-shaped boss with a diagonal cross-section at the center of the echinus; the second capital has disproportionally-large bosses, deeply-grooved volutes, a six-petalled rosette at the center of the echinus, and an astragal decorated with a beads-and-reels motif);65 Deir Rafat (plain bosses);66 Beit Jimal (bosses decorated with Greek crosses within medallions, and echinus decorated with a cross monogram. The lower part of the echinus is rather broad and convex; fig. 10: 2);67 Beit ‘Anun (plain bosses, echinus decorated with a rhombus motif. The sides of the echinus and the abacus are divided into three rectangles by a pair of vertical grooves; fig. 10: 3);68 Khirbet Teqo‘a (plain bosses, echinus decorated with a sun-circle motif carved on a round boss);69 Khirbet el-Ḥarayiq (four pairs of plain, semi-circular bosses that protrude from the four corners of the capital [again, with no pulvinus] and are decorated with atrophied rosettes, and echinus decorated with a Greek cross within a protruding medallion. The band under the echinus is decorated with rope motif; fig. 10: 4);70 Khirbet Kafr Jul (plain, disproportionally-large bosses);71 Khirbet Tawas (two capitals: one has bosses decorated with a central round cavity, possibly representing the sun disc motif, and pulvinus with balteus decorated with a diagonal coil motif. The second capital has plain [?] bosses and pulvinus with a schematic balteus that continues as a plain strip down the echinus; fig. 10: 5);72 Khirbet Umm Deimine (two capitals: one has bosses decorated with Greek crosses, and another freestanding cross is carved on the echinus-abacus joint [fig. 10: 6]. The second capital has bosses decorated with rosettes, echinus decorated with a Greek cross within a protruding medallion, and pulvinus decorated with a Latin cross that replaces the balteus [fig. 10: 7]);73 lower Herodium (central church; bosses decorated with atrophied volutes, echinus decorated with a freestanding Greek cross);74 and Maḥatt el-Urdi (bosses decorated with Greek crosses within medallions, echinus decorated with an identical cross flanked by asymmetrical acanthus leaves).75
The capitals from Khirbet Nasus, Jifna, Deir Rafat, Beit ‘Anun, Khirbet Teqo‘a, Khirbet el-Ḥarayiq, Khirbet Kafr Jul, and Khirbet Tawas (the examples published by Mader and Kochavi) were surface finds or reused items, though some were found within or near remains of Byzantine-period churches, mostly unexcavated. The Type 5 capitals from Ḥorvat Zikhrin were found in the excavations of the above-mentioned peristyle dwelling (where one Type 4 capital was also found) and village church. The construction of both these structures was dated to the mid-fifth century. The capital from Mazor was excavated within the remains of a Byzantine-period structure (a monastery?) which was probably built in the late fifth/early sixth century, though the capital itself was in secondary use in the medieval period. The capitals from Khirbet Umm Deimine were attributed to the second, monastic phase of a large structure, the beginning of which was dated to the fifth century. The Beit Jimal, lower Herodium, and Maḥatt el-Urdi capitals were found within or nearby excavated churches, the first dated to the fifth or sixth century and the other two to the sixth century. Noteworthy regarding the capital from Maḥatt el-Urdi is the fact that this site is located on the outskirts of the city of Eleutheropolis, which means that this capital is probably the only one reported from an urban or urban-annexed context in central Palestine. Finally, the other capital from Khirbet Tawas (published by Peleg) was also found in the excavation of a church. The foundation of this edifice was dated to the seventh century. Like the case of the Khirbet Istabul church, this capital is one of the latest known examples of Ionic-style capitals documented in the region, unless it represents an earlier item (apparently from the fifth or sixth century) which was reused in the seventh-century church.
This type is the rarest, and its common characteristics include a short, conical shaft, defined echinus, and a more or less continuous decoration which covers the echinus and the front of the shaft.
The first example comes from Khirbet esh-Sheikh Meisar in the northwestern Samaria Hills. It has rather deeply-grooved volutes, a pulvinus with balteus, and a narrow abacus; the area between the volutes and the front of the shaft is decorated with simple geometric/vegetal motifs which might represent a stylized palmette (fig. 11: 1). The capital was a surface find and therefore its exact date and context are unknown. Ne'eman dated it to the Byzantine period, while according to Zertal and Mirkam it is a Roman-period element.76 However, the close similarity between the front decoration of this capital to the decoration of two locally-produced Corinthian capitals from the fifth or sixth century found at other sites in western Samaria allows attributing the Khirbet esh-Sheikh Meisar capital as well to the Byzantine period.77 In Late Antiquity, this region was inhabited by both Samaritans and Christians, which makes the attribution of the Khirbet esh-Sheikh Meisar capital to either a synagogue or a church equally valid.
The other Type 6 capitals known to me were found at Khirbet al-Karmil. The first of apparently two or three similar capitals was published by Mader. According to its illustration, it has rather deeply-grooved volutes and a decoration of what can be defined as four narrow leaves or pointed aediculae on the echinus. Four other somewhat broader aediculae-like motifs decorate the upper part of the shaft and another row of a similar though upside-down pattern appears on the lower part of the shaft (fig. 11: 2). This capital was a surface find, albeit documented within the remains of a church dated to the fifth-sixth centuries.78 There are two photographs of similar fragmentary capitals in the British Mandate-period file of Khirbet al-Karmil in the Israel Antiquities Authority Archive. One of the fragments may actually represent the complete capital documented by Mader in the early twentieth century; its preserved volute appears atrophied, though this could also be the case in the capital published by Mader, whose drawing is not necessarily accurate. The second fragment also presents an atrophied volute, but it has aediculae-like decoration only on its upper shaft and not on the echinus.79 With respect to the echinus decoration of the capital published by Mader, the pattern may have stemmed from a Hellenistic-Roman element, namely the fluted leaf decoration. It appeared especially in Greece and Asia Minor, and continued into Late Antiquity, in a capital type termed by Kautzsch “Kapitell mit Schilfblättern.”80 However, the pointed leaves on these capitals are quite elaborate (in the least detailed case, the leaves have a midrib), and appear as a continuous encircling wreath placed over a lower wreath of acanthus leaves. This pattern, which obviously derived from the Corinthian order, is of course much different from the Khirbet al-Karmil capital, whose short row of plain “leaves” is not more than a remote echo of the Aegean prototype.
Chronologically speaking, the above review indicates that each of the capital types was well-represented in the countryside of central Palestine by the fifth century, with some (if not all) apparently already introduced into the rural settlements of the region in the fourth century. It is unclear whether Type 1 capitals continued to be produced after the fifth century, though there is rather solid evidence for the continuous manufacturing of Types 2 to 5 and possibly Type 6 as well during the sixth and maybe even the seventh century. As mentioned, the discovery of a Type 3 capital in the early eighth-century church of Khirbet Istabul is not enough to confirm production of Ionic-style capitals at such a late date, and for the time being I prefer interpreting this isolated item as a reused fifth- to seventh-century capital. At any rate, it should be emphasized that although Type 2 to 6 capitals appeared in central Palestine not before the fourth century, they or their prototypes already existed in rural settlements in parts of northern Palestine and Syria in Roman times (see discussion below). Altogether, the above database of published and unpublished capitals, which includes a large number of decontextualized and spoliated items that cannot be securely dated, demonstrates the great diversity of the capitals. This diversity reflects a large number of local workshops and a broad range of skills, sources of influence, and localized artistic tastes and interpretations, alongside the stylistic simplification of the capitals compared to their predecessors. Therefore, the general chronology of fourth/fifth to sixth/seventh centuries suggested above cannot be finalized and no clear development trends can be identified.81 The only exceptions in this regard are relatively well-dated capital finds which lack close parallels in earlier or later contexts, such as the unique group of Type 3 capitals from the Monastery of Martyrius and, to some extent, the Type 5 capitals from Ḥorvat Zikhrin.
From the viewpoint of settlement and architectural contexts, Ionic capitals were documented in villages and in isolated monasteries. In the former they were usually in use in communal, basilical church buildings; however, they were also found in at least one synagogue (at Eshtemo‘a), and on occasion in private dwellings (Ḥorvat Zikhrin and Khirbet el-Jiljil) and in rock-cut burial caves (Ḥorvat Burgin). In the isolated monasteries, which were usually characterized by narrow, column-less chapels, capitals were likely to be found in peristyle courtyards, porticos and relatively spacious units such as refectories.
Here one should also mention the diversity in the size of the documented capitals. The items whose dimensions are known (based on their publication or on my personal observation) can be roughly divided into two groups—those with a base diameter of ca. 0.3–0.35 m and those with a base diameter of ca. 0.4–0.45 m (a few examples are ca. 0.5 m in diameter). Based on the context of the capitals whose architectural affiliation is known or reasonably-estimated, it can be concluded that capitals of both groups were used in the interior decoration of public buildings (notably churches), while columns in external/open-air spaces (peristyle courtyards and porticos) tend to be thicker, hence they were likely adorned with capitals of the second group.
The diverse typology and widespread distribution of Ionic-style capitals in the countryside of the hill country of late antique central Palestine should be interpreted within the local and pan-Levantine historical and socio-cultural contexts. First, the varied design of the shape, decorative components, and proportions should be attributed to two main factors: first, the nature and intensity of influencing sources and architectural/artistic traditions and second, the apparent large number of local workshops/artisans producing the capitals. A third, though likely lesser factor was the differences in the characteristics of the various local limestone formations, which may have had some influence on the quality and appearance of the finished items. Obviously, the few Type 1 capitals reported from late antique rural contexts of the discussed region form a negligible local expression of an artistic style well-rooted in the broader Mediterranean region (and beyond) since at least Hellenistic times. The predominance of this classical/quasi-canonical style among the Ionic capitals documented in Roman- and Byzantine-period architecture from northern Palestine and southern Syria82 can be seen as the main source of influence on the late antique stone artisans from central Palestine who produced a limited number of Type 1 capitals. However, it should be remembered that classical Ionic capitals also adorned some of the built and rock-cut structures, including monumental tombs, from the Early Roman period (notably the late first century B.C.E. to the late first century C.E.) in urban and rural contexts throughout the central hill country.83 Many of these centuries-old remains were surely visible in late antique times, and it is not unlikely that they too constituted some source of inspiration for the Byzantine-period stone artisans.
Nevertheless, in contrast to locally produced Corinthian capitals, the late antique rural artisans had virtually no close acquaintance with imported Ionic capitals that would represent classical prototypes which might have formed a basis for imitations in local stone and a background for the formation of a much more uniform architectural vocabulary. Furthermore, it seems that the presence of classical Ionic capitals in ancient ruins and tombs located within and around late antique rural settlements in the central hill country and in contemporaneous buildings in more northern regions was not attractive enough to encourage copying and assimilation—at least not on a wide scale—into the architectural landscape of the central Palestinian hill country. Apparently, in the case of Type 1 Ionic capitals, the main reason for this unsuccessful penetration of the classical Ionic style was the time distance that separated their creation (and the cultural and geopolitical world they represented) and the late antique population living on the same land, a distance which—contrary to other regions—was not bridged by a constant use of this artistic order.
But what about the quite prominent difference with respect to the extent of the use of the classical Ionic style between the populations of late antique northern Palestine-southern Syria and that of the central Palestinian hill country? Was this merely a greater artistic conservatism of the former compared to the latter? Indeed, regionalism and even sub-regionalism in late antique material culture are clearly recognizable in the local archaeological record, yet numerous material media do reflect a relatively uniform expression across the southern Levant and often beyond, regardless of the existence of past social, ethno-religious and, at times, geopolitical boundaries. People have constantly traversed the different territories of the southern Levant, carrying with them ideas and influences, while often making the geographical distance between the Golan or Ḥauran and the southern Palestinian hill country not more considerable than that between, say, the northern hill country and most of the Galilee, which nevertheless present different artistic preferences in the case of Ionic capitals.
I therefore suggest that the specific socio-political history of central Palestine and especially its hill country during the first two centuries C.E. had a direct impact on the extent of continuity of monumental public structures in the countryside and can partly explain the marked distinction between this area and the regions to its north regarding the evolution of the Ionic style in late antique times. In the countryside of northern Palestine there were monumental pagan temples since at least the second century, and virtually all of them existed well into the fourth century. Some of these temples were replaced within the course of the fourth or fifth century by Christian churches at a time when many other churches were founded in rural and urban contexts elsewhere in the region.
It seems likely that richly-decorated (and monumental in their own way) Jewish rural synagogues also already existed in eastern Galilee during the third century. However, all of the documented examples of Jewish rural synagogues continued to function during at least part of the Byzantine period. Archaeological evidence suggests that synagogues were founded in the Jewish Golan only in the fourth century, but these structures too continued to exist for several centuries; of course, both regions have yielded evidence for the existence of synagogues already in the Early Roman period.84
In other words, this more or less consecutive existence of monumental cultic buildings in the northern Palestinian countryside preserved the tradition of classical architecture and decoration trends—in this case, the Ionic style—which prevailed in this area during the Roman period. This continuity, on the one hand, prevented the disappearance or fading of the classical traditions from the cultural memory of the local population (be it pagan, Jewish or Christian), regardless of the political or social changes that these populations might have experienced.85 On the other hand, it left a relatively narrow space for the development of new provincial trends. More often than not, the classical basics of the Ionic-style capitals were preserved, alongside an understandable evolution in the form of stylistic changes and additions which occurred in the course of the Late Roman and Byzantine periods.
However, as is commonly agreed, the two Jewish revolts against the Romans in 66–73/4 and 132–35 C.E., and especially the latter, resulted in a substantial demographic setback as evidenced by a sharp settlement decline among the Jewish communities of central Palestine, notably in the southern hill country and its foothills. The Jewish settlement in northern Judea (including Jerusalem and its environs) never regained its former intensity, while the rehabilitation of the Jewish communities in southern Judea and probably also along the coast apparently occurred after a near-total settlement gap of some two generations, namely towards the end of the second century (a period during which the center of Jewish life in Palestine moved to the Galilee). At any rate, a true vibrancy of the Jewish settlement in central Palestine is archaeologically and textually attested mostly from the fourth century onwards. The former Jewish territories as well as some of the areas with a continuous Jewish presence were gradually occupied between the second and fourth or fifth centuries by migrant pagan and Samaritan populations who originated in other parts of the country and in neighboring and more remote regions. The pagan component of the population became, starting in the fourth but most intensively since the fifth century, the basis of the steadily-growing Christian community in central Palestine and the country in general.86 Indeed, it seems that monumental rural pagan temples of the kind familiar from northern Palestine and Syria never existed in the countryside of central Palestine, either before or after the Jewish revolts. Yet, Jewish synagogues which pre-date the second and even first revolt against the Romans are documented in the region, and these structures functioned as carriers of their contemporary architectural and decorative styles, albeit on a modest scale.87 Following the second revolt, Jewish monumental architecture virtually vanished from rural central Palestine, to reappear only around the late third/early fourth century at the earliest. This is more or less the period when the first monumental Samaritan synagogues were founded in the Samaria Hills and on the Carmel; strikingly enough, the archaeological evidence has not yet confirmed the existence of earlier Samaritan synagogues in these regions,88 contrary to the situation reflected by the Jewish synagogues of Judea, Galilee and the Golan. And, as already mentioned, a wide-scale construction of Christian churches accompanied the region's built-up topography only around the fifth century.
In my view, it is therefore possible to posit two related agents which left their imprint on the evolution of the Ionic capital (and probably additional media of architectural decoration) in the central Palestinian countryside. The first is a vacuum of monumental architecture, which occurred specifically in the typical former Jewish territories and lasted for a period that can be estimated between about a century and half, if not more than that, to at least two centuries and apparently longer. In southern Judea synagogues reappeared around the late third/early fourth century and in northern Judea and parts of southern Judea, the renewal of monumental public architecture is marked by the foundation of the first churches not before the mid- fourth/early fifth century.89 This architectural renaissance functioned as a most convenient background for local artisans to develop new, more or less original decorative styles on the basis of the classical characteristics, as well as to adopt and modify other provincial styles that already existed in rural areas farther to the north.90
Stemming from this situation is what I identify as the second agent, namely a consolidation of an independent or semi-independent cultural identity of the rural population in parts of the hill country of central Palestine. Evidence of this cultural identity is notable in the case of the Ionic capital (notably Types 2 to 6) and other rustic capital types not discussed here.91 This identity developed, apparently both consciously and unconsciously, as an ideological declaration of uniqueness and local patriotism not only toward the local urban centers (and more remote, including typical rural regions), but also toward certain sectors of the region's own rural population.92 The fact that the lion's share of Ionic capitals in the central Palestinian hill country was associated with Christian communities suggests that this endemic artistic trend was initiated and developed by the local Christian population, and that its physical manifestations—the capitals themselves—were only rarely adopted by Jewish and Samaritan communities living in the same region.93 It can be rather safely concluded that this architectural-artistic medium has contributed to a certain awareness of togetherness at least among part of the Christian rural population of that geographical territory.94 Whether this new, Christian-oriented rural identity (which may have had some fourth- to fifth-century pagan roots) was an outcome of the above-mentioned demographic changes cannot be proven. Yet, it is possible that these changes which impacted the origin of the region's pagan and Christian populations contributed to the creation of some unique artistic dialects within the local architectural vocabulary.
Although relatively few examples of late antique public structures in the region under discussion are documented in which all or most of the decorative array (notably column capitals) was preserved, it is well known that contemporaneous capitals of different artistic orders (and subtypes of the same order) often adorned the same structure. And, of course, contemporaneous buildings in the same settlement (or in neighboring settlements) might have been adorned with completely different capital types. These were represented not only by Ionic-style examples, heterodox capitals and/or rural derivatives of the Corinthian order, but very often also by canonical or quasi-canonical Corinthian capitals, usually made of limestone. In other words, the localized architectural identity cultivated by the population of the central hill country, especially the Christians, was well embedded in the broader Mediterranean late antique world, and was never truly separated from its most prominent decorative expression—the Corinthian capital. Nor was this population—regardless of its varied geographical and ethnic origin—totally severed from the past architectural heritage of the region. The rural milieu of the central Palestinian hill country was therefore in a constant negotiation between the pan-Mediterranean koiné and the “provincial” interpretation of the latter, which in itself combined localized and broader Levantine trends.95
Interestingly, some lines of similarity and difference can be identified when comparing the situation presented here to a model offered by Moshe Fischer regarding architectural decoration, including column capitals, in second- and third-century Roman Palestine.96 According to Fischer, during that period “official provincial architectural decoration was, at its beginning, based almost exclusively on imported art.” However, during the third century, when imported marble items were rare, “architectural decoration based on local stone became the exclusive official art of the country. … Architectural decoration of this period bears the general lines of the ‘classical’ design … Later on, these artisans started to neglect those principles,” as reflected especially by Galilean synagogues of the third and fourth centuries. Fischer attributed this “disintegration of ‘classical’ forms” to the weakening of centralized rule, which formerly “enabled regulation of the production and the commercialization of art.” Thus, in these politically-turbulent times “a real provincial art, with strongly individualized artistic trends, flourished. … On this level, therefore, a more important role is to be attributed to environmental, sociological and religious factors.” Indeed, as noted above, a similar contrast (though by no means full dichotomy) between provincial trends and classical traditions occurred also in the case of late antique column capitals in the central Palestinian hill country, both on the stylistic level (i.e., Ionic vs. Corinthian order) and contextual level (i.e., countryside vs. urban centers, Christian vs. Jewish and Samaritan rural settlements). However, contrary to the situation in the third and probably also the fourth century, the marked provincialism, and—at times—individualism of column capitals craftsmanship in rural central Palestine between the fifth and seventh centuries occurred alongside an influx in the import of marble items, notably Corinthian capitals. In other words, the common use of imported marble capitals and limestone imitations which represented canonical trends (or “official” ones, to paraphrase Fischer's terminology) in the central Palestinian countryside did not interrupt the vibrant activity of local craftsmen who at the same time manufactured a variety of rustic Ionic and other capital types. Furthermore, as suggested above, this independent artistic vein may have even been intensified not despite but because of and as a reaction to the popularity of the canonical Corinthian capitals.
As noted in the introduction to this article, recent archaeological studies (specifically on the classical and medieval world) have demonstrated the use of the code-switching notion in order to interpret the influence of various contemporary cultures on constructing the identity of given societies.97 The case presented in this study can therefore be seen as yet another example for this cross-disciplinary model—apparently the first time to be applied on a southern Levantine context, while demonstrating the conscious and unconscious choices made by people in shaping their identity.98
I wish to thank the various scholars and institutions who kindly allowed me to republish some of the capitals mentioned in the text, and to Natalia Zak (Israel Antiquities Authority) and Yulia Gotlieb (Institute of Archaeology, Tel Aviv University) who prepared the illustrations for this article. I am especially indebted to Moshe Fischer for his many thoughtful comments on earlier versions of the article, which no doubt greatly improved the final result.