From Paganism to Christianity in the Imperial Mausolea of the Tetrarchs and Constantine"

Ni{ i Vizantija V 115 Mark J. Johnson “FROM PAGANISM TO CHRISTIANITY IN THE IMPERIAL MAUSOLEA OF THE TETRARCHS AND CONSTANTINE” Truly remarkable discoveries of Tetrarchic imperial mausolea at Romuliana/Gamzigrad and Sarkamen made during the last twenty years have shed new light on our understanding of imperial mausoleum building during the early fourth century A.D.1 These important finds have rendered obsolete a number of theories concerning the burial places of Galerius and other members of the imperial family and have added much to the existing knowledge base concerning the specific architectural design of late antique imperial mausolea. During the same twenty years numerous articles on Constantine’s mausoleum/ church of the Holy Apostles in Constantinople have called into question previous theories regarding this lost monument. In addition, the Mausoleum of Helena in Rome has undergone recent excavations and a restoration that have brought new insights to our understanding of the first Christian imperial mausoleum. An examination of these discoveries and new scholarship will demonstrate that while important similarities exist in this group of buildings, significant differences also exist between the Tetrarchic and Christian mausolea that are best explained in terms of Constantine and his family’s adoption of Christianity. Diocletian’s mausoleum at Split was the first of the Tetrarchic funerary structures to be built and remains the best preserved of the group (Fig. 1).2 Constructed as part of the emperor’s retirement palace to which he retired in 305, it stands within a temenos wall enclosure to one side of the open courtyard that led to the entrance of the main part of the palace. The tomb is octagonal in plan and stands on a podium of the same shape that contains a crypt and is preceded by a rectangular pronaos. The octagon is 20.30 m wide measured corner to corner and is surrounded on all sides by a columnar portico 3.50 m wide. The interior of the building is circular in plan with a diameter of 13.35 m with seven rectangular and semicircular niches and the entry articulating the interior wall. The crypt is also circular with niches and entered through a narrow passageway that bends at a right angle. 1 See my forthcoming book, tentatively titled Sepulcra Divorum: Mausolea f the Later Roman Emperors, ca. A.D. 220 to 470. In the interest of space I will limit citations here, but full documentation may be found in the book. 2 H. Kähler, “Domkirche,” Mansel’e Armagen. Mélanges Mansel, II (Ankara, 1974), 809-20; S. MacNally, „Introduction: State of Scholarship,“ Diocletian’s Palace: AmericanYugoslav Joint Excavations, 5, ed. idem, et al (Minneapolis, 1989), 1-43. 116 Mark J. Johnson A dome covers the building, but is not visible externally as it is covered with a pyramidal roof of terracotta tiles. A carved cornice at the top of the wall is the only decoration on the exterior. In contrast, the interior is decorated with columns placed between the niches and with a band of figural relief sculpture at the top of the wall depicting putti hunting and three portraits usually identified as Hermes Psychopompos, Diocletian and his wife, Prisca. Both ancient sources and modern scholars generally agree that this was the burial place of Diocletian. Fragments of a porphyry sarcophagus in Split may well have come from the monument and originally held his remains. A major point of disagreement has been over the function of the two spaces. It has been argued that the lower chamber was not used for burials but Fig. 1. Mausoleum of Diocletian, Split, was only constructed to support the exterior upper room, which functioned as both . 1. , , the repository for the sarcophagus and as a setting for cult functions to honor the deceased emperor.3 It has long been my conviction, however, that the sole raison d’être of the lower room was to contain the burial, following the model of other two-storey Roman tombs. Our understanding of the monument at Split has been enhanced by the important discoveries at Gamzigrad. After several years of archaeological campaigns within the walls of the fortified palace there, Dr. Dragoslav Srejovic and his team discovered the remains of two mausolea and two consecration mounds on the Magura Hill one kilometer to the east of the palace.4 These were identified as belonging to Galerius and his mother, Romula. Like Diocletian, Galerius planned his burial in his homeland and in planning his tomb, Galerius looked to the architectural model of Diocletian’s monument at Split. The mausoleum to the north (Mausoleum I) is in a poor state of preservation with only the core of its podium remaining standing. A few fragments of its architectural decoration were found on the site, permitting a reconstruction of the building (Fig. 2). The podium was square, measuring 9.00 m per side above the second course of limestone ashlars. Near the center of the core of the podium is a small chamber, or better, burial pit, constructed of brick and 3 McNally, “Scholarship,” 22. D. Srejović and Č. Vasić, Imperial Mausolea and Consecration Memorials in Felix Romuliana (Gamzigrad, East Serbia) (Belgrade, 1994). I am grateful to the late Dr. Srejović for his kind generosity in sending me a copy of this book. 4 117 Ni{ i Vizantija V Fig. 2. Mausoleum of Diocletian, Split, section . 2. , , at one time covered with a segmental vault. It would have held either a small sarcophagus or an urn and remained inaccessible following the burial and the completion of the tomb. Based on an angled block discovered on the site, it appears that the upper level of the building was octagonal, with extant pieces of the cornice indicating a width for each side of 3.00 m and an exterior diameter, measured corner to corner, of about 7.80 m. Other than a cornice or two, the exterior does not seem to have had any embellishment. It is estimated that it would have been about 8 meters high. A door led into a small circular cella, about 4.90 m in diameter, which may have been covered with a dome. Fragments of marble and tesserae found here indicated that the interior was decorated revetment and mosaics. About 45 meters to the southeast was a larger and more elaborate structure, identified as Mausoleum II, also in a ruined state. Its foundation is in the form of a ring with an outer diameter of 11.30 m. The inner space was domed and served as a crypt, for once again is found a rectangular brick tomb enclosure, orientated east-west and large enough for a sarcophagus. As in the other mausoleum, this tomb chamber remained inaccessible following completion of the burial and construction. The exterior of the podium is dodecagonal with a diameter of 11.00 m and each side measuring 2.85m in width. It was covered with carefully dressed limestone blocks, a few of which are still in situ. On the west side is a narrow staircase, encased within the podium and giving access to the upper floor. Little remains of the upper story of the building, but once again fragments found on the site give a clear picture of its original appearance (Fig. 3). It was circular with an exterior diameter of about 6.40 m and, like Diocletian’s mausoleum, 118 Mark J. Johnson was apparently surrounded in its lowest portion by a colonnaded porch, as indicated by the discovery of fragments of marble columns and an ionic capital. A portion of entablature from the porch bears figurative decoration. The cella was also circular, measuring 3.75 m in diameter, and presumably domed. Fragments of porphyry found here probably formed part of its original decoration or furnishings. On the basis of archaeological evidence, Dr. Srejović deduced that Mausoleum I was slightly earlier in date than its neighbor and proposed that it was built for Romula while the more elaborate Mausoleum II was constructed for Galerius. Fig. 3. Mausoleum I (of Romula), Gamzigrad, recon- Both have obvious parallels to the Mausoleum of Diocletian struction of exterior at Split: two stories, ashlar ma. 3. I( ), , sonry construction, polygonal e and circular forms. Dr. Srejović was also responsible for the excavations at Šarkamen that led to the identification of another imperial mausoleum.5 As at Gamzigrad a fortified palace was planned, though only partially constructed in this case. The remains of the mausoleum are located some 246 meters to the northwest of the western gate of the fortification. They consist of the concrete rubble core of the podium of the original structure (Fig. 4). This podium is square, measuring 10.65 m per side. In its center is a burial chamber of a roughly circular shape, approximately five meters in diameter. The exterior was decorated with granite ashlars from a local quarry. The maximum preserved height of the remnant is 5.52 meters, most of which corresponds to the crypt/podium of the structure. A second level seems to have stood above the square podium but it is difficult to say whether the upper part was circular or perhaps octagonal. The lower chamber was vaulted with a dome constructed of square bricks. A burial pit was cut into the bedrock but no additional construction in the crypt is evident. The chamber does 5 D. Srejović, D. M. Tomović and Č. Vasić, “Šarkamen. Tetrarchial Imperial Palace,” Starinar, 47 (1996), 231-43; M. Vasić and M. Tomović, “Šarkamen (East Serbia): An Imperial Residence and Memorial Complex of the Tetrarchic Period,” Germania, 83 (2005), 257-307. Ni{ i Vizantija V 119 not seem to have been provided with any access, so once the entombment was effected, the intent was to leave it untouched and protected within the walls of the mausoleum. The most remarkable discovery was made in the excavations of the crypt. The burial pit had been plundered, but a small amount of ashes and burnt bones, possibly human, remained. About 1 meter from the pit in a small niche cut into the rock, excavators found 38 gold objects.6 For the most part these were jewelry for a woman: three necklaces, three rings, two earrings, a diadem, two hair-fasteners and a pendant. In addition, there were nine gold foil votive plaques, decorated Fig. 4. Mausoleum II (of Galerius), Gamzigrad, reconstruction of exterior with linear designs with four of them bearing the impressions . 4. II ( ), , of the obverse of gold solidi of Tetrarchic date. In size, construction materials, and design with a closed crypt in the podium surmounted by a second-level structure, this building is very close to the two mausolea discovered at Gamzigrad. The similarities with Gamzigrad in design and construction throughout the complex led the excavator to believe that the same architect and construction crews did the work at Sarkamen.7 The Mausoleum of Helena, located outside of Rome, was the first such building constructed for a Christian member of an imperial family (Fig. 5).8 Built in the period between ca. 315 and 327, it inaugurated an important innovation in planning which was then followed in most other Christian imperial mausolea. Unlike the preceding examples that were freestanding structures, this 6 I. Popović and M. Tomović, “Golden Jewelry from the Imperial Mausoleum at Šarkamen (Eastern Serbia),” Antiquité tardive, 6 (1998), 287-312, on 307. 7 Srejović, Tomović and Vasić, “Šarkamen,” 233. 8 Most recently see L. Vendittelli, “La conservazione e la valorizzazione del Mausoleo di Sant’Elena. Nuovi dati dai lavori di scavo e di restauro,” Ecclesiae Urbis. Atti del congresso internazionale di studi sulle chiese di Roma (IV-X secolo), Roma, 4-10 settembre 2000), ed. F. Guidobaldi and A. Guiglia Guidobaldi (Vatican City, 2000), 771-97; and M. G. Filetici, “Il Mausoleo di S. Elena: il recupero dell’edificio ed il cantiere di restauro,” Manutenzione e recupero nella città storica, progetto e intervento. Atti del II convegno nazionale, Roma, 12-13 settembre 1995, ed. M. Segarra Lagunes (Rome, 2005), 347-54. 120 Mark J. Johnson Fig. 5. Mausoleum (of the mother of Maximin Daia?), Sarkamen, section . 5. ( ?), Ш , mausoleum is attached to another building: a Christian basilica, dedicated to SS. Marcellinus and Petrus. According to the Liber Pontificalis, Constantine built the basilica and then added to it the mausoleum for his mother.9 Excavations by Deichmann and Tschira in the 1950s and by Guyon in the 1970-80s have clarified the architectural form and history of the complex. It consists of an occidented coemeterium basilica, a courtyard enclosed by a portico on the south flank of the church, a walled enclosure to the north and the circular mausoleum with its rectangular vestibule on the east as well as numerous smaller mausolea attached to the basilica and the portico. An extensive catacomb lies under the building and was directly accessible from the portico. The mausoleum is circular in plan with an exterior diameter of 27.74 m. The west flank of the rotunda was flattened in the area where it abutted the vestibule. The vestibule, rectangular in plan and measuring 9.50 by 28.40 m, survives only in foundations. It was entered through the narthex of the basilica, to which it was connected by three wide openings, or through a door in its south wall. The interior of the rotunda, like the exterior, is also circular, with a diameter of 20.18 m. The usual semicircular and rectangular niches are disposed around the interior wall with the widest one placed opposite the entry. A square staircase, entered through the niche north of the entrance, gave access to the upper reaches of the building. The exterior of the mausoleum has a two-tiered elevation. The lower tier, corresponding in height with that of the interior niches, is a solid cylinder without any windows or architectural articulation except for a cornice of marble corbels. At this point the exterior wall is set back, reducing its diameter substantially. The drum is taller than the lower part of the building and is pierced by eight windows set in curved niches. The upper level terminates with a second cornice of projecting marble corbels. The dome has collapsed for the most part with only part of its lower reaches on the north side remaining. 9 Liber Pontficalis., ed. L. Duchesne, I, (Paris, 1886),182. Ni{ i Vizantija V 121 The building is constructed of brickfaced concrete, with the brick facing originally stuccoed over with the stucco drafted to look like ashlar masonry. The stucco was then painted a deep red, attested by various remaining fragments.10 The use of hollow jars, or pignatte, in the construction of the dome gave the building its popular name of Torpignattara. The interior of the Fig. 6. Mausoleum of Helena, Rome. reconstruction mausoleum consisted of . 6. Х , , a single spacious and well-lit room. The elevation of the interior is divided into three zones: the niche level, a clerestory with a window placed above each niche, and the dome which rose to a height of about 25.50 m and was decorated with mosaics. Recent reconstructions of the mausoleum have included an opaion at its apex, though there is no evidence whatsoever of such a feature.11 The Liber Pontificalis, a text compiled in the sixth century from earlier donation documents gives the most complete picture of the kind of furnishings found in the Christian imperial mausolea when it lists the gifts made by Constantine for the mausoleum of his mother: „A paten of purest gold, weighing 35 lbs.; 4 silver candelabra overlaid with gold, 12 ft. in height, each weighing 200 lbs.; a golden crown, that is, a chandelier, with 120 dolphins, weighing 30 lbs.; 3 gold chalices, each weighing 10 lbs., set with prases and hyacinth gems; 2 gold pitchers, each weighing 40 lbs., an altar of purest silver, weighing 200 lbs., (placed) in front of the tomb [sarcophagus] of the blessed Helena Augusta, which is of porphyry with images, 20 silver chandeliers, each weighing 20 lbs.“12 As no lower chamber existed here, it is supposed that the porphyry sarcophagus stood in the niche opposite the entry. The other items are exactly the kinds of objects that would be found in a church, demonstrating that the eucharist was celebrated in the mausoleum and blurring the distinction between tomb and church. 10 Venditelli, “Conservazione,” 780. Filetici, “Mausoleo,” 347; Venditelli, “Conservazione,” 788. 12 Ed. Duchesne, I, 182; tr. R. Davis, The Book of the Pontiffs (Liber Pontificalis) rev. 2d ed. (Liverpool, 2000), 23-24. 11 122 Mark J. Johnson Whatever lines of separation existed between the two building types were completely obliterated in Constantine’s plans for his own burial which took place in a church dedicated to the Apostles in Constantinople. The Apostoleion was located inside the new city walls on the highest hill of the newly founded city. It was probably not begun before the summer of 334 as the Eusebius places his account of the church after that of Constantine’s Persian campaign, which took place in the spring of that year. According to Eusebius, Constantine at first hid the fact that he intended the church to serve as his mausoleum, though this later became evident. His purpose in doing this, Eusebius explains, was to share in the devotion offered to the Apostles. Therefore he ordered that twelve sacred thekai, interpreted variously as „cenotaphs“, „pillars“ or „plaques,“ in honor of the Apostles to be set up near his sarcophagus, six on each side.13 A new reading of the sources indicates that Constantine was thinking of the church as the actual memorial for the remains of the apostles. Relics of Andrew and Luke were located and transferred to Constantinople and introduced into the church on June 22, 336.14 Evidently, he intended to fill the other thekai with the remains of the other apostles as they were discovered.15 All of this was located near the altar of the church, where Constantine was entombed in late May or early June 337. Thereafter, he and the Apostles „were honored with the performance of the sacred ordinances and mystic services.“16 There have been numerous interpretations of the form of the original building over the years with some arguing that Constantine built a cruciform church to which his son later added a circular mausoleum. Recently Cyril Mango has argued that Constantine did not build just a church but a church-mausoleum and this is what Eusebius described.17 He notes that the description of Eusebius seems to fit a domed circular building rather than a basilica, for it is the height, not the length, of the building that elicited comment. Nor is there any mention of columns, a standard in descriptions of basilicas. Therefore, the original building was circular in form, while Constantius added the cruciform church of the Holy Apostles to it later, specifically to house the relics of the apostles. If this interpretation is correct, then Constantine was following his predecessors in constructing an imperial mausoleum in the form of a domed rotunda. All late antique imperial were built as domed rotundas, either circular or octagonal, a type first adopted by the Emperor Gallienus in the 260s and used in the last Roman imperial mausoleum, that of Honorius in the early fifth century. While sharing a common form, there are important differences between those built for pagan and Christian rulers as illustrated in the examples under discussion here. The pagan imperial mausoleum is a freestanding building with two stories. The lower storey is often partially interred, dark with only tiny openings that seem to be more for ventilation than for light if there are openings at all and with limited or no access following the burial. The upper storey consists of a domed room that is decorated and, if it has windows, they are small, leaving the interior somewhat poorly illuminated. The upper room was not used for burials 13 Vita Constantini, 4.60; tr. A. Cameron and Hall, 176. 14 R. W. Burgess, “The Passio s. Artemii, Philostorgius, and the Dates of the Invention and Translations of the Relics of Sts. Andrew and Luke,” AnalBoll, 121 (2003), 5-36, on 29. 15 Burgess, “Dates,” 29. 16 Vita Constantini, 4.71; tr. A. Cameron and S. Hall, (Oxford, 1999), 181. 17 C. Mango, “Constantine’s Mausoleum and the Translation of Relics,” Byzantinisches Zeischrift, 83 (1990), 51-62;“Addendum,“ 434 123 Ni{ i Vizantija V but for cultic functions, a place where the deceased was commemorated and offerings made to his or spirit. The evidence from Gamzigrad and Sarkamen demonstrates that the lower and upper chambers at Split must have served the same functions. In contrast, the Christian mausolea were for the most part, attached to churches, or in the case of the Apostoleion, combined the function of church and mausoleum. They. too, were domed rotundas, either octagonal or circular in plan but, unlike their pagan counterparts, were not provided with a crypt. Instead the functions of the two superimposed spaces in pagan mausolea were combined into a single room in the Christian examples. Furthermore, these spaces were not dark and subterranean but flooded with light from large windows. One might suppose that this architectural difference was a reflection of religious belief concerning the afterlife: the pagan Hades was a dark and gloomy subterranean place while the Christian paradise was filled with light. As the builder of both his mother’s mausoleum and his own, Constantine was responsible for both maintaining the traditional building type for imperial tombs, but introducing significant changes that produced a symbiosis of architectural form and his Christian beliefs. Ма к Џ. Џ А А А ШЋА А А А А А А Ј А , Ш / А . . / К Ј , К . ђ , . ђ , К . : , , , , . Ш . К ђ . ђ , . , . А , , 124 Mark J. Johnson