A Blocked-Out Capital from Berenike (Egyptian Red Sea Coast)

3.1 The Discovered Fragment of the Capital

The fragment that was discovered is a product of a vertical split of a complete capital and accounts for less than a half of the whole. Although the top and the bottom sides of the capital are too weathered to reveal any tool traces, the break clearly maintains the stone layering direction, proving the unintended character of the split. The fragment is well preserved, with clearly eroded top layers and chipped corners. The stone has been finely prepared, with gentle chisel marks visible on the curving surfaces. There are no traces of whitening or polychrome on the original stone surface.

The fragment of the capital has pseudo-Corinthian features and is highly geometrized (Figure 5). Based on measurements of the preserved substance, the dimensions of the whole capital are estimated to have been 31.0 cm in height, 42.5 cm in lower diameter, and 94.0 cm in maximum upper diagonal. The object was sculpted from anhydritic gypsum^[3] as a single-element capital. In its upper part, it has volutes reduced to non-sculptured, elongated corners with slanted side-faces tapering towards the top, with their bottom face sloping towards the shaft’s centre. Fleurons are reduced to cylindrical protrusions. The lower section of the capital’s kalathos has a diameter of 48.0 cm (Figures 6 and 7) and underneath it are two ring-like cut outs or steps, each of sequentially decreased diameter and height (respectively, 1.5 and 3.7 cm for the top one and 1.5 and 4.5 cm for the bottom one).

Figure 5

Blocked-out capital from Berenike trench BE19-122 photograph (not orthophoto view), scalebar measures 50.0 cm (photography: M. Bergmann).

Figure 6

Blocked-out capital from Berenike trench BE19-122, artistic reconstruction (render: S. Popławski).

Figure 7

Blocked-out capital from Berenike trench BE19-122: left, reconstruction; right, existing state of preservation (F. Mi, S. Popławski).

3.2 Blocked-Out Capitals in the Ancient World

Blocked-out capitals are a particular type of architectural decoration characterized by a simplified form. Their shape is based on Corinthian or Ionian capitals and achieved by replacing floral ornamentation with geometrical forms. The concepts of simplification might have been inspired by capitals in their roughed-out state (see Brzozowska-Jawornicka, 2019; Czerner, 2009; Grawehr, 2017; Grawehr & Brzozowska-Jawornicka, 2020; Laroche-Traunecker, 2000; McKenzie, 1990, 1996; Patrich, 1996; Pensabene, 1993, 2018). They were first discovered in the areas inhabited by the Nabataeans, where they may be found in the stone monuments of Petra (Jordan) and Bosra (south Syria). This was reflected in their original designation as “Nabataean capitals,” but subsequent findings in Egypt, Israel, and Cyprus have led researchers to associate their forms with the area of influence of Ptolemaic Egypt, particularly with Alexandria (Pensabene, 1993, pp. 114, Tav. 44–45; Sabottka, 1983, pp. 195–203).

This stylization was initially believed to be the result of unfinished stonework, as was the case with the capitals from the facade of the Temple of Augustus at Philae. Borchardt (1903, pp. 73, 83–84) explained their form by the difficulty of processing a very hard stone, diorite. Similar examples occurred in Mons Porphyrites and Mons Claudianus (both in Egypt), where granite details were produced (Kraus, Röder, & Müller-Wiener, 1967, pp. 108–205; Maxfield & Peacock, 2001, pp. 36–39). At the moment, the simplification tendency is widely noticed in soft and easily weathering material as well. Researchers justify it, among other things, by a small and fragile detail’s limited resistance to external conditions (Brzozowska, 2016, p. 63; Grawehr, 2017, pp. 107–108).

The intentionality of stylizing an architectural decoration in this way is not questioned anymore. Based on similar proportions of subsequent capitals and their elements, three basic types of blocked-out capitals were differentiated in Petra, along with their biomorphic equivalents (Figure 8). These are Nabatean Types 1, 2, and 3, which correspond to high floral capital, low floral capital, and Alexandrian Corinthian type IV capital, respectively (McKenzie, 1990, pp. 116–117; Patrich, 1996, pp. 199–201). The purpose for styling the capitals discovered in Marina el-Alamein, on the other hand, is proven by the form given to them. They were fashioned by flattening the disks and removing a part of the stone required for carving a fully decorative volute, rendering them unfit for the production of a “Normal” Corinthian capital (Czerner, 2012, pp. 46–47).

Figure 8

Evolution of capitals forms. “Normal” Corinthian, Alexandrian, Floral, and Nabatean Types (drawing: U. Kraśniewska, based on McKenzie, 1990, p. 190; 2007, p. 85).

It may be tempting to look for signs of simplification in undoubtedly unfinished capitals, such as those discovered in Dendera, Kom Ombo, or Cherchell (Figure 9; Castel, Daumas, & Golvin, 1984, pp. 21–36; Laroche-Traunecker, 2000, p. 209; Pensabene, 1982, p. 59, Figure 168). Instead of the intended geometrical shape of blocked-out capitals, they are characterized by sections of completed ornamentation along with some portions left unfinished, most likely as a consequence of the unintentional interruption of work. Because of this apparent similarity, it is reasonable to assume that the source of inspiration resides in the particular stages of the capital formation process.

Figure 9

Selected unfinished capitals: (a) Kom Ombo, (b) Dendera, and (c) Cherchell (drawing: S. Popławski, based on the studies by Laroche-Traunecker, 2000, p. 210; Pensabene, 1982, p. 59, Tav.56, 168b).

3.3 Architectural Analysis of the Capital from Berenike

The capital fragment discovered in the 2019 season demonstrates clear characteristics of simplification. It is not the only architectural artefact of this kind found at the site; in 1995, another capital with similar features was unearthed in the site’s northeastern part. The capital was excavated in a Late Antique structure (BE95-4) around 150 m from the Isis Temple (Figure 2) and had been probably repurposed as a statue base. It was photographed and described in a report from the excavation works (Sidebotham, 1996, pp. 61, 71) as measuring c. 54.0 cm by c. 54.0 cm with a large circular depression of c. 32 cm in diameter curved in its top section. It is possible to deduce from the published image (Figure 10) that the forms of both capitals are similar. Their common features may be seen in the simplified volutes and fleurons. Although the length of the upper diagonal, calculated at c. 80.0 cm, indicates that the 1995 capital is smaller, it might be due to erosion and weathering of the original stone surface. Despite the simple and geometrized form of both capitals, much can be inferred about the architectural decoration that influenced the shape.

Figure 10

Blocked-out capital from Berenike trench BE95-4, scalebar measures 50.0 cm (photography: S. E. Sidebotham).

The two bottom rings are the typical elements of the capital discovered in Berenike in 2019. They would have been used to create two rows of acanthus leaves in the process of making a Normal Corinthian capital. Identical rings can be seen on the incomplete capital from Cherchell (Figure 9c; Pensabene, 1982, p. 59, Figure 168), which was most likely in a half-finished state due to a change in its intended function already during carving. According to Pensabene, it was initially intended to be a full capital, but shortly before its completion, it was changed to a semi-column crowning. As a result, a comprehensive study of the technique for carving acanthus leaves is now available. It is reasonable to assume that the ring’s lower surface, prior to further processing, corresponded to the bottom ends of downward-facing leaves. If the same approach had been applied to the capital from Berenike, it would have been impossible to achieve fully plastic, sculptured leaves of a Normal Corinthian capital due to the low and shallow placement of those rings.

For this analysis, it is necessary to think about how an Alexandrian capital was constructed (Czerner, 2012, p. 43). According to its reconstruction, acanthus leaves on Corinthian capitals typical of Alexandria and the surrounding region (Figure 8) were carved from a single ring that was shared by both rows. This was achievable due to the spatial separation of the leaves and the lack of overlap between neighbouring wreaths. As seen at the site of Marina el-Alamein, this method of forming details became the starting point for styling separate discs, with their amount matching the intended number of acanthus leaf wreaths. Each disc might then be carved to have properly separated leaves. Unfinished capitals carved in this way were discovered among the architectural artefacts of the Roman nymphaeum in Dendera (Castel et al., 1984, pp. 21–36; Pensabene, 1993, pp. 220–226, Tav. 31) as well as in Kom Ombo (Figure 9a and b) (Laroche-Traunecker, 2000, pp. 209–211; Pensabene, 1993, p. 293, Tav. 38; 2018, p. 420).^[4]

The analysed capital has no astragal, which in a classical form serves to visually separate the capital and the shaft of the column. The lowermost ring, which was the foundation for carving leaves, and a scarcity of material surplus in the bottom section hindered future stages of shaping the column and rendered obtaining a Normal Corinthian capital impossible. At this point, it is worth mentioning capitals constructed from two distinct pieces, the top of which generally corresponds to the volutes and the bottom to the acanthus leaf wreaths. A less-common separation occurs between successive rows of acanthus leaves, as seen, for example, in the capital of the Alexandrian Pompey’s Pillar from the Diocletian’s era (Pensabene, 1993, p. 39, Tav. 7; Sabottka, 2008, pp. 288–298). The application of this technique to the Berenike capital would have resulted in its fragmentation into disproportionate components and would have made carving and assembly more difficult. Although it was also feasible to carve the astragal directly in the column’s shaft, the transition from the shaft to the capital was most likely denoted by a slight overhang of the latter (McKenzie, 2007, p. 86, Figure 128, pp. 166–167, Figure 286).

Elongated corners are another characteristic feature of the Berenike capital. This kind of shape is typical of one of the steps in the creation of a Corinthian capital’s volute and abacus. However, this similarity does not entirely explain the form used in this particular case. Because the corners are so elongated and tapered towards the end that forming a volute would have required removing a significant amount of excess material. Such elongation might indicate that the sculptor intended to make an abacus that would extend outside the volutes, with its projection in the shape of a square with concave sides. Corners fashioned in this manner are even referred to as “horns” (Grawehr, 2017, p. 105; Patrich, 1996, pp. 199–201); however, this does not seem to properly reflect their character.

Halfway between the corners, as far as they have been preserved, are fleurons reduced to the form of cylinders. They might be used in preparing the remaining material for carving floral ornamentation known from the classical Corinthian order. Their shape was commonly used in the ornament carving process.

Knowing the diameter of the capital allows for estimating the probable size of the column on which it was originally installed. According to Vitruvius (Vitr. 3.3.10), the height of the column shaft in the Corinthian order is eight to ten times its diameter depending on the size of the intercolumniation. The application of these proportions to the pseudo-Corinthian order with simplified capitals was confirmed in Marina el-Alamein, where the column shaft height amounted to 17 modules or 8.5 times the diameter (Czerner, 2009, p. 25). In the case of Berenike, the column shaft might therefore have a height of between 3.20 and 4.00 m, which together with the capital would result in an estimated size of 3.50–4.30 m.

3.4 The Capital from Berenike in the Context of Blocked-Out Capitals

The discovery of blocked-out capitals in this area is hardly a surprise, given the presence of Ptolemaic – and subsequently Roman – influences are over both southern Egypt and Berenike itself. In the context of the object in question, the analysis of the capitals with distinctive two lower rings and elongated corners is crucial.

A resemblance to the Berenike capital may be observed in the details from Karnak, Medinet Habu, and Kom Ombo, among others, due to the appearance of the simplified volutes. For instance, two capitals with prominent corners and rectangular protrusions, as well as column drums measuring 55.0 cm in diameter each, were discovered in the peristyle of the Roman baths in Karnak (Figure 11a). They most likely formed only the upper part of the capitals. Plaster remnants with traces of red and yellow pigment were found on their surface (El-Masekh, Fournet, Piraud-Fournet, & Boraik, 2017, p. 224). According to the discoverers, elements of the capitals had been reused, and thus did not allow for accurate dating.

Figure 11

Selected blocked-out capitals: (a) Karnak; (b and c) Medinet Habu; (d and e) Kom Ombo; (f) Didymoi; (g and h) Karanis; (i) Coptos; (j) Kourion (drawing: U. Kraśniewska, S. Popławski, based on the studies by Boak & Peterson, 1931, Figure 77; Brun et al., 2011, pp. 33–47; El-Masekh et al. 2017, p. 224; El-Nassery et al., 1976, pp. 239–271, Pl. LXVIII: 47; Grawehr & Brzozowska-Jawornicka, 2020, pp. 181–200, Figure 2; Hölscher, 1954, p. 190. 46 a,b; Laroche-Traunecker, 2000, p. 210).

Two other capitals of similar forms were reported nearby, in the area of the Medinet Habu complex, in a small church located outside the Eastern Gate (Figure 11b and c; Hölscher, 1954, pp. 55–56, Pl. 46 a,b). Based on the released photograph, each of them has a diameter of c. 40.0 cm. The first is characterized by a significant simplification and the absence of the sculptural ornament. The other is slenderer and decorated with a delicate floral detail. Both consist of only a few basic elements: bulky corners, protrusions between them, and plain cylindrical shafts.

Many capitals with significantly simplified volutes were found in Kom Ombo (Figure 11d and e) (Laroche-Traunecker, 2000, pp. 209, 213; Pensabene, 1993, pp. 601–607, Tav. 73; 2018, pp. 407–422). Despite the fact that they were discovered at the same site, their styling differs. Their corners vary in shape, some being simple and others being more complex. This might be attributed to the varied levels of development of what are ultimately the same capitals (Laroche-Traunecker, 2000, p. 209). It seems considerably more plausible; however, these are the complete blocked-out forms.

Finally, a capital of comparable characteristic was unearthed in the north-eastern corner of the Roman Fort at Didymoi (Brun et al., 2011, pp. 33–47). It had been re-used in the first quarter of the third century AD. The capital is smaller (diameter c. 23.0 cm) and simplified to a roughed-out form consisting of only fundamental elements (Figure 11f).

It is important to note that all of the described capitals from southern Egypt are characterized by a simple form and extensive reduction of classical elements. In several cases, the capitals are decorated with colour or florals, but without additional sculpting of the simplified form. Unlike the Nabatean and Marina el-Alamein capitals, their stylization is not based on further interpretation of the classical elements. It is worth mentioning that the capitals from Douch are an exception – despite their South Egyptian provenience, their decoration is more elaborated and closer to those found in Nabatea and el-Alamein (Laroche-Traunecker, 2020, pp. 95–112; Pensabene, 2018, pp. 407–420).

Capitals with acanthus leaf wreaths characteristically reduced to two ring-like steps were discovered in Karanis. They were found as crownings of pilasters surrounding niches in Late Roman houses (Figure 11g and h; Boak & Peterson, 1931, pp. 57, 60, Figure 77; El-Nassery, Wagner, & Castel, 1976, p. Pl. XLVIII). Considering their relatively small size, they seem to be more sculptural and decorative than architectural elements. Although they are made of mud brick, smaller details are shaped in plaster. This demonstrates intentional simplification as their form was achieved in part with materials other than stone. These capitals are significantly less prominent, with their top part dominating the bottom. It might have been caused by the dimensions of brick used for their construction, as it was most likely the case in house C119 (Gazda, 2004, p. 30, Figure 54).

A capital with corners and ring-like steps most similar to the fragment found in Berenike was discovered in Coptos (Figure 11i). It was most likely built during the extensive construction phase during the first century AD and then reused in the construction of a later building located in the northern part of the site (L. Pantalacci, personal communication). Despite being rather bulky and of higher proportions, it contains only basic elements. The presence of masonry tool marks on its surface and an elongated groove at the bottom of the base diameter might indicate that it was not completed. These defects might have later been concealed with a layer of plaster, but this cannot be confirmed due to a lack of evidence. The stepped arrangement of the rings in its bottom section bears a striking resemblance to the capital from Berenike. In this case, however, the rings appear slightly further up, potentially allowing more of their surface to be used for carving the acanthus leaves. There are simplified fleurons in the middle of the distance between the corners. The corners themselves are elevated and narrow upwards in a similar fashion to those of the Berenike capital. Unfortunately, due to the breaking of the corner ends in the latter, it is not possible to accurately compare the ends between the two capitals.

Outside Egypt, blocked-out capitals were widespread in Cyprus, for example, in Salamina, Amathus, and Nea Paphos (Brzozowska-Jawornicka, 2019, pp. 177–178; Wright, 1992, p. 460), where simplified detail is quite common. Their stylization is based on both Ionic and Corinthian capitals. Pseudo-Corinthian capitals of Cyprus differ significantly from the Berenike capital. The cross section of the double rings, positioned at a considerable height from the capital’s bottom, resembles the shape of an acanthus leaf. The top part consists of four corners, with a near-classically shaped volute ending with a plain scroll. What stands out among the instances of this common form is the detail used in the capital located at the front portico of the Temple of Apollo Hylates in Kourion (Figure 11j). Its corners, like those of the Berenike capital, have been cut down to the ruled surface, although in this case they have been ended with a straight cut. Despite the evident differences in other elements, it may be considered comparable to the capital from Berenike due to the shape of its corners alone.

3.5 The Development of the Corinthian Capitals in Egypt

In the context of the elements discussed earlier, the simplified model for the design of the Berenike capital can be traced back to the Corinthian order rather than the Ionic or Composite order. The Normal Corinthian capital (Figure 8) is defined by a shaft called a kalathos. It extends upwards from the ring separating the column shaft and the capital, called an astragal, and is covered with an abacus. Two rows of acanthus leaves decorate its bottom part. Cup-shaped stalks (cauliculi) rise from the upper row, curled at the ends, creating volutes in the corners and helices in the centre. Fleurons are positioned above the helices and partially blend with the abacus. The Normal Corinthian capital may be observed as early as the third century BC (McKenzie, 2007, p. 85). It was later standardized and widespread by the Romans.

The period of the conquest of Egypt by Alexander the Great and the ensuing reign of the Ptolemaic dynasty coincide with the development of the Corinthian order. Due to the lack of the standardized form, Alexandria developed its own order. The form of helices distinguishes it from its classical archetype. Ronczewski (1927), followed by McKenzie (2007, pp. 85–86), identified four types of the Alexandrian capitals (Figure 8). They are all characterized by the acanthus stalks rising directly out of the leafage and by the absence of cauliculi. Helices of type I curl towards each other. They curl outwards and away from each other in types II and III, but in type III, there is more space between them. Acanthus leaf wreaths are removed entirely in type IV, and the volutes spiral into the helices, which curl towards each other in the centre of the capital.

McKenzie distinguishes two major directions in the development of the capitals in Egypt during the Roman period. One relates to the Normal Corinthian capitals, while the other develops the designs originated in the Ptolemaic Alexandria (McKenzie, 1996, p. 135; 2007, pp. 223–227). One subtype of these new Corinthian capitals^[5] is characterized by plain acanthus leaves curved sharply downwards, as well as simple and narrow volutes. Most capitals of this subtype are notably shorter than the Normal Corinthian capital, which may result from the late origin of their design. As an example of this kind of decoration, McKenzie identifies the capital from the “Imperial Cult Chamber” in Luxor^[6] (AD 300) (Figure 12b) and the earliest known capitals of this type from the Serapis Temple in Mons Claudianus (AD 118) (Figure 12a). This development group is most closely related to the capital from Berenike (Figure 13), primarily because of identical components and similar proportions that are so different from those of the Normal Corinthian capital, in which the height of a capital should be equal to the lower diameter of the column shaft.

Figure 12

Corinthian capitals distinctive to Egypt in Roman period – McKenzie type 3: (a) Mons Claudianus and (b) Luxor “Imperial Cult Chamber” (drawing: S. Popławski, based on McKenzie, 2007, pp. 223–224).

Figure 13

Comparison of the form of the blocked-out capital from Berenike and the Corinthian capital from the “Imperial Chamber” at Luxor (drawing: S. Popławski).

This form might have inspired the capital from Berenike in the same manner as the capitals from Petra have their equivalents in the types of floral capitals. Also, similar to the capitals from Petra, it is possible to demonstrate the intentionality of the stylization of the capital from Berenike rather than the incompletion of its design.