This document is part of the online version of the book An Oasis City by Roger S. Bagnall, Nicola Aravecchia, Raffaella Cribiore, Paola Davoli, Olaf E. Kaper and Susanna McFadden, which is available at http://dlib.nyu.edu/awdl/isaw/oasis-city/. It is published as part of the NYU Library's Ancient World Digital Library and in partnership with the Institute for the Study of the Ancient World (ISAW). Further information about ISAW's publication program is available on the ISAW website. Please note that while the base URI of this publication is stable, the specific content available online will be updated to include more links to supplementary material and related digital resources.
Egyptian Religion at Trimithis during the Roman Period
4.1. The temple
Paola Davoli and Olaf E. Kaper
Figure 85. Plan of the temenos.Figure 86. Two reused temple blocks in el-Qasr (photo O. E. Kaper).
The temple area is located on top of the central hill; it was enclosed by a temenos wall of irregular perimeter (about 66 × 122 m), of which only part of the plan is recoverable and one segment about 5 m high (Fig. 85) survives. It probably had three gates, on the east, north, and west sides. At present the area is flat and harshly eroded; only a conspicuous shapeless mass of mud-brick walls marks its center. The Roman temple proper was built of sandstone blocks, but it has been completely dismantled down to its foundations. Most of the blocks and architraves have been removed and presumably reused in buildings of the Ottoman period in nearby El-Qasr, as several blocks in situ testify (Fig. 86). The area has been ruthlessly dug up, too, mainly in search of sebbakh (natural fertilizer) or clay, and possibly of treasures; consequently, its surface is dotted with circular pits of varying width and depth. These pits were cut in the deep stratigraphy formed during the pharaonic period (§2.1, 2.4) and were filled with wind-blown sand and collapsed stone blocks from the temple.↩#p162
The excavation started in 2005, and since then ten squares of 10 × 10 m have been explored. It has been estimated that most of the surviving blocks have now been recovered. Among these, few belonged to the upper courses—like the cavetto cornices—while the majority formed parts of the foundations, the paving, and the lower courses of walls and columns. So far only small sections of three collapsed walls have been found, but they were enough to allow us to build up a hypothesis about the temple's orientation and dimensions. Two of these collapses contained the lower parts of corners, with blocks forming the torus molding so characteristic of Egyptian architecture. From these indications we assume that the temple was oriented toward the east–southeast and extended over an area of about 24 × 14.5 m.↩#p163
Figure 87. Two stone blocks with baboon decoration, still together despite reuse and demolition (photo O. E. Kaper).
The style and masonry of this temple are common in Roman-period Egyptian temples. The Roman temple was built largely with blocks retrieved from earlier ones (see §2.2, 2.3), which had to be adapted to their new functions. Some of these blocks had already been reused in older buildings and so were now adapted for a second time for use in the new walls. The reutilized blocks (spolia) were placed mainly in the core of the walls, sometimes together with the blocks originally abutting them, as is evident in the case of some decorated items (Fig. 87). Because of this massive recycling, the wall masonry, comprising stones of different shapes and sizes, was neither uniform nor solid. For this reason great quantities of gypsum mortar and sandstone chips were used to fill the gaps between the blocks.1 The newly worked blocks are of small and medium size, the outer surfaces having been cut with rough bosses, a well known building technique used in Egypt since the Late Period. Not all the external walls had been smoothed, as some bosses with "pilgrim grooves" (gouges where visitors have scraped out a bit of stone to use for healing or other magical purposes) testify. Dovetail joints are rare, as is evidence of anathyrosis (the hollowing-out of joint beds). So far we cannot say how many columns there were, but some were probably located in front of the temple and included in the screen walls of a porch; others were in the front part (or courtyard), and yet others would have stood in front of the naos (the inner sanctuary). A possible contra-temple (an external shrine accessible to the public) was constructed abutting the rear wall. The layout and style of the temple were probably similar to those at Deir el-Hagar (§1.2), but on a larger scale.↩#p164
The furniture is almost completely missing, except for fragments of two stands for sacred barks and a few fragmentary statues, among which are some baboons, altars, pedestals, and thymiateria, or incense burners.↩#p165
Mud-brick buildings were also constructed in the area; these were probably annexes to the temple, as is commonly found in other temple enclosures; but the presence of fourth-century pottery, ostraka, and a couple of Greek Christian graffiti2 traced on a cavetto cornice and a stone block suggest the presence of people possibly living in this area during the late Roman phase.↩#p166
The decorated relief fragments contain the names of two Roman emperors: Titus and Domitian.3 Of the sanctuary decorated under Titus, only part of the rear wall of one room remains. It seems that the temple was demolished in antiquity, so that only the lower courses of the Roman-period temple remained, together with some collapse from higher up that had fallen to the ground. Very few stones from the higher levels of the building have been found.↩#p167
Figure 88. Goddesses in the sanctuary of Domitian.
The reliefs from the sanctuary decorated under Domitian depict a row of goddesses in Egyptian style placed all around the room (Fig. 88). These are the goddesses of the year, and each is shown presenting the hieroglyph for "year", which is a palm-rib, to the god Thoth. The inscriptions with each of them specify that the year they offer has a certain attribute. The first goddess offers the "good year", the second "the pure year", the third "the peaceful year", and so on. They can be compared to fairies bringing their gifts to the god of the temple. In the large temples of Edfu and Dendera in the Nile valley, the same goddesses are represented in long rows of thirty-nine, but in Amheida their number was more restricted on account of the limited space on the bottom courses of the sanctuary walls. On the basis of the reconstructed decoration of the rear wall, we can reconstruct the width of the sanctuary as 3.5 m, which is still considerably larger than that of any of the other known temples in the Dakhla Oasis from this time—ʿAin Birbiyeh, Deir el-Hagar, and Ismant el-Kharab (Kellis). Only the temple in Mut may have been larger.↩#p168
Figure 89. A pedestal fragment (standing on its right side) showing two Roman emperors as pharaohs holding up the sky.
In 2014, we found a large stone block with a representation of two Roman-period pharaohs, holding up the sky with their hands (Fig. 89). This decoration identifies the block as a bark pedestal, upon which the processional statue of Thoth would have rested inside the sanctuary. The relief was gilded, and its date can be estimated to be the second century CE, owing to its similarity to the bark stand at Deir el-Hagar, which dates to Hadrian's reign.↩#p169
4.2. Funerary life: the pyramids
Olaf E. Kaper
The cemetery of Amheida extended south of the city, but there were also funerary monuments to the north. The Dakhleh Oasis Project survey team excavated two tomb chapels of the Roman period there, in which were wall paintings with scenes of Egyptian gods, scenes from the Book of the Dead, and from other religious sources.4↩#p170
The principal structure in the cemetery is a large pyramid, noted earlier, and already described as a pyramid by the earliest modern visitors, Archibald Edmonstone and Bernardino Drovetti, in 1819. Herbert Winlock studied it in 1908, noting: "Around it there are the remains of mud-brick walls, and there are a great many human bones scattered on the slope on which it stands."5 He also noted another structure at the north end of the site, "a towerlike building, measuring 4 by 6 m. in plan. It is vaulted—the vault springing from pendentives which in turn rest on arches corbeled out in several steps from the walls. Inside these are traces of white stucco." A recent study by the architect Nicholas Warner, who was in charge of its restoration, has suggested that this may be another funerary monument.6↩#p171
Figure 90. Amheida Pyramid before restoration.Figure 91. The same pyramid after restoration (photo Christopher Kleihege).
The main cemetery to the south of Trimithis contains at least two mud-brick pyramids that stood above mausoleum tombs. The larger of these, mentioned in the accounts of those early visitors, survives largely intact and has recently been restored (Fig. 90, Fig. 91). It measures about 7 m in width and survives up to a height of 8 m. The pyramid is set on a square base 2.2 m high and has a steep angle of some 60 degrees. Interestingly, the Egyptian Ministry of Antiquities has in recent years excavated in a cemetery of ancient Mothis (Mut), at Bir el-Shaghala, where another series of elaborately decorated pyramid tombs has come to light.7 Here, the pyramids are slightly different because they have chambers inside. Similar chambers have not been found at Amheida for either of the two preserved pyramids; they must be underneath the pyramids, but these have not yet been excavated. It is clear from the surface remains that there was a complex of chambers immediately around the monument. The excavated pyramid tombs at Bir el-Shaghala show that their chambers could have been extensively decorated with wall paintings on plaster. Those pyramids were also covered on the outside with a "two-coat render of mud plaster covered with a lime based finishing."8 As Nicholas Warner has put it, "The [Amheida] pyramid is the best-preserved and largest example of a Roman pyramid in Egypt."↩#p172
Figure 92. A reconstructed pyramid at Deir el-Medina (photo O. E. Kaper).
There are in fact few comparable pyramids in Roman Egypt, certainly not of the size constructed at Amheida, but the shape, of course, has a long history. Originally treated as an exclusively royal prerogative, from the mid-18th Dynasty (about 1400 BCE) onwards—when the royals had stopped building pyramids for themselves—small versions of these regal monuments came to adorn private tombs. The last royal ones were steep, at an angle of approximately 60 degrees, and the private pyramids adopted the same, as at Amheida, or a similar angle. It seems that the larger pyramids of earlier times were built at a lesser angle owing to their size. The Bent Pyramid at Dahshur, erected by King Snofru, shows clearly that a steep 54-degree angle was attempted but given up around the halfway point; the top section is built at an angle of 43 degrees, which would become the norm henceforth. The private pyramids being smaller, it became possible to realize the ideal steeper angle in them (Fig. 92). The tips of obelisks seem to reflect the same concept, in that they too were always steep. Some private pyramids of the New Kingdom were built on a small podium, and others on the roof of a funerary chapel. A few were solid, but most seem to have contained a chapel, which is why so many of them have collapsed and disappeared. The Theban necropolis of the New Kingdom must have been dominated by the bright whitewashed pyramids of its many tomb chapels. There is no evidence for pyramid buildings from the Libyan period of Dynasties 21–24, but the kings of the 25th Dynasty, who hailed from the Sudan, took up the pyramid for themselves and their wives. The Sudanese versions were steep, and always set on a low pedestal. After the end of their rule over Egypt, the Sudanese kings and queens continued building pyramids into the fourth century CE.↩#p173
In Egypt, this tomb type was taken up again in the 26th Dynasty in the cemeteries of Thebes and Abydos. All known examples are built of mud brick, set on pedestals of up to one meter high, and contain a domed chamber. The tip of the pyramid was made separately, of stone, and decorated with funerary imagery referring to the sun god or to Osiris. These Late Period pyramids are clearly the model for the one at Amheida. Its size is similar to the pyramid erected in Thebes by an official named Padineith. A puzzle, however, is posed by the length of time between the 26th Dynasty and the Roman period, during which no other pyramid structures are known, and by the fact that the tradition seems to have come to an end elsewhere in Egypt.9↩#p174
The oases certainly had a distinctive cultural outlook and traditions. There are Roman-period pyramids at other sites, most notably Tomb 7 at Tuna el-Gebel,10 but it is remarkable that this type of tomb has not been preserved elsewhere in the Nile valley or in the Fayyum, except for the small mud-brick pyramids with Coptic monograms at Karanis. The smaller structures at Tuna el-Gebel just cited and the mention of a contemporary pyramid in a third-century-CE papyrus from Oxyrhynchos (P.Lips. 30) offer us limited evidence that the architectural type had not disappeared altogether. Something similar may be observed in the Fezzan region of Libya, where mud-brick pyramid tombs were built by the Garamantes, a desert people, at the sites of Charaig and El Hatir during the Roman period.11 In the papyrus from Oxyrhynchos, a testator gives instructions for the construction of a pyramid, along with other provisions including a foundation to support funerary feasts.↩#p175
Nonetheless, the introduction of large pyramid tombs at two of the major cemeteries of Dakhla is worth remarking on, pointing to a local tradition that existed nowhere else. The unique nature of the oasis may be the reason for this extraordinary independence of mind—it was characterized by an island mentality, and its cultural traditions were deliberately linked to ancient Greece on the one hand (§7.3), and to ancient Egypt on the other.12↩#p176
4.3. Onomastics
Roger S. Bagnall
Apart from the temples, the funerary monuments, and the churches (for the latter, see §5.1), we have another route to studying the religious commitments of the oasis population in the Roman period—their names. To a large extent, the names in papyri and ostraka are directly or indirectly theophoric ("god-bearing"), referring to divinities and often specifically to local cults. These typically assert that an individual is the gift of a particular god or "belongs to" the god. The repertory of such names is often highly localized, just as cults were; when documents from a previously silent part of Egypt become available, we usually reap a harvest of otherwise unknown names. Such theophoric names dominate the documents found in the Great Oasis about as much as they dominate in any other region. Most of the names in question and the divinities from which they derive are Egyptian, relatively few Greek. But the names themselves are sometimes highly Hellenized in form (Ammonios, Sarapion, for instance), most often linguistically Egyptian but given Greek endings (Petosiris, Psenamounis), and less commonly just transcriptions of Egyptian into Greek letters (for example, Pamoun).↩#p177
Because the oasis texts come mainly from the period between the mid-third century and the beginning of the fifth century of our era, the picture for personal names is somewhat different from what we might have found in the first two centuries of Roman rule, or even earlier, when the giving of theophoric names was at its peak; and the oasis picture will continue to develop as the Ptolemaic ostraka from Mut are studied by Günter Vittmann. The documents from Trimithis belong to the era of transition, between the flourishing of the indigenous Egyptian cults and the dominance of Christianity in Egypt.↩#p178
In the case of Trimithis, we have found a major divide between the texts assignable to the time of Diocletian (284–305) and the immediately following years, and those connected with the period of the occupation of Serenos' house (about 340–370). The first group is almost entirely traditional Egyptian, the second not. In the first group, we find a few gods dominating the names in the well tags (see §5.2) and in texts contemporary with them. Names formed from Amun are common, in both its Egyptian form (Psenamounis, "son of Amun," is by far the commonest) and its Hellenized form Ammonios. Amon-Nakht, the god of ʿAin Birbiyeh, may be represented by the name Pinachthes. Horus, mostly simply in the form of the god's name, is also very widely found. The third very popular theophoric name is Psais, "the Shai," referring to the snake-form divinity of good fortune. The patron divinity of Kellis, Tutu, is also well represented in the form Tithoes. These two demon-taming popular gods of the oasis were still a major source of names down to the first quarter of the fourth century. Petosiris, "the gift of Osiris," is also still widely used at Trimithis.↩#p179
Many traditional names, however, had started to go out of use at this time, and the range of divinities after whom parents named their children narrowed significantly. But alongside the regiments of people named Horos, Tithoes, Psais, Ammon, and Psenamounis, we still find in the early fourth century a certain number of names drawn from cults not at that point as influential in naming as they might have been earlier. They are still present, but now distinctly rare. The junior member of the Theban triad, Chonsu, is invisible at Kellis, and found at Trimithis only occasionally in Petechon ("the gift of Chonsu"). Bes and Sarapis make an appearance at Kellis, but Bes is invisible at Trimithis, while Sarapis is not uncommon. Oddly, Tapsais, Tutu's consort, is rare at Kellis, and Neith, his mother, is absent; neither figures at Trimithis, either. Seth, the chief god of Mut, appears still fairly commonly in the name Pisechthis, but apparently hardly ever in any other form.↩#p180
Thoth, the main god of Trimithis, is fairly well represented in the fourth-century texts, mainly in the form Pathotes ("the one belonging to Thoth"). At the same time, a certain number of names based on his Greek counterpart, Hermes, are attested. The Trimithitan onomastic repertory also includes derivatives of divinities such as Isis, Nephotes, Rait, Atum, Apis, Renenutet, Anoubis, the Nile, and a few others.↩#p181
In Serenos' circle, by contrast with the almost entirely Egyptian theophoric repertory of the well tags and the tenants represented on them, we find many Greek and Roman names: apart from Serenos himself, these include Gelasios, Gerontios, Domnion, Zoilos, Herakleios, Theodoros, Iulianus, Claudius, Nikokles, Ninos, Sarapion, Faustianus, and Philippos. There are signs here of the imprint of Greek classical literature, discussed in §7.3. And the Roman influence too is noteworthy.↩#p182
It is not yet an especially Christianized set of names, but outside the inner circle of Serenos and his agents, individuals connected with his household or its period of occupation include Makarios, Martyrios, Paulos, Dorotheos, Theodoros, Ioannes, Matthaios, Timotheos, Papnouthes, and Psenpnouthes. These are all Christian names, and the Hebrew names from the Old Testament found there too, like Jacob, Ephrem, Jonah, Elias (Elijah), and Joseph, also no doubt belong to Christians.13 Some of these names are borne by multiple individuals. By 350–370, then, a fair number of Christian names were in use; and as these are the names of adults, they were no doubt for the most part given at least twenty or thirty years earlier.↩#p183
In the late third and the first quarter of the fourth century, even though the names of Trimithitans were not yet to any great degree Christianized in the way that starts to be visible in mid-century, neither do they reflect much continuing investment in most of the large range of cults known in the oasis in previous centuries. The repertory, narrow to the point of monotony, may point to inertia, the tendency to pass names on from generation to generation in the absence of any reason to do otherwise, rather than to current thanksgiving to the gods for safe childbirth.↩#p184
Chapter 4 Notes
1 No pottery sherds were used in the masonry of the temple walls, in contrast to their massive use in the mud-brick buildings.↩#footnote-4.1
