Hidden Mediterranean History/Histories: The Church of the Panagia tou Potamou in Kazafani (Ozanköy), Cyprus

ABSTRACT

The following article approaches a historic monument on the island of Cyprus, the sixteenth-century Panagia tou Potamou church, with an interdisciplinary methodology. An in-depth study of its history, architecture, and paintings leads to a new evaluation of the church's value for Cypriot and Mediterranean research. The church has proven to be a space for burial and private memory of a sixteenth-century semirural community, reflected in the staging of a prominent burial and the iconographic topics underlining ideas of intercession and salvation. Ultimately, this enables an enhanced appreciation of “minor monuments” in general. The other angle of approach concerns heritage questions: in precarious state for most of the twentieth century, particularly the wall paintings are in urgent need of restoration. In 2015–2017, an emergency intervention secured the most fragile parts and evaluated the state of the church, proposing future ways to ensure the survival of this monument.

Traditionally our understanding of the past, of societies and their artistic production, has been largely dependent on major events, monuments, and works of art. More recently, scholarship in the histories of art and architecture has turned towards previously neglected smaller-scale historic realities. A certain “anthropological turn” has led to a focus on minor works, sometimes fragmentary or of limited artistic excellence, recognizing them as valuable sources that provide glimpses into the local historical and cultural context of the time of their creation. Furthermore, those buildings and objects were part of often rapidly changing sociohistorical contexts throughout the centuries. In this study, the result of an interdisciplinary scientific project, one of those long-neglected monuments on the island of Cyprus is reapproached: the church of the Panagia tou Potamou in the village of Kazafani (Turkish Ozanköy)¹ on the Pentadaktylos (Turkish Beşparmak) foothills, just below the important thirteenth–fourteenth century Premonstratensian monastery of Bellapais.²

The church and its extant decorations require the following scholarly description and analysis, not only for the benefit of cultural historians of Venetian Cyprus but also because the rate of decay of the building and its murals is so rapid. This article represents a transdisciplinary study of a culturally significant monument at a given moment in time, leaves future conservators a record of what interventions were undertaken by the World Monuments Fund project, and offers guidance to local authorities about immediate rudimentary measures that can be taken to protect this almost unknown Cypriot heritage site.

The church, tucked away in a courtyard in the upper quarters of the village and almost invisible from the street, dates to the late medieval period during which it was richly embellished with now only fragmentarily preserved wall paintings. Early scholarship took notice of the church largely due to two exceptional funerary monuments: the larger one was briefly described by George Jeffery in 1907 (Jeffery 1907: 20; Jeffery 1918: 323); the other one, a small tombstone, was found by Jérôme Peristiany and Luigi Baldassare before 1925 (Baldassare 1925; Enlart 1925: 169–71). These will be revisited in the second part of this article, preceded by an analysis of the building's architecture, which is the first one ever undertaken. The third part offers an in-depth iconographic and stylistic evaluation of the remaining painted decoration, while the final section presents the results and recommendations of the emergency intervention undertaken on the murals in 2017, funded by the World Monuments Fund. This multifaceted approach enables the authors to demonstrate the role the church played in the local community during the time of its erection, and at the same time it lays out a path to reestablish consciousness of the building's potential for future community building.

The Architecture: A Showcase Example of a Minor Extra-Urban Church of the Sixteenth Century

The church is situated within an irregularly shaped courtyard, which can be accessed from the east (Fig. 1). Towards the west, the plot on which the church is built descends steeply into a small creek, which gave the church its name (Panagia tou Potamou = Our Lady of the River). Towards the south/southeast, where the village is overlooked by the Pentadaktylos/Beşparmak mountain range, the ground ascends. The core of the building is formed by a single nave space with barrel vault and eastern apse (Fig. 2). The plan is heavily irregular: due to the complicated plot disposition, the western end of the nave is wider than the eastern end. At some point during the Ottoman period a porch with a flat wooden roof was added to this core, which runs along the western and northern sides and is articulated only by means of a few low buttresses. In the same phase, the southern wall was strengthened with an additional shell of ashlars. The simplicity of this expansion suggests a lack of financial means during the time it was built; yet, it does not appear to have profoundly altered the rather austere character of the exterior. Already in its initial state, the exterior walls were plain and only interrupted by window and door openings.

Today, the eastern side is the most “decorated” part of the building (Fig. 3). The apse possesses three round arched windows next to each other: a comparatively rare feature among the minor churches of Cyprus built during the Latin period (Kaffenberger 2020: 1:113–19). The triforic apse window of Saint George, Akrotiri (Kaffenberger 2020: 2:77–78), shows one large monolithic “lintel” from which the three arches are carved—clearly a different approach to the Panagia tou Potamou. Here, the large windows, not connected to form a group window, evoke Middle Byzantine models such as for example of the Holy Apostles church in Pera Chorio (on the architecture Papacostas 1999: 2:41; Prokopiou 2006). While the older examples usually use individual small voussoirs or bricks to shape the arch, the windows of Kazafani possess individual large lintels from which the rounded upper part is carved—a technical feature that, for windows of this size, suggests a late medieval imitation of earlier models rather than the inclusion of an older apse.³ The lateral windows appear to have been partly blocked at some point, presumably to provide more surface for the application of (now lost) wall paintings inside. The apse cornice is pronounced but lacking any more elaborate molding profile. A small window slit in the gable above, perhaps once with a blind arch carved above, is flanked by the only sculptural element: two rounded corbels with a hole, certainly once serving as flagstaff holders. While all of this is hardly spectacular, it does nevertheless reveal an attempt to employ details of certain sophistication, the flagstaff holders even indicating the use of urban practices for ephemeral building decoration.

The current church portal in the south, a simple chamfered opening with segmented arch, was created together with the porch in the eighteenth or nineteenth century. Inside the porch, the three original portals are preserved: one in the west, two in the north (Fig. 4). They differ slightly in size but show identical construction principles. Each is shaped as a simple pointed arch made from large voussoirs, set into the outer shell of the wall without any molding. On the inside, the openings are framed by slightly larger rectangles, providing space for the wooden doors. The wooden lintels above these openings still show remains of the original hinges, though the doors are lost.⁴ The interior of the porch is very simple; the northern gallery roof is held by three diaphragm arches on flat, softly waved corbels. A larger pointed arch separates the northern from the western gallery, which contains one of the above-mentioned tombs (Fig. 5). The original western threshold between porch and nave is made from ashlars. Those of the northern half of the doorway are heavily abraded, indicating that this half of the door used to be opened for the passage of people more often. A possibly ancient molded pier, resembling a late-antique templon pier, is embedded into the inner half of the threshold, apparently in an attempt to visually enhance this place of transition into the sacred space of the church interior.

The interior of the nave is structured in a more complex way than the exterior (Fig. 6, see Fig. 2): the barrel vault emerges seamlessly from the thick walls, which are interrupted by deep, squat-arched recesses. They differ in size—on each side two larger ones in the west and a slim, high one in the east behind the iconostasis—and are separated by large expanses of wall. Thus, we should not speak of a blind arcade, even if the structural idea is similar. The resulting plan carries a faint memory of the typically Byzantine dome-hall church, but the seemingly random choice of dimensions for the niches, and of course the lack of a dome, betray that this might be more of a coincidence. The easternmost recesses are merged with the apse wall, so that the small prothesis niche to the north of the apse is placed fully within the recess (Fig. 7). Curiously, a smaller, pointed niche sits at a higher level on the opposite side in the southern recess. Two transversal arches underpin the barrel vault; they are aligned with the wall piers/segments between the arched recesses. A subtle difference in the sculptural treatment indicates a purposeful increase of sophistication towards the bema in the east. The western arch has a rectangular profile and rests on simple corbels, a quarter circle in the south and a heavily damaged trapezoidal one with what seems to be dogtooth molding in the north. The eastern arch shows a chamfered profile and rests on double quarter-circle corbels (Fig. 8). This distinction between single corbels for the western arch(es) and double ones for the eastern arch is a common feature throughout late churches on the island (Kaffenberger 2020: 1:120 and 276), a prominent example being the sixteenth-century rebuilding of the Panagia Chrysiotissa in Afentrika (Kaffenberger 2020: 2:28–30).

This leads us to the question of the building date. Jeffery (1907: 20, 1918: 323) suggests the fifteenth century without providing specific arguments, while R. Gunnis (1936: 260) and, much later, A. Papageorghiou (2010: 135) content themselves with generally indicating a medieval origin. Only B. Imhaus, discussing the tombstones found in the church, suggests that these were contemporaneous with the architecture and thus dates the entire church to the fourteenth century (Imhaus 2004: 1:338). This has to be rejected, however, as several arguments point towards a later date, during the Venetian period. The most important one is the distinction of the eastern and western vault corbels, which is widespread among buildings clearly attributable to the sixteenth century. Furthermore, similar wooden door hinges as those preserved in the northeastern portal appear in Venetian-period churches throughout the island. This woodwork is, however, not contemporary with the heavily damaged remains of the original iconostasis. Despite formally belonging to the group of screens employing shell decorations above the upper row of icons, which has origins in the late Venetian period, its flat rendition of the leaf and scroll ornaments rather points towards a late seventeenth-century date (see Fig. 8).⁵ This might indicate a larger remodeling phase, also including the porches and the latest parts of the painted decoration. A further chronological anomaly appears in the apse: If the windows of the Venetian-period apse were blocked to enhance the surface for paintings, when would these modifications have been executed? As nothing of these supposed paintings remains, we can only speculate that they were begun not too long after the church was finished, also in the Venetian period—just like the rest of the painted decoration, which will be discussed in detail below. To conclude, it seems probable that the single nave church was built in the late fifteenth or early sixteenth century and received an L-shaped porch in the Ottoman period, perhaps already in the late seventeenth century. However, we might speculate whether the current building was perhaps not the first on this site if we consider the testimony of the two preserved tombstones.

The Tombs: Testimony of a Noble Patron

The southern end of the western porch wing is occupied by a most unusual feature (see Fig. 5): a low, pointed arch with a depth of ca. 50 cm placed against the wall, forming an arcosolium. Certainly, it once housed the large tomb slab, which is now installed on two octagonal column fragments right in front of the niche and has the appearance of an altar mensa. The tomb slab shows a flat and somewhat basic carving of a bearded nobleman with hands folded in front of his chest (Fig. 9). As Jeffery already mentioned, the person's clothing, a “civilian costume [with] a round cap [and] a large gaberdine” (Jeffery 1918: 323), shows the monument's medieval origin.

Funerary niches with horizontal tomb slabs were introduced to the architecture of the island in the early fourteenth century, most prominently in the urban architecture of Famagusta and Nicosia. In Famagusta, they usually line the nave or aisle walls, prominent examples being the early one in Saint George Exorinos (Syriac) or those of Saints Peter and Paul (Nestorian) and Saint George of the Greeks (Greek Orthodox)—the latter testifying to an untroubled appropriation of this habit by the Greek community.⁶ Alternatively, they are placed in canopies in separate chapels, such as in the case of the Latin cathedral of Famagusta. Their appearance in porches is slightly less frequent, the most important example being the tomb of the abbess Eschive de Dampierre in the Latin church of Our Lady of Tortosa in Nicosia, placed against the eastern wall of the northern porch (most recently discussed by Olympios 2018: 230–32). In Bellapais Abbey, nearby the Panagia tou Potamou, the open porch, added onto the church in the fourteenth century, possesses four arched recesses at the lateral ends. While these are shallow and do not show any remains of tomb fixtures, and thus cannot be connected with a funerary or commemorative context with certainty, they clearly operated as visual inspiration for the Panagia tou Potamou porch niche.⁷ In its current state, the tomb niche can also be connected with burials of privileged persons, often “founders, patrons, as well as eminent ecclesiastics, monastics … and aristocrats” in arcosolia in the narthexes of Greek churches, for example in Constantinople (Marinis 2014: 74).

Here we reach a relatively common situation for scholars of late medieval Cyprus: working with material remains of an artistic practice that is connectable to various interacting cultural traditions but itself so entirely devoid of historical context that the precise interpretation has to remain somewhat speculative. Luigi Baldassare only mentions the niche and the large tombstone briefly and instead focuses on the interpretation of the second tombstone, found in two fragments in “a corner” of the church around 1925 (Fig. 10). This tombstone (last located in the village church of the Archangel Michael in the village center) is more elaborately decorated in that the central figure, a young woman in a long dress, hands folded in front of her chest, is surrounded by a polylobed frame containing a lengthy inscription. This inscription tells us in medieval French that it is the tombstone of Joanete, the niece of a certain Martin Carmoun, who died on April 22, 1348—probably a victim of the plague ravaging Europe that year.⁸ We will not repeat here the extended discussion of possible roots of the “Carmoun” family name given by Baldassare. Instead, we shall content ourselves with his conclusion that it is most likely the same family as the “Carmain” mentioned in several historic chronicles for the mid-fourteenth century (Baldassare 1925: 122–25). Those, in particular a certain Sir Juan and Sir Raffé, were knights belonging to the social group of “white Genoese”: Genoese citizens of Eastern Christian, presumably Syrian origin. While this provides a possible historic context for the second tombstone, it does not necessarily provide the same context for an early phase of the church. The rather small format and decontextualized state of the sculpted piece could well mean that it was brought here from Bellapais—a somewhat more likely context for an elaborate mid-fourteenth-century tombstone.

Considering these arguments and the difference in artistic quality, we must question a direct link between the two tombstones and thus also the identification of the nobleman depicted on the first tombstone with the uncle Martin Carmoun mentioned in the inscription. Notwithstanding, Imhaus suggests that the depicted person might indeed be of Syrian origin (Imhaus 2004: 1:129), a possibility supported by the houppelande, the garment worn by the deceased. Salzmann points out that in the late fourteenth and early fifteenth century this attire, usually connected with the social roles of civil servants and merchants, was the preferred choice for tombstones of Greeks and Syriacs, while Latins favored depictions in knightly armor (Salzmann, forthcoming: 132–35).⁹ When considering different possibilities of who might have been buried in this prominent place, Baldassare suggests that the shape of the headgear differs from that of Greek dignitaries, described as “double-arched and surmounted by a cross” (Baldassare 1925: 121).¹⁰ Yet, while his drawing shows a strange domical cap on top of a crown-like lower part, only the latter is actually visible on the tombstone. Baldassare does not err when he suggests interpreting this as some form of a toque (headgear worn by laymen throughout the medieval period). Ioanna Rapti has pointed out the similarity of the headgear worn by the Armenian soldier Philippe (died in 1351) on his tombstone in Tarsus with several other examples found in Cyprus—among which we can count Kazafani.¹¹ However, more than indicating a link to the Armenian or oriental Christian culture, this parallel indicates that the same fourteenth-century models were used for both tombstones. Furthermore, G. Meyer-Fernandez recently concludes from the evaluation of Cypriot donor portraits that even if “le vêtement d'un Chypriote peut indiquer son origine ou son ouverture aux apports extérieurs, il ne permet pas toujours de l'identifier à un Grec, à un Latin ou à chrétien d'Orient. Cette difficulté … se généralise à l'époque vénitienne” (Meyer-Fernandez 2019: 222). We can, therefore, conclude with certainty that the funerary niche and tombstone were installed to commemorate a lay patron of the church, though we cannot determine definitively his denominational or ethnic identity.

A major chronological problem is presented by the fact that the tombstone follows fourteenth-century models, while the church is datable to the period of around 1500. Additionally, the niche was apparently built in a different period than the western church wall and does not bind into the porch wall either. It is highly unlikely that a niche of this kind would have been built in the Ottoman period to house a medieval tombstone. Furthermore, it shows faint remains of painted decoration, presumably executed in the Venetian period, so its erection precedes that of the porch. It appears most likely that the niche is either the only remaining part of a predecessor to the current church or was built in this place specifically to house the tombstone. In any case, the niche would then have been encased in some form of a narthex, the predecessor to the current western porch wing. This complex building sequence underlines that the tombstone was considered a feature of importance and received a specific staging to emphasize its historicity. Such processes of displaying tradition are fairly common in Cyprus throughout the fourteenth to sixteenth centuries (Kaffenberger 2020, vol. 1, esp. chs. 6 and 7), but in this particular context the precise implications are speculative. Was the original patron's memory being maintained when the church was (re)built around 1500 and the tombstone displayed so prominently? The memory of this founder would then only have been lost after the Ottoman-period remodeling, when the importance of the niche was clearly preserved. Should we assume appropriation of a nameless fourteenth-century tombstone by a sixteenth-century patron? There is no proof for such practices in late medieval Cyprus, but the general idea of reuse of valuable elements of material culture in order to enhance prestige is certainly a common phenomenon. Or does the very moderate sculptural quality of the stone indicate the simplified imitation of fourteenth-century (originally Latin) funerary culture by a sixteenth-century patron? This interpretation might seem tempting; however, most preserved Greek tombstones from the Venetian period made use of more contemporary, “Renaissance” stylistic features.¹²

In any case, all of these possible interpretations are in keeping with other artistic phenomena of the Venetian period of the island when many of the privileged Greek families contributed to a visual culture of a “fourteenth-century revival,” including for example the copying of Gothic portals at the Greek cathedral of the Odigitria in Nicosia, in order to celebrate the island's glorious past but at the same time negotiate the special role within the Serenissima (Olympios 2019; Kaffenberger 2019a).

The Paintings: Fragmentary Splendor of a Sixteenth-Century Cycle

Originally, large parts of the interior of the church of the Virgin Mary in Kazafani were decorated with wall paintings. Only fragments have been preserved, a semblance of the once much richer appearance. The largest painted surface remains on the western wall, the southwestern vault, and around three recesses in the nave. The lower wall depicts different saints from various periods, and the bema lacks any painting at all. What follows is a description and preliminary analysis of the extant murals, starting with the southern wall (Fig. 11).

The lowest part of the wall of the southeastern recess, right in front of the iconostasis, is blank. The conserved paintings in the upper part show a complex composition in two registers (Fig. 12). The central part of the lower zone depicts St. George on horseback, moving westward, and in the act of slaying the dragon, which is now lost.¹³ The life-size figure holds a spear in his raised right hand and the reins of the white horse in his left, while a red, wind-swept cloak billows behind him. Fragments of the blessing hand of God appear in the upper right corner.¹⁴ The fragmentary inscription of George's name is visible to the right of his halo (… EOP …). was originally a long inscription of two lines with a name and an epithet, but at present it is not identifiable.¹⁵ The portraits of equestrian saints (in contrast to the frontal, standing position common in Byzantine art) is a phenomenon that became prominent in border regions of the empire and in areas under Crusader control, such as Cyprus.¹⁶ As S. Gerstel suggested, the equestrian portrayal of St. George reflects well-established Latin devotional practices, expressing cultural identity or emulation (symbiosis; Gerstel 2001: 266–80).¹⁷ The placement, format, and life-size scale of the equestrian saint provided this devotional image with disproportionate importance vis-à-vis the remaining program. Given the funerary function of the Panagia tou Potamou, it is reasonable to assume that the representation of St. George alluded not only to his miracles and apotropaic power but also to his role as intercessor (Velmans 2009: 239).

Above the equestrian portrayal of St. George, eight scenes depicting episodes from his synaxarion are enclosed within frames, including: George appearing before the Emperor Diocletian; his imprisonment (Fig. 13); George being stripped and scraped with rakes (Fig. 14); being sawn, tortured on a wheel, buried in fire, and in a cauldron of lime.¹⁸ Each scene contains small letters, but none of this text is currently legible. The scenes are detailed and narrative in nature, for example the guard sleeping outside the cell. The martyr's repeated sufferings, God's miraculous interventions on his behalf, and his own miracles performed for the benefit of others were all assurances to the donor (and the viewer) of St. George's proximity to the divine and of the effectiveness of his intercessions (Maguire 2000: 192–93). The selection of the saint and the scenes of his vita imply that the donor chose to invoke St. George's protection as well as his intercession for the afterlife.

The top center of the recess features a scene that apparently depicts the Annunciation to the Virgin (Fig. 15). The Theotokos, seated on a red throne with her right hand raised, receives the angel's visit. In front of her an amphora is placed on the floor filled with white flowers. There is no liturgical relationship between this feast's scene and St. George, but this particular integration is probably due to the important role played by St. George, and that in turn is enhanced by the presence of a Mariological scene.

The equestrian saint is flanked on either side by the hieratic depiction of two standing, front-facing saints. The saint on the left side is identified as St. Nicholas of Myra due to the episcopal attire, the physiognomy, as well as the partially preserved inscription (… NIKOΛA/O …).¹⁹ The popularity of depictions of St. Nicholas is based above all on the fact that he provided help in a variety of situations: he watched over travelers, he provided money to the needy, he expelled demons, and he helped people in trouble with the law. He is the model of a bishop saint who embodied a philanthropic role and had a capacity to act as an arbiter of social security and order (Mouriki 1993: 256).

On the other side, the female martyr St. Paraskeve is shown, clad in a dark maphorion as is typical and holding in front of her an icon of the dead Christ depicted as the Man of Sorrows (Fig. 16) (Mouriki 1993: 253–54; Connor 1999: 219, 221; Kalopissi-Verti 2012: 173–74). The inscription on the left side of her halo reads: AΓIA … ΠAPA/CKE … I. Saint Paraskeve is the patron saint of Good Friday and prominently associated with the cult of the Holy Cross.²⁰ In Cyprus, she is represented with this specific iconography in many church decorations and icons from the thirteenth century on, most dating to the post-Byzantine period.²¹ Her cult was widespread not only in Byzantium but also in southern Italy (Andronikou 2017: 11–13). Nevertheless, in Cyprus, the popularity of St. Paraskeve from the thirteenth century on acquired a local character through both the repetition of the same typical iconographic formula and also through the presence of her relics on the island.²² Her unambiguous iconography connected with the Man of Sorrows is particularly relevant to liturgical texts indicating her role on the Day of Judgment.²³ As such, her portrait fulfills the needs of the viewer in three respects. It is linked to the Last Judgment (depicted on the west wall), it is connected with the intense local cult of the cross, and it motivates personal devotion to the intercession of the saint.

Originally there were twelve rectangular frames containing busts of saints in the soffit of the recess (Fig. 17). All twelve, turned outwards in three-quarters position, present an open rotulus with one hand raised. On the east side, the figure once displayed at the bottom has entirely disappeared; the other five above it are in poor condition. The inscription behind the bottom, vanished figure is still preserved, but it is faulty: COΛAMOC. Although no crown can be seen on the youthful figure's head, it would have depicted Solomon. The two inscriptions on the left side behind the halos of the two lowest figures on the west side prove that they are prophets: O ΠROΦITIC.²⁴ The fourth prophet counted from the bottom has the inscription near his halo: MOYCIS. Like the figure two frames below him, he is individualized with a long gray beard. Finally, under the frame of the lowermost prophet on the west side, the nimbus of a front-facing, standing saint is faintly discernible, an arrangement that was most likely analogous in the east.

On the east spandrel above the recess, directly in front of the iconostasis, there are some fragmentary remnants of a frontally depicted saint (Fig. 18). There is no further information that would reveal its identification, though a gray cloth framed by a nimbus points to a headdress such as that of St. Athanasia, who was widely venerated in Cyprus.²⁵ However, due to the lack of further clear evidence, this must remain hypothetical.

Above the pier between the two recesses of the south wall, St. Mamas (the protector of shepherds) is represented frontally riding a lion while holding a lamb and a pastoral staff (Fig. 19).²⁶ The youthful and beardless saint moves east towards the apse. The Cappadocian saint was known in Cyprus from the late seventh century due to migration between Anatolia and the island (Seibt 1991), though the earliest attestation of the cult of Mamas on Cyprus is only found in the late twelfth century when Queen Tamar renovated the Georgian Monastery of Yialia, which is dedicated to him (Snelders and Immerzeel 2012–2013: 95). At the beginning of the fifteenth century, a legend circulated claiming that the Cypriot village of Morphou (Guzelyurt) possessed the body of the saint.²⁷ Cypriots “appropriated” St. Mamas by inventing a local legend for him including the element of the lion, which had not been canonical before. In iconography, the specific Cypriot variant then fuses the two earlier versions: Mamas riding a lion and Mamas the shepherd.²⁸ The high popularity of icons and mural paintings started in the Lusignan period but became especially pronounced from the sixteenth century onward (Solomidou-Ieronymidou 2012: 338; Philothéou 2012: 171; Snelders and Immerzeel 2012–2013: 95–96). An explanation for this phenomenon can be found on three fourteenth-century panels on the saint's marble sarcophagus, depicting St. Mamas mounted on a lion between St. George and St. Demetrios (Kalopissi-Verti 2012: 163). This particular setup indicates the favorite composition of the trio, which also appears on the south wall in Kazafani: beside St. Mamas, St. Demetrios is shown on horseback killing the soldier Kalojan.²⁹ The rendering of Kalojan, who, between the hooves of his horse, is pierced by the saint's spear, is the best-preserved part of the composition (Fig. 20). Like St. George, St. Demetrios moves westward, away from St. Mamas. Demetrios being one of the most popular Byzantine saints, he is depicted in many Cypriot churches from the early twelfth through the early sixteenth centuries.³⁰

A front-facing figure depicted with a tonsure is preserved on the western end of the wall, on the right spandrel of the western niche (Fig. 21). He wears a stole over his left shoulder, which suggests that he is a deacon, and in his left hand he holds a closed book. To the left, the inscription reads ὁ ἅγιος, on the right, IΩA ….³¹ Also on his orarion some letters are still readable, namely the ΓH …³² from AΓHOC AΓHOC AΓHOC, though misspelled with an H (E) instead of an I (AΓIOC). The words of the seraphic hymn heard by Isaiah in his vision of the Lord in the temple (Isa 6:2–3) introduces the consecration of the bread and wine in the Eucharist, which is performed by the deacons.³³ We see St. Ioannes Lampadistis, who is sometimes depicted tonsured and represented as a deacon.³⁴ Saint John, we must also note, was venerated as a native saint in Cyprus and is believed to have been buried in St. Herakleidios, the katholikon of the Monastery of St. John Lampadistis (Stylianou and Stylianou 1997: 292–322). He is one of the most important saints in the period of Crusader Cyprus (Weyl Carr 2009).

Above the western blind arch on the south wall, the only preserved decoration of the vault can be found. The large narrative panel depicts the Nativity of Christ combined with the Adoration of the Magi (Fig. 22). In the center of the composition the Theotokos sits enthroned in a cave, which is hardly recognizable at present. Virtually no traces of the depiction of the son on her knees have survived either. Behind them, the ox and the donkey look out from behind the manger and Joseph—turning away from them—is deep in conversation with an old man leaning on a staff. The bird depicted at Joseph's feet enhances the folk-art picturesque, narrative style of the painting as do the animals grazing and fighting on the rocky hillside behind them (Fig. 23). Three angels appear above: the one on the left bears a scroll with “Glory to God in the Highest and peace on earth to those on whom his favor rests” (Luke 2:14) clearly written on it in Greek [Δόξα ἐν ὑψίστοις Θεῷ καὶ ἐπὶ γῆς εἰρήνη ἐν ἀνθρώποις εὐδοκία]; the angels in the center and on the right announce the “good news” to the shepherds, one of whom shields his eyes from the blinding heavenly light. The largest part of the composition (in the lower right) is occupied by the three magi. The first, only his long gray hair is now visible, offers Christ a diadem, undoubtedly a Western-type crown, and, like his two companions, he is dressed in elaborate clothes and decorated shoes (it is not the conventional Persian costume). Corresponding to Western iconography, the first of the magi kneels in front of the seated Theotokos after having laid the crown on the floor.³⁵ The second, standing, magus turns to the third and points with his raised right hand to the eight-pointed star above the head of the Theotokos. In his left hand he holds a pyxis. This magus has brown hair and a beard, and he wears a magnificently ornate crown (Fig. 24). Unfortunately, the head of the third magus, who is also depicted kneeling, is lost, though it is clear that he had an elaborate headdress like a turban³⁶ and that he is holding a gift similar to the second magus. A man standing directly behind him holding the horses' reins wears elaborately decorated trousers and an extravagant fur hat (Fig. 25).³⁷ The trousers are embroidered with flowers similar to the robe of the first figure to the left of the Theotokos in Oikos 17 of the Akathistos Hymn in the Latin chapel of the monastery of St. John Lampadistis in Kalopanagiotes (ca. 1500; Bryer and Georghallides 1993: fig. 35). Identical hats are worn by some figures in the Crucifixion in the church of Panagia Podithou in Galata (1502; Stylianou and Stylianou 1997: 105, fig. 49). The figure of the groom was common in the religious art of Italy and France from the fourteenth century onward, and examples in Orthodox churches are known from St. Nicholas Orphanos in Thessaloniki and in the katholikon of Chilandar on Mount Athos (Parani 2003: 233–35, fig. 205).³⁸ As is the case elsewhere, it can be assumed that this figure depicted the donor, who wanted to be integrated into the biblical scene and so joined the worship of the magi.³⁹ Thus he forms an unusual, anachronistic extension of the iconography of the narrative scene.⁴⁰ Although the secondary scenes such as the annunciation to the shepherds and Joseph belong to the Nativity, the Theotokos is shown enthroned, rather than reclining. This is a deliberate choice meant to give concrete expression to the representative character of the scene.⁴¹

The preserved painted decoration on the northern wall is less elaborate than that of the southern wall (Fig. 26). On the soffit of the western blind arch on the north wall is a painting of an unidentified male saint (Fig. 27). He stands, front facing, with short brown hair and a trimmed brown beard and tonsure. Above the blue tunic he wears a brown pallium. In the left hand he holds a book, the cover decorated with gems, and his right hand is raised in a gesture of blessing. There is no evidence that would help to further identify this saint, however. In the eastern part of the recess, to the right of the door, there is a fragmented bishop figure with a closed book in his left hand (Fig. 28), but since neither his head nor any inscription has survived, the identification remains uncertain.

Plaster overlap indicates that one of the figures on the upper part of the pilaster supporting the west and central arches of the north wall was painted at a later stage, probably replacing an earlier figure (Fig. 29). Saint Spyridon, with an ornate frame and recognizable by his basket hat, was a native of Cyprus (ca. 270–348) and one of the early saints to have a local cult.⁴² In this composition, he holds a document pertaining to, as the inscription reads, the “oneness of God and the divinity of Christ.” Next to him stands St. Nicholas (part of the inscription on the right side of his halo is still discernible: NIKOΛA), who holds a book in one hand and blesses with the other. He has a gray beard and hair and wears an omophorion (pallium), sticharion (long tunic), epitrachelion (stole worn round the neck), and epigonation (rhombic cloth hung over the girdle). A stylistic comparison with the St. Spyridon to his left suggests that St. Nicholas belongs to the earlier scheme.

In the central northern recess, the figures of St. Constantine and his mother, St. Helena, can be discerned, each holding a piece of the True Cross (Fig. 30). From the frontally facing figure of Constantine on the eastern side of the recess, only his bright loros, his halo, and fragments of his crown can be identified. With his left hand, he holds the shaft of the cross beside him. Helena stands on the corresponding soffit on eye-level, meaning that she would have been visible to an entering beholder from the main doorway, and in a way visually expands the iconostasis into the recess. She too is dressed in imperial garb, crowned, and with a nimbus. With her right hand, she also presents a large cross. The holy pair is often represented in Byzantine art (Teteriatnikov 1995), and in Cyprus they (especially Helena) assumed particular prominence under the Lusignans in connection with political propaganda.⁴³ This could be an explanation for why Helena was depicted without Constantine in several churches from the Lusignan period.⁴⁴ Similar to these examples, it seems likely that Helena was singled out in Kazafani for her role in the legend of Cyprus's relics of the True Cross (Connor 1999: 222–24). The cult of the relic of the True Cross in Cyprus is one of the earliest outside of Constantinople (Weyl Carr 2009: 475–76, with bibliography).

On the wall directly east of the niche, fragments of two female figures with long, red clothes and halos are still discernible. The two women embrace each other in the scene of the Visitation, which is very rarely depicted in isolation. As such, it indicates a former larger cycle that was created independent from the Nativity, one that clearly adopted different dimensions for the depicted figures.

The main focal point of the painted decoration used to be on the western wall (Fig. 31). Here, a monumental Last Judgment composition was displayed of which significant fragments remain. In the churches of Cyprus, especially during the era of Latin rule, the Last Judgment was a preferred motif (Weyl Carr 2020). The elements of these scenes are adapted from and inspired by Byzantine examples.⁴⁵ Gunnis (1936: 260) described the Last Judgment on the west wall as the best-preserved painting in Kazafani, and so we must conclude that the composition is certainly much more damaged today than in the 1930s when he inspected it (Fig. 32). Though the image of Christ as judge in the center, below the window, has been lost, there are still many characteristics visible, such as the Psychostasia (weighing of the souls) at the bottom center. Directly above it is the Hetoimasia (prepared throne), which is guarded by two seraphim. On the empty throne a dove, a book, and the instruments of the Passion of Christ can be seen. Adam and Eve—on both sides of the judgment scale—are in proskynesis, adoring the throne.

Horizontally, the wall is divided by three red lines into four zones. A sun is painted directly above the round window at the zenith of the western wall, and other remnants make it clear that there were stars here too, flanked by two Old Testament prophets (Fig. 33). The one on the north side turns towards the center in proskynesis, but only fragments of his royal dress and crown are visible. He holds an open rotulus whose inscriptions cannot be decoded any more. This figure represents either David or Solomon. The prophet on the south side holds an open rotulus too and performs a gesture of blessing with his right hand. He does not look up, rather, he looks down to an eight-rayed golden star that is placed between him and the rotulus. A red line separates this segment from the large composition below. In the narrow strip below the prophets, two angels appear in half length before a dark blue background to the left of the oculus; the northern side of this strip is destroyed but presumably showed a similar composition.

The next register is reserved for the Heavenly Judgment, though this area of the composition has been greatly damaged. Nonetheless, seated apostles can still be seen at both ends, individualized and with raised hands. Together, with the central (lost) group—probably consisting of Christ, the Theotokos, and John—they formed the Great Deesis. The half-figured righteous, standing at the right hand of Christ, are divided into four groups and organized in horizontal rows (Fig. 34). Closely crowded together, they turn towards the center with their hands raised. The four groups can be identified on the basis of their clothing: at the top are the martyrs and saints (St. Paraskeve can be seen on the left margin holding an icon of Christ in her hands, and in front of her stands a female martyr in royal dress); among the prophets below, at least one king and Daniel can be identified; in the register below are the hermits, who in turn are individualized and wear different headgear; and the lowest zone is occupied by bishops. In front of the bishop and the hermits, a vertical line is drawn which separates them from the central composition of the Psychostasia (Fig. 35).

The damned are illustrated on Christ's left in a field comprised of two sections (Fig. 36). The upper section is wider, featuring Hades with a central monstrous figure that represents evil and with many hands captures the damned in hell. The latter appear as a group of heads, squeezed into a deep dish, ready to be consumed by the devilish creature above. From the twelfth century onwards, individual punishments for specific acts, long known from texts, enlarged the iconographic repertoire of the Last Judgment (Mouriki 1976: 161–62; Garidis 1982; Gerstel 2002: 207–11). In the lowermost zone, corresponding to the bishops on the opposite side, the individual torments are shown (Fig. 37). Six naked sinners are moving towards the center, red flames licking up between them. There are no labels, but the implements and symbols of their torments make the identities of some sinners clear. The first is bitten by a snake in the genital area and therefore probably a whore, and the man next to her can be recognized as a male whore. The third sinner is “he who ploughs over the boundary line,” because a plough is attached to him. A millstone hangs from the neck of the next sinner, who is depicted upside down, representing the dishonest miller, and the last two are fallible farmers, recognized through attributes like the scale and the barrel. It is striking that the sins mainly affect an agricultural community specifically. These pictures were clearly calibrated to local historical and social conditions in rural areas (Gerstel 2002: 216–17). The individual punishments are prominently displayed at eye level, so those leaving the church would see them and be reminded of the results of leading a sinful life. Such an intent to convict, or at least to intimidate and prevent, became known in Cypriot iconography only in the late Byzantine period (Kalopissi-Verti 2012: 146–47). Interestingly, a great number of individual punishments were also depicted in countryside churches of Venetian Crete (Tsamakda 2012: 205–8; Lymberopoulou 2020; Weyl Carr 2020: 375–76). The iconographic similarities are perhaps not surprising as both islands were part of the Serenissima at the time (Kalopissi-Verti 2012: 146–48) and shared analogous rural social environments.

Additionally, two front-facing representations of saints flank the door of the west wall. To the south there is a repeated rendering of St. Paraskeve (Fig. 38), the lower part of which is lost. The saint is dressed in a dark-colored maphorion, holding an icon of Christ in both hands. The picture, which is clearly recognizable as Imago Pietatis, has no frame.⁴⁶ As mentioned above, since the saint is the personification of Good Friday, she is particularly associated with the Man of Sorrows. The relevant liturgical texts refer both to this as well as to her important role in the Last Judgment, and so her presence on the west wall below the Last Judgment is an intentional extension of the theme.

The archangel depicted on the northern pendant is preserved only up to the chest (Fig. 39). His wings are still recognizable behind him, he is dressed in the imperial loros, probably held a staff in his right hand, and in the position of his left hand a small head surrounded by a cross nimbus can be seen. This is, in analogy to St. Paraskeve, an unframed, icon-like representation of Christ, rarely attested as part of an archangel's iconography.⁴⁷ Although there is no longer an inscription, the iconography and the location suggest that the archangel is Michael. Michael's cult was widespread in the Orthodox world (Pallas 1978; Martin-Hisard 1994; Kalopissi-Verti 2012: 168–70) as he fulfilled several functions. Like all other heavenly powers, Michael is first and foremost a minister of God and a messenger of his will on earth. He is considered a protector of the faithful against evil powers and a guardian of churches. This is of course the main reason why he is often depicted in the narthex and close to or facing the entrance doors of churches (Kalopissi-Verti 2012: 169–70). As Michael is also associated with the cult of the dead and with the hymn of the dying, he is regarded as the companion of souls. This is the explanation for his constant presence in scenes of the Last Judgment (Pallas 1978: 45; Kalopissi-Verti 2012: 170, with the relevant texts). Kalopissi-Verti pointed out that in all of these capacities, Michael takes a meaningful place in the narthex at Asinou (Kalopissi-Verti 2012: 170), and we suggest that the same conclusion may apply to his depiction in Kazafani. He is the protector of the church, and as an intercessor for humankind and a companion of the souls, he is a part of the Last Judgment displayed above him. The icon of Christ in his left hand parallels the icon in the hands of St. Paraskeve. The iteration of St. Paraskeve is related to her specific role in the context of the Last Judgment, and so is her counterpart, the archangel Michael. Our assumption that these two images were conceived as a pair is supported by the fact that they carry the same type of icon.

The Paintings: Stylistic and Technical Remarks

Only a few christological scenes have remained in the church, all of them from the childhood of Christ. The Annunciation is small and inconspicuously located in the southwestern recess, in the midst of the biographical episodes of St. George. The Visitation is placed opposite it, on the north wall, and the detailed Nativity featuring the Adoration of the Magi is on the southwestern arch. It is clear from a stylistic point of view, as well as from techniques adopted, that several different painters' workshops were involved over the years. The Annunciation and the Visitation probably belonged to the first phase, as do most representations of the saints on both walls (St. Nicholas on the north wall belongs to a later period; only St. Spyridon on the north wall was, in all probability, painted during Ottoman times). Parallels with other Cypriot churches point to a date of around 1500, indicating that Frankish impact was still strong. The artists achieved contrasts in light areas with thick, dark brushstrokes, white dots highlight jewelry and ornaments, corporeality does not occur, yellow halos are bordered with a black and white line, and figures are flat and schematic. To this group belong: Constantine and Helena, the Visitation, the uncertain saint and the fragmentary bishops in the northwestern recess. A second artist had a more advanced understanding of the workings of light and shade in the drapery as well as on the faces, as can be seen for example in the portrait of St. Paraskeve in the southeastern recess. This artist also encircled the halos with white dots. All representations on the south wall belong to this group. Distinguishing technical characteristics that remain visible are the rather thick (5–15 mm) whitish rough plaster, and the use of “cord-snapping” (on wet plaster) to trace the main construction lines. The Nativity was therefore done after the paintings on the lower wall, which is clearly indicated by the overlap of its plaster onto the panels with saints. While the saints exhibit a rather stiff formality, the artist of the Nativity succeeds in creating a dynamic, differentiated, multi-figure composition with differences in size, articulated attitudes, and lighting effects. The composition is a good example of the consistency with which the imagery adheres to Byzantine tradition. Enriched with new iconographic components, the painting became increasingly narrative-driven and more diversified, exhibiting a variety of stylistic currents.⁴⁸ So, while the artist followed Byzantine iconography, he added significant details that reveal the knowledge of Venetian artistic standards. Technically we may observe the presence of a reddish rough plaster that was used to even the stone masonry, elaborate preparatory drawings (red and yellow lines, which were then colored by using lime paint), and compass incisions for the nimbuses, which were not used on other paintings. The Last Judgment composition and the two saints below it on the west wall were executed by an artist or by a workshop that clearly did not work in other parts of the church. Jeffery suggested a date in the fifteenth century (Jeffery 1907: 20; Jeffery 1918: 323), but stylistic and iconographical parallels indicate a sixteenth-century date, probably even somewhat later than the scene of the Nativity and the Adoration of the Magi.

The saints, who all have a strong local cult, contribute to the devotional and eschatological character that is prominently evoked by the large, detailed Last Judgment scene. These images indicate a strong relationship to the respective individual donor. It is almost certain then that the burial/tombstone in the porch may be situated within the context of one of the more prominent painting phases. Perhaps this was the St. George in the southern recess, or the panel of the Nativity with its strong “intercessive” character, or the monumental Last Judgment on the western wall with its message of eternal salvation for the righteous. The other phases would have been commissioned by other individuals whose burials, in or around the church perhaps, have not survived the course of time, or by smaller groups from the community. In any case, all these activities have to be dated to the Venetian period, confirming a late date also for the porch burial and presumably the tombstone. It appears that the church had become a site to which powers of facilitated access to salvation where attributed. The purposeful staging of the burial and the practice of embellishing the niches in the nave underline the patrons' wish to create a somewhat historic setting, one that resembles two centuries older, urban settings. The Panagia tou Potamou must have thus played a role not only in the context of private memory and care for the afterlife but also for establishing, demonstrating, and authenticating social status within the complex, multilayered society of late medieval Cyprus.

The Panagia tou Potamou Today: The Emergency Intervention to Protect the Wall Art 2015–2017

The current, heavily damaged state of these important paintings necessitated an urgent intervention. From interviews with the villagers we learned that the structural stability of the church had been in question in the 1920s and 1930s and that it was already partly abandoned then (used only at Christmas and Easter). According to the Annual Report of the Department of Antiquities, the last restoration—emergency works on the “verandah” (porch) in the north—was carried out in 1973 (Annual Report of the Director of the Department of Antiquities 1973–1974: 17). After the events of 1974, a local member of the village community, Ilker Nevzat, voluntarily took on responsibility for the structure. He was assisted by artist Emin Çizinel, in an enhancement intervention, involving the removal of a huge quantity of trash, the reconstruction of the collapsed roof of the porch in accordance with the original construction technique (wooden beams and cane-mesh mats), and the partial cleaning of the interior murals, which took about 18 months. In the intervening years, however, plant roots penetrated the decayed masonry and are now found in fissures and gaps behind detached painting plasters (Fig. 40). This, of course, is of grave concern for the well-being of the art. Additionally, the soil accumulation in the southeastern enclosure now reaches to about half of the height of the interior walls, and thus moisture problems deriving from this persistent situation are signaled by the total loss of painted plasters on the south wall up to a height of about 1.5 m and by active decay of the remaining painted surfaces in this area. Inside, severe structural problems are indicated by a complex network of pronounced cracks. Whether movement is ongoing or whether it was provoked by a single event (for example an earthquake) and has since stabilized, is unknown. Cracks in grayish lime/river-sand plaster used during the 1930s–1940s intervention suggest that the building has continued to move since. There is a pronounced longitudinal fissure that runs all along the climax of the barrel vault and has caused the displacement of one row of stone ashlars.

Most of the paintings are obscured by a heavy, black soot deposit. This situation is very common in Greek Orthodox churches due to the extensive use of candles and incense during liturgies, though the deposit was apparently thinned during the partial cleaning of the paintings in the 1980s. During an interview in July 2017, the artist Emin Çizinel explained how he used cotton wool soaked with natural turpentine, sometimes adding some poppy oil, for the removal of soot deposits from the lime-based wall paintings. It is likely that the main outcome of the treatment was an optical one, due to the permanent wet effect produced by the oily substances that were employed. There are also pronounced plaster detachments especially along the edges of fragments and in the vicinity of structural cracks (Fig. 41). The deformation, lifting and cracking of painted plasters, caused by the compaction of the masonry has created extremely fragile areas. Especially dramatic is the situation of the paintings on the west wall, where the upper left part is completely separated from the stone masonry, forming a suspended slab (Fig. 42). Both the paint-receiving plaster and the paint layer are generally strong. Considering that all paintings were executed with lime-milk as additional binder, this seems to indicate a sort of lime-fresco technique that formed “one body” between the paint layer and the lime-rich plaster. Only in a few areas is the lime-color flaking and/or showing lack of adhesion. In a small area on the right side of the central part of the north wall the phenomenon is more pronounced and clearly associated with crystallization of soluble salts. Green algae present in some areas of the vault and the above-mentioned roots seem to indicate that there are active moisture problems. Numerous areas with deformed painted plaster (blisters, bulges) cannot be flattened; neither can accumulations of dust and debris in the gaps behind the detached plasters be extracted. Tests were made with an acrylic resin emulsion (Primal B-60A) to bind the loose particles behind the detached plasters by first removing the dust and debris from the open edges, then pre-wetting them before injecting Primal B-60A. The reinforcement of the open edges and holes with lime plaster (lime putty / marble dust / fine brown quartz sand) was then attempted by pressing the mortar as much as possible below the lifted plaster. A hydraulic grout (Ledan TB1) was injected, then gently pressed to slightly reduce deformations (Fig. 43).⁴⁹ Where necessary, the consolidation of the paint layer was carried out by pre-wetting the area with water and alcohol (3:1), applying Primal B-60A diluted with water at 1:3 (v/v), then Japanese tissue with water, and by pressing with wet cotton to compact the surface and absorb excess resin before final rinsing. First attempts to remove the soot deposit with water-based slightly alkaline cleaning agents were not successful because of the water-repellence of the painted surfaces (following the intervention in the 1980s with poppy oil and turpentine). The use of a commercial nonpolar cleaning gel has provided promising results, but the method needs further studies before widespread application (Fig. 44).⁵⁰

Turning to the future, the following recommendations seem apt to help preserve the decorated interior. While the present conservation campaign was successful in stabilizing all paintings in the church, it will be important to improve the poor condition of the building that houses them. In order to protect the paintings and eliminate active sources of decay, the removal of weeds and the filling of all open gaps in the masonry is necessary, especially on the roof.⁵¹ The soil must be removed from the southern part of the enclosure and, if possible, followed up with the construction of a ventilated trench at the bottom of the walls. Looking ahead, the building must be properly surveyed, and monitoring must begin to understand if movements are still happening as this is central to the well-being of the murals inside. Lastly, attention must be paid to the carved wooden polychrome iconostasis and the painted crucifix above it, which are in very poor condition and in need of urgent conservation.

Concluding Comments

The results of the scientific investigation have made evi- dent that such laborious measures are more than worthwhile, even if the church might at first appear simple and unassuming. Built during the Venetian period, its architecture is entirely in keeping with a number of churches of the time around 1500. The building has served a com-paratively rich (semi)rural community as a preferred place of burial and personal memory as well as to demonstrate and authenticate social status. The presence of multiple phases of painted decoration and the unusual arcosolium in the (later) porch, as well as the tombstones, are a testimony to this. The painted decoration has proven to be unusual in a number of respects. First, in each of the decoration phases, which happened in rapid succession in the first half of the sixteenth century, only a limited part of wall surface was treated. Many iconographic choices underline the idea of intercession (the strong focus on the Virgin in the Nativity and the prominence of interceding saints) and salvation (St. Paraskeve holding an icon of the Man of Sorrows and the Last Judgment scene). We can thus suggest that the paintings were executed thanks to private donations, likely connected to testaments of the deceased buried in or around the church. Second, the iconographic peculiarities of the Last Judgment surpass the local Cypriot context and suggest parallels to contemporary Cretan compositions. This testifies to the opportunities of exchange and contact that rural communities in the eastern Mediterranean had (be these through the transmission of drawings or artists'/patrons' mobility). While the ethnic and confessional identity of the patron connected to the tombstone is unresolved, it has become clear that he was a typical member of sixteenth-century society in his aspiration, common to many wealthy (Greek) families across Cyprus, to assume elements of fourteenth-century Latin visual culture by material appropriation or aesthetic imitation.

The emergency intervention of the years 2015–2017 aimed to secure the most endangered parts of the painted decoration as well as provide strategies for future maintenance and well-being of the church. Indeed, despite the significant care the building received since 1974 by members of the village community, it is nevertheless in a precarious state. The rapid loss of painted plaster has been stopped for now, but the structural problems that accelerated the degradation will need to be addressed in the near future. The close examination and careful preservation of the Panagia to Potamou will ultimately serve as a case study for the importance of such minor monuments in unraveling the oftentimes complex late medieval history of the island. It will also act as a model for future local and international interventions to protect heritage even while new, “modern” histories are being written in the eastern Mediterranean.

Notes

References