Printer Friendly

Imagined Empire: representing imperial communities in Roman Dougga.

Civic Representation: A Portrait from the Temple of Saturn in Dougga


This paper is the second in a series of inquiries into the visual culture of the Roman Empire under the Severan emperors (193-235 CE), each asking how residents of the empire experienced or formulated the very idea of empire. (1) The proper subject of this paper is a temple complex honoring the goddess Caelestis, built by a patriotic aristocrat in North Africa under the last of the Severan emperors. Although I wish to examine how this temple complex worked to create in its visitors an idea of the larger imperial world, primarily through its use of a series of place personifications, I begin by looking at a curious portrait perhaps with no direct relationship to the topic.

This large marble statue of an unknown man (fig. 1) was discovered in a niche within a peripheral cella in another of Dougga's temples, this one belonging to the consort of Caelestis and one of the most celebrated gods in North Africa, Saturn. (2) When I first confronted this object at the Bardo Museum in Tunis, the oddity of the portrait struck me immediately. A dignified, craggy-faced man stood togate before me, approximately life-size, in the manner of a proper Roman citizen--and atop his head was placed a mural crown. It looked ponderous, in the way of priestly crowns worn by the men and women of imperial Asia Minor, and did not seem to sit comfortably. Adding to the apparent discomfort was the misgendered use of the mural crown, otherwise worn almost universally by female personifications of GrecoRoman cities. (3) Yet for all the strangeness of the juxtaposition, the man in the portrait sacrificed none of his authoritative presence--perhaps he became even more notable, more memorable for it. I asked myself why the iconography of the mural crown would have appealed to this ancient male patron, and wondered how legible such an image was to its contemporaries.

The portrait's known facts are easy to rehearse. The temple of Saturn was one of Dougga's most prominent temples, situated above the city near the apex of its dominating slope. This temple was also remodeled in the Severan period, although several decades earlier than the remodeling of the Caelestis temple and by a different set of patrons. Built atop an ancient precinct of the Phoenician god Baal, the temple stood as both a modern space and one that preserved the memory of pre-Roman times. During the reign of the Gordians, following the death of Alexander Severus in 235, the mural-crowned togatus was erected to honor a man now presumed to have been a priest in the cult. He stood in a niche in one of the side cellas of the main temple, where excavators found him fallen and buried, but still in fairly good condition.

The mural crown, of course, occasioned immediate comment; and the notion that he was some sort of personification, that he must be some sort of personification, would have moved into position as the most likely explanation had it not been for his veristic portrait features. (4) The strange combination of masculine portrait and feminine mural crown led discoverers to assume that the portrait misinterprets the archaic (by the time of this statue's production) military honor of the corona muralis, granted to soldiers who were the first to scale an enemy wall. (5) But his civilian garb alone makes this interpretation unlikely, to say nothing of the statue's date and location. Claude Poinssot definitively rejected the military theory in his 1955 re-examination of the statue and suggested that the man saw himself as in some sense a tutela, or guardian spirit, of his community--and in a correlated footnote said that such a notion would be readily available to everyone in Dougga because of the personification series on display in the city's temple to Caelestis. (6) Given the common reuse of extra or repurposed marbles, there is even a possibility that the priest appropriated for his portrait a stone originally intended to form one part of the Caelestis temple personification series--perhaps one blocked out but never finished, perhaps flawed in the manufacture, or otherwise finally deemed unnecessary for the temple complex.

There are many reasons to be cautious about this theory of the priest-portrait's origins (we will come to the issues surrounding the personification series itself later on), including the difficulty of access to the statue itself. At the moment, the portrait is difficult to subject to close scrutiny--it is raised atop a high pedestal and placed within a niche in the Bardo that imitates its original display context. Certainly, the museum presentation conveys the necessary experience of dignity and honor, stemming from the very elevation and distance that now hinders the interested viewer. Although I was unable to examine it up close myself, and in spite of a general lack of good published photographs, there is a close-up black and white photograph in Ranuccio Bianchi-Bandinelli's overview of the late Roman world available to consult. This high contrast photograph may reveal signs of recutting. The craggy portrait face appears to exist at a slightly smaller scale than the massive mural crown and indented marks on the folds of the toga could derive from an earlier, different draping. (7) It may seem far-fetched to suppose that an elegantly draped, voluptuous female form could so successfully be rendered into a ponderous togatus; but on the day that I first saw the priest of Saturn, I had been only moments earlier engaged in a group discussion of another of the Bardo's strange late imperial marbles--the so-called Tunisian farmer. This group of classicists and ancient art historians, led by Natalie Kampen and Clemente Marconi to benefit graduate students at Columbia University, stood before the Tunisian farmer for perhaps 45 minutes and reconstructed its transformation from a large statue of Ceres into its present condition. (8)

Whether our curious statue of the priest of Saturn derived from an earlier marble or whether it was a new creation, its visual vocabulary ends up tapping into the ideas embodied by place personifications. That is to say, the portrait in some way reflects the experience of an imagined imperial community. Given the recency and proximity of the Caelestis personification series and given that temple's status as pendant to the temple of Saturn, the priest's statue in one way or another relates to the mix of provinces and cities in the Caelestis series. This relationship prompts several questions. What were the likely goals or the varied effects of such a sculptural series on the residents of Dougga? How did communities in general imagine their connections to the larger empire? And how did individuals engage with such imagery to form self-images as members of local and imperial communities? It is to the work of the late Severan period that this paper now turns, and to its creative installation of the empire writ large to benefit one small corner of the world.

Late Severan Dougga: Traces of the Past in the Urban Landscape

Dougga today is an archaeological site in northcentral Tunisia, located a few hours' drive from Tunis. As an ancient city it was subject to at least three Mediterranean powers: first as a Libyan settlement incorporated into the Carthaginian empire; afterwards as a residence for a branch of the Numidian ruling family; and finally, refounded as an outpost of Roman culture in the imperial province of Africa. Today it seems remote and vigorously bucolic, far away from strategies of conquest or assimilation. Yet the city's remains speak very strongly of its connections to imperial communities in the ancient Mediterranean. (9)

The most prominent monument of Dougga is the mausoleum of a Numidian prince, presumably a lesser member of the royal family that came to power after Rome's destruction of Carthage.10 His name was Ataban, son of Iepmataph, son of Palu. His nomenclature establishes a sense of generational continuity, and the formal structure of the monument seems simultaneously to implant a strong local presence and to reach outward to a cosmopolitan world. It stands in a line of local dolmens, weathered stones that marked a cosmic pattern onto the valley's topography. At the same time it honors in its architectural and sculptural program that wonder of the ancient world, the Mausoleum in Harlikarnassos. Its bilingual inscription purposes a broad reception. Its sculpture blends and interprets Egyptian, Punic and Greek motifs. The monument stood and continued to stand at the point of transition, where the highway wandering southward through the valley, began its climb to the hillside community of Dougga above.



Another of Dougga's monuments also claims a connection to the larger Mediterranean world, although many centuries later and under very different local politics. Dougga's temple to the goddess, Dea Caelestis (fig. 2), was built between 221-235 CE.11 These remains allow for a stimulating reading of collective identification practices under the Severans. After a brief contextual sketch of the temple, I will present its sculptural program and suggest how it articulates a connection between Dougga and the wider world, between individuals and society. Testing its interpretive flexibility against concepts developed in Benedict Anderson's post-colonial theory, I close with some thoughts on how the program, once established, inflected the understanding of Dougga's residents as participants within the empire. (12)





The Architectural Experience of Dougga's Temple to Dea Caelestis

The temple of Caelestis was a relative latecomer to Dougga's urban infrastructure, although its goddess had long held an honored place in the local pantheon as a Roman-era successor to the Phoenician Tanit. (13) Her sphere was broad and included healing, nurturing, and both human and agricultural fertility. Sailors had long looked to her for protection; and her associations with water continued in North Africa, where she occurred in connection with both harbors and springs and is frequently described as "bringer of rain." She was the object of both personal and civic cult and hundreds of extant stones and terracottas attest to her enduring local popularity. (14)

The Caelestis temple in Dougga was built to the southwest of the city's municipal center, on the outskirts of the community, atop a semicircular terrace that negotiated a gently sloping hillside (fig. 3). The terrace, 52 meters in diameter, was entirely enclosed by architecture--on the northeast (where the platform was cut into the existing hillside), by a semi-circular annular vaulted portico; and on the southwest (the orientation of the temple's main facade), by monumental architecture built atop vaulted substructural terraces and fill, now missing. An unpaved section likely supported some sort of monumental gate, which must have been more ornamental than functional, although the collapse of the slope below the temple platform opens this part of the reconstruction to various interpretations. While it is tempting to imagine the platform has having enjoyed the sweeping views of the valley available today, the strength of the foundational work seems to indicate some kind of enclosing architecture.

The temple precinct, however imposing the architecture may have been, was not accessed through this axial facade structure. The entrances were placed symmetrically at the northeast and southwest corners and were marked by aedicular gateways with large hinged doors and pedimental entablatures that bore dedicatory inscriptions. The western entrance stepped down onto the adjacent terrain; and its inscription gave the name of "Quintus Gabinius Rufus, Felix Beatianus" and announced to the visitor that the temple was built by him to augment the generosity of his own parents, who had earlier vowed sums for the benefit of the community. (15) The main inscription on the semicircular portico behind the temple (on which more later) supplements this story of inherited liberality and indicates as well that the temple complex was built on and donated from Gabinius' own land. (16) We can assume that this doorway opened onto a landscape vista as breathtaking as the one that currently exists there--a dramatic juxtaposition of land and marble that would have enhanced the impression of Gabinian wealth and marked the transition from urban fabric to bucolic cultivation.

The eastern entrance was negotiated by two steps that led up to the temple from a series of halls and annex chambers, one with a large pool, probably used for ritual ablution.17 Although it is now clear that this entrance was adorned with an identical inscription to its opposite, the topographical associations on the eastern side have long suggested the temple's connection to the emperor. An early excavator, reconstructing some fragments said to be found "dans les oliviers a peu distance de l'entree [nord-est]," suggested that this inscription informed the visitor that the temple was built on behalf of the emperor Alexander Severus.18 It now seems that these fragments, which are lost, could not have crowned the doorway in question as by now two identical inscriptions originally crowning both entrances have been recovered from the site. Perhaps the questionable fragments adorned an annex building or existed as an isolated marker of the imperial presence within the complex. The eastern entrance, as mentioned above, was the city-side entrance; and, whether or not there were intermediate inscriptions honoring Alexander Severus in this area, there was certainly an unmistakable topographic reference to him available to all temple visitors. The temple to Caelestis was linked directly to Dougga's forum on a road which, at a time contemporary to the construction of the temple, was given the standard imperial punctuation of an monumental arch to that emperor (figs. 4-5). The emperor's inscription on the temple pediment, in fact, styles him "M. Aurelius Alexander Pius, Felix Augustus," a cadence precisely matching that of Gabinius' on the temple entablatures. (19) Donor and emperor are thus, in their nomenclature, made pendant to one another in the heavenly precinct. The two entrances to the temple platform, therefore, may be interpreted as underscoring the double and richly layered symbolic programs built into this monument, the local and the imperial, and they speak to the imagining by Gabinius and his collaborators of a purposeful connection between the two.

Having gained entrance to the open courtyard the visitor was presented with a clear outline of his potential range of movement. Paving stones (fig. 3) articulated a broad initial open area which must have contained the altar to the goddess and served to pool celebrants in front of a peripteral temple in a traditional Corinthian order. Whether or not a celebration was taking place, this plaza was certainly the ideal point from which to admire the temple's imposing architecture. From this plaza there were three possible avenues of movement. On the central axis of the plaza there was a paved, direct approach to the temple. The temple can be described as thoroughly informed by standard Roman temple construction, not only in its columnar order, but also in its position atop a podium several meters above the courtyard, strongly rectangular within its curved terrace.

The other two avenues of movement led the supplicant into the symmetrical openings of the semi-circular portico that ran the length of the curved external wall. This space (fig. 6) was also marked architecturally as separate from the courtyard, being raised slightly above that level and initiated by massive piers with sculptural niches. The colonnade ran for twenty-two intercolumnar bays and was paved with a mosaic of white marble tesserae arranged in radiating concentric arcs. The arced pattern formally corresponded to the lightweight annular vault (whose moorings are still visible in the back wall) that ran above the walkway, together enhancing the relative intimacy of the space and mirroring the combination of rectilinear and curvilinear forms found on a larger scale in the courtyard below. The rectilinear format of the Corinthian colonnade, of course, provided an open, yet intermittent view of the courtyard and temple; and the interior wall must originally have been covered in painted frescoes, the design of which was very likely vividly colorful and busily geometric in the style of the Severan period. (20)

The remaining area of the courtyard was not paved, and most scholars have agreed that it was planted, perhaps even with trees that formed a sacred grove. Archaeology is unlikely to determine the look of this unpaved space, as its layers have been too badly disturbed to provide any proof. Taller trees would have increased the visual complexity of an already crowded field and reduced visibility of several components of the program, which causes me to prefer a reconstruction of low plantings. (21)

Finally, a fully articulated curving limestone entablature ran the length of the portico, with architrave and high inscriptional frieze. This entablature survives for over three-quarters of its length and sets out the goals and intentions of the temple donor, Gabinius, though it now rests on the ground in front of the walkway (fig. 7). In spite of some lacunae, we learn from it that the Republic of Thugga (Dougga's ancient name) offered this temple to honor the goddess, the city, and the Gabinius family, who were obligated perpetually to maintain it. The building cost 60,000 sesterces, with 30,000 more going toward silver cult statues of Caelestis. (22) It closes thus: "Q. Gabinius Rufus Felix Beatianus alone has granted this from his own money and lands and also given statues and other adornments, a feast, and a gymnasium, constituted from his generosity, as additional gifts." (23)

The gymnasium may have been part of the as yet unexcavated annex structures, combining with piping and pools to emphasize bodily health. Yet most important for this inquiry is the mention of "statues and other adornments"--CVM STATVIS CETERISQ[ue]--that decorated the precinct. There is, of course, no way to know precisely which statues these were, and the relative absence on site of sculptural fragments leaves the matter open for speculation. The entablature of the curving portico, however, provides hints that prompt us to imagine these statues in the form of a series of city and province personifications. The original excavators recognized several such names carved into a blank band of the upper cornice molding, occurring at regular intervals above the main inscription. These names survive poorly as most of them straddled the joined entablature blocks and the collapse has damaged the majority beyond legibility. The original surveyors detected the city names of Thugga and Laodicea, and the province names of Dalmatia, Judea, Mesopotamia and Syria. Since their discovery the city of Carthage and the province of Hispania have also been recognized, bringing the known total to eight. As for the remainder, it would be pointless to make suggestions other than to note the strong likelihood that the city of Rome and the province of Africa would have been included. (24)

Serialized Geography and Architectural Reception

The placement of the inscriptions, which were incised and painted red for visibility, makes fairly certain that statues stood at intervals atop each columnar order, in a manner consistent with standard Roman plaza decoration such as with the caryatids from the Forum of Augustus or the Dacian captives in the Forum of Trajan. The discovery of a few limestone fragments of draped female figures may indicate that these statues were full-bodied personifications. (25) Logically, there may originally have been twenty-three to twenty-five of these statues, one above each column (and perhaps the two end pilasters); only a third of the series remains.

Explanations for this series of provinces and cities have moved in different directions. Claude Poinssot, summarizing the site, suggested that the personifications marked points of cultic intersection throughout the Mediterranean. (26) Dougga's physical and social proximity to Carthage (whose patron deity was Juno Caelestis and whose waterfront boasted a large hilltop complex to the goddess) make cultic connection between these two cities, at least, very likely. Perhaps the complex of Dougga was somehow a satellite of the one in Carthage? (27) Reasoning that Caelestis was perhaps more popular throughout the Mediterranean under the Severan dynasty than at any other time, Poinssot roped all of her surviving inscriptions from around the empire into a pan-Phoenician assembly of devotees. Laodicea was taken to be the port of Apamea in Syria, long ago a Phoenician city, very probably with a cult to Tanit (though there has been found as yet no evidence of it). Dalmatia, as it happens, hosts one of the few non-African dedications to Caelestis; Hispania, of course, was once under Punic control. Mesopotamia, Judea and the fifteen missing personifications were then left to somehow fall within this pan-Mediterranean religious community.

Poinssot never referred to a Severan period series of oracular statements from the temple of Apollo in Klaros, but these at least do provide evidence that devotees of contemporary cults took steps to establish links throughout the Mediterranean. These inscriptions, acknowledging the gift of advice from Klarion Apollo, have surfaced in several provinces, including the African ones of Mauretania and Numidia, as well as in Britannia, Dalmatia, Sardinia, and Syria. There may never have been any parallel list of locational inscriptions in Klaros itself, but the sanctuary must have housed oracular votives from all over--objects that, as a collection, are open to interpretation as a geographic series. In either case, through inscriptions in provincial locales or as material objects in Klaros, the signs of Apollo's power manifest a spatial network of cultic connections. (28) A similar understanding could well have animated the Dougga series.

On the other hand, the prominent position of Alexander Severus within the monument's program has prompted others to claim that the Dougga series reflects Severan military and diplomatic activity. (29) Syria, Mesopotamia, and Illyricum (if not precisely Dalmatia) are known to have been military theaters during his reign, while his family's ties to Apamea justify references to Syria and Laodicea and his personal curiosity about Judaism could have prompted the inclusion of Judea. The reference to Laodicea, in addition, may somehow relate to Septimius Severus' decision to elevate that city to colonial status. (30) This explanation, though sound in principle, fits the surviving personifications no more aptly than the religious explanation, it being entirely unclear, for example, how Hispania may fit. Perhaps it is not farfetched to think these two interpretations need not be mutually exclusive. The latter theory leans too heavily on the idea that the imperial government had a directive hand in the installation (even implying that Gabinius took prompts from the sources now at our disposal), while the former relies upon a relatively modern understanding of institutional religion, assuming that cultic installations were branches of a more centrally organized unity. But if we set aside a desire for a comprehensive solution, other ways to explain the series emerge.

To this end I put forth a third interpretative approach for the Dougga personifications, one prompted by elements with a presence at the temple equal to the goddess or the emperor--that is to say, prompted by the role of Gabinius and his family. The donative history of the Gabinii is fairly well known and too complex to share here in full.31 Suffice it to say that they began to build or repair several prominent monuments in Dougga during the reign of Trajan and continued to be active until the mid-3rd century. Included among their benefactions is a modest temple dedicated jointly to the deities Concordia, Frugifer, and Liber Pater built, like the temple of Caelestis, on family land. (32) This temple housed another small sculptural series, a set of portrait statues of prominent Gabinii. These statues, presumably still on display at the time of the Caelestis temple's construction, would have made an interesting parallel to the later personifications--yet another equation of land and persona. In a gesture that underscored their interest in the local, the Gabinii also decorated the city's forum with its portico; while alternately (closer in time to the Caelestis monument) a familial connection to the Severan house was demonstrated by one Gabinia Hermona, who built a grand pseudo-peripteral temple to the deified Caracalla. This last stood very near to where the later Arch of Severus Alexander and temple of Caelestis would stand. Finally, the aqueduct and two of the city's largest public cisterns were situated within this family's lands. (33) Our Gabinius, as perpetual priest of the cult, was the principal architect of the Caelestis program and we are entitled to intuit within its personifications a network of his own conception, even if we cannot demonstrate his selective process or fully reconstruct it.

A reconstruction of the visual and ideological program of the Dougga personification series is, of course, served by literary or sculptural comparanda in Rome, such as the Portico of Pompey, the Augustan Forum, and Portico of Nations, or the Hadrianeum reliefs (all of which presented a series of provinces or peoples). (34) Alternately, to better match the social conditions of Dougga, we may compare series in provincial settings such as the sixty Gallic tribes inscribed on an altar to the imperial cult at Lyon, capital of Gaul, or the Sebasteion complex at Aphrodisias in Turkey. (35) Modest parallels in North Africa itself include the odd personifications carved in relief in the theater at Sabratha and the temple pediment in Oea, among others. (36) However, I suspect that even if we suddenly knew all of the constituents of the Dougga program, we would still not have discovered an unambiguous interpretation. In fact, the most striking feature of the programs we know about is their diversity, a diversity that must spring from the specific agenda of their builders and the imagined connections each locality was making to the larger world.

That this larger Mediterranean habit of serialized geography was sometimes put to use in individualized interpretation by residents of the empire is evidenced by another North African artwork, dating approximately to the time of Gabinius and the Caelestis temple. Although it does not use geographic personifications to accomplish its goal of presenting select places from around the Roman world, a large floor mosaic from what was most likely a privately owned villa in Haidra (ancient Ammaedara) does create the impression of a personal journey. (37) Decorating the floor of a large reception room (5.3 x 6 m.) with niches on its back and side walls, and preceded by a marble sill with incised decoration at the room's entrance, the mosaic presents a squared version of the Mediterrannean sea, with five groups of three islands placed in shifting relationship to the room's four walls. These islands, twelve of which survive, were labeled with the names of cities and islands from the central and eastern Mediterranean. (38) There is no attempt at a true depiction of geographical relationships on this floor (city representations are featured, for example, as independent islands even when the island on which they are actually situated is depicted elsewhere in the mosaic), nor do the "islands" occur in an order that even approximately accords with their actual distribution in the Mediterranean. They are assembled, rather, to allow people sitting or standing in different parts of the room to always have access to a portion of the display. In order to take in the full series, as most thoughtful visitors to the room surely were tempted to do, one had to island-hop, as it were, and move around the room on a journey with no prescribed order.

Without rehearsing the full argument for the interpretation of this floor as a personalized geography, most likely reflecting a real or a hoped-for journey to a series of holy sites, I will simply note that ten of the surviving twelve places have strong connections to the cult of Venus/ Aphrodite. Whether we imagine that the design derived from a pilgrimage itinerary, from a sacred geography, or from the owner's personal experience, it in every case supports the interpretation of the Caelestis personifications as similarly representative of cultic connection. In this reconstruction, Gabinius and his family approved the series in order to promote in the minds of all who visited the complex a particular imagining of the distances bridged. The political framework of the empire, of course, was the architecture through which these bridges were intuited, a necessary framework that even in the case of the Haidra mosaic was used to link the local to the imperial, for the marble sill leading into the island-strewn reception room was incised with a curious collection of motifs. Measuring instruments (of the forge or of navigational science) appear on one end of the sill, flanking a centrally placed head in rather cartoonish profile, while on the other end appears a large triumphal arch with a road schematically receding beyond it. It is not unreasonable to suppose, as does Bejaoui in his analysis of the room, that the arch in question is the local monument in Ammaedara honoring the emperor, Septimius Severus. The crude figure evokes portraiture generally, perhaps that of imperial coin types or perhaps marbles of the villa's owners. In any case, as one stepped over the threshold of the Haidra mosaic room, one was stimulated simultaneously to imagine several aspects of the world beyond the confines of one's hometown: that the route to this world was open to the cosmopolitan visitor, that it was secured by an imperial presence, and that the experience of it, when it came, was a serialized geography.

Benedict Anderson's Imagined Communities and the Dougga Series

To return to Gabinius and his temple, it is also interesting to set aside questions of programmatic intent and ask instead how the complex may have functioned socially within the civic and cultic communities of Dougga. This approach brings three effects quickly to the fore: first, the program worked architecturally, placing visitors to the precinct into a structured set of relationships; second, the personifications rely upon and help to perpetuate a pre-existing conception of worldliness; and finally, the anthropomorphic representations of place functioned to link visitors corporeally with empire, adding to the spatial and intellectual cues a bodily one. These effects can be explored using some of the theoretical structures mentioned above, offered by Benedict Anderson to decipher what he terms "the development of collective subjectivities" in the modern world. (39)

In his work on the Southeast Asian experience of the colonial and post-colonial periods, Anderson coined the term "imagined community," (the title of his major work of 1988) by which he means nationalist identifications. (40) For Anderson, print-media and bureaucratic practice formed the building blocks of the international idea that states coexist in relatively equal space and that individuals "naturally" belong to one or another of them. In his more recent collection of essays, The Spectre of Comparisons, Anderson refines his description of collective subjectivity by considering "the material, institutional, and discursive bases that necessarily generate two profoundly contrasting types of seriality," which he calls unbound and bound. (41) Unbound seriality is exemplified in his thinking by "open-to-the-world plurals" such as nationalists, bureaucrats, or member states of the United Nations. Every participant in an unbound series holds a theoretically equal place with all the others, and these others may be added indefinitely as need arises. Bound seriality is exemplified by closed memberships, articulated by social practices such as the census or by identity groups such as Asian-American or Gay and Lesbian. Participants in a bound series must possess or perform the necessary credentials to occupy his or her spot; they are not provided equally to all comers. Both forms of seriality are employed to minimize individual specificity, and each, in its own way, serves as an avenue to collective identification. (42)

Anderson's persuasive thesis presents a challenge to those of us who want to know how pre-modern societies imagined themselves into being. Art historians, who posit visual experience as a major component of ideological formations, have an opportunity to apply his ideas directly to architecture and other visual media. (43) For this reason, I read the Dougga personifications as an unbound series, an open invitation to all those who visited the temple to imagine themselves as part of a bigger world. In no way do I assert an ancient parallel to modern nationalism or identity politics, nor do I hope to obtain a definitive interpretation of the Roman monument. Rather, I construe an ancient imagining of empire from the traces of the Dougga series.

First, we should recall the architectural placement. I have already spoken of the secluded nature of the sanctuary interior, as well as of the textual and topographical configurations of the two entrances.

Together these features suggest a controlled, even a limited, access to the temple precinct, one that, incidentally, would certainly aid in the preservation of its expensive silver cult statue(s). Additionally, the plaza pavement is somewhat limited in extent, not capable of housing a very large crowd, perhaps eighty to one hundred people at most. The peripheral portico served as a sort of ambulatory, perhaps channeling excess crowds during celebrations, but more probably directing supplicants in purposeful movement or, at night, providing a space for those sleeping among its columns to await a divine communique (a common practice in healing cults). In addition, the high temple would have prevented unified views of the portico exterior and thus encouraged an interior experience, a placement among columns reserved to those who gained admission. As a result, it would not have been possible to take in the whole of the personified series ornamenting the portico roofline. Yet from any vantage point within the portico or in the courtyard below, several of these draped statues would have been visible. The columnar spaces would thus have been visually marked, at times of peak activity, by bodies both above and within. (44)

The installation of personifications in architecture that itself provided a bodily sense of passage was not accidental. Serialized geography was equally embodied and most visitors to the temple would already have had some experience with the concept, whether they had fingered the reverses of Hadrianic province coins or witnessed festivals in Carthage during which allegorical figures paraded by, or even had heard a list of Roman legionary names (which made a mixture of geography and virtue). (45) Perhaps, to a civic way of thinking, the statues may have suggested something like the Greek Panhellenion (a collective of cities) or, if literary, the sequential locations described in travelogues or romance novels. (46) However it was intuited, the provinces and cities above the Caelestis portico brought each visitor into an architecturally parallel relationship with a representation of Mediterranean community. An imperial, or cultic, or even a resistant reading of these figures remained available; what mattered was the illusion of sameness that they provided, opening up the world in an inclusive manner.

The corporal presence of these figures, it seems to me, played a crucial role in this process of visual identification. Contemporary with the Severan dynasty is a series of honorific and funerary steles from Ghorfa in neighboring Algeria (fig. 8). These steles are representative of a type that is found throughout North Africa and present the dedicant in an architectural relationship to the goddess Caelestis. On one of the steles, now on display at the Bardo Museum, a togate man stands between two columns in a temple-like facade. The niche that frames him is evocative of statuary displays, even perhaps suggesting the placement of a cult statue. The architecture is seen from multiple viewpoints, with the soffit below the pediment providing a non-mimetic, but compelling representation of dimensional space. Above the pediment can be seen the goddess Caelestis, bearing the so-called egg cluster and vessel that are typical of her iconography, with attendant figures and flowing water below her. (47) Other stones show the honored human raised up on a podium, one supported by yet more bodies in the form of corpulent caryatids. (48) I do not read these steles as literal translations of experiences such as those provided in the Caelestis temple complex at Dougga. Rather, I note their architectural and figural vocabularies as suggestive of the types of bodily associations devotees were able to make. These stele are self-placements both within architecture and within divine hierarchies.



That the dedicator of the Caelestis stele illustrated here saw a connection with the larger imperial world of Rome, for instance, is made clear by the inclusion of yet another iconographic symbol. In the honored space of the pediment there stands an imperial eagle with wings unfurled, a frequent symbol around the Mediterranean for the cult of the emperors. In Dougga, such a stele would have provided an irresistible association with the city's own Capitolium, with its eagle-borne apotheosis of Antoninus Pius (fig. 9). (49) Through the agency of the goddess and within the political structure of the empire, a devotee thus imagined for himself a successful translation into the afterlife.

Finally, suggesting that the Caelestis complex personifications functioned as bodily invitations for the residents of Dougga, I return to the statue of the priest of Saturn (fig. 1). Claude Poinssot, as mentioned earlier, suggested that the man saw himself as in some sense a tutela, or guardian spirit, of his community; and, in a correlated footnote, he said that such a notion would be readily available to anyone who knew the Caelestis complex.50 Poinssot was certainly correct; and the experiential analysis of the Caelestis temple provided here has demonstrated how such an association was likely conceived. The priest of Saturn has placed himself, in a rather creative way, within the unbounded series of empire articulated by the Caelestis personification series, in which it became possible for him to imagine himself as an equal participant in worlds both human and divine.

The formula for implementation of personification series will not be found in some list, discovered or reconstructed from monuments such as this one. The formation lay in something more ephemeral--in the collective imaginings of empire, which originated in specific circumstances, but which relied on a more or less unproblematic assumption that regions existed as stable and comparable (if not always fully equal) units. The representation of these units as human bodies opened the way for a corporal identification with cities, landscape and spheres of influence, making of a person a living embodiment of the very stuff of empire.

UNC Charlotte

(1.) See J. F. D. Frakes, "Monuments of Passage: Roman North Africa and an Emperor on the Move," Arris: The Southeast Chapter Journal of Architectural History 19 (2008), 53-69.

(2.) For the statue, see C. Poinssot, "Statue du temple de Saturne (Thugga)," Karthago 6 (1955): 32-7. The information in the following paragraphs derives from this source unless otherwise indicated. For the cult of Saturn generally, see M. Le Glay, Saturne Africain (Paris: BEFAR, 1966).

(3.) For a rich consideration of several aspects of this iconographic phenomenon in the ancient Mediterranean, refer to S. Matheson, An Obsession with Fortune: Tyche in Greek and Roman Art (New Haven: Yale University Art Gallery, 1994).

(4.) The only other instance known to me of a male portrait with mural crown are obverse coin portraits of Agrippa. Agrippa seems in these to be posthumously honored as a restorer of the city of Rome. These are discussed by P. B. F. J. Broucke, "Tyche and the Fortune of Cities in the Greek and Roman World," in Matheson 1994, 45. Another male shown several times with a mural crown is the hero figure Aristaios of Cyrene, also discussed in Broucke (1994). See Matheson1994, figs. 13-14 for illustrations.

(5.) Poinssot 1955, 58ff.

(6.) Poinssot 1955, 66.

(7.) R. Bianchi-Bandinelli, Roma: la fine dell'arte antica (Milan: Feltrinelli, 1970), 217-219. For Bianchi-Bandinelli, who accepts the juxtaposition of mural crown and sacrificing togatus without comment, the statue serves as an example of the sophistication of African sculptural workshops in comparison to earlier "provincial arts" from other provinces.

(8.) Contributors to this conversation, in addition to the two mentioned above and myself, were Giovanna Assenso, Celeste Lovette, Elizabeth Marlowe, Annalisa Marzano, Rebecca Molholt, Jean Sorabella, and Herica Valladares.

(9.) For Dougga's history and circumstances, see the excellent treatment of J. B. Rives, Religion and Authority in Roman Carthage from Augustus to Contantine (New York: Clarendon Press, 1995).

(10.) C. Poinssot, Les ruines de Dougga (Tunis: Institut National d'Archeologie et Arts, 1958), 58-61.

(11.) The following descriptions rely primarily on my own experience of the site, but rest upon the excellent architectural study of the temple by J.-Cl. Golvin and M. Khanoussi, Dougga: etudes d'architecture religieuse. Les sanctuaries des Victoires de Caracalla, de "Pluton" et de Caelestis (Bordeaux: Ausonius, 2005). See also R. Cagnat and P. Gauckler, Les monuments historiques de la Tunisie: premiere partie, les monuments antiques: les temples paiens (Paris: Leroux, 1898).

(12.) B. Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1989) and B. Anderson, The Spectre of Comparisons: Nationalism, Southeast Asia and the World (London: Verso, 1998).

(13.) For Caelestis, see U. Antonelli, "Tanit-Caelestis nell'arte figurate," Notiziario Archeologico 3 (1922): 41-69 and G. H. Halsberghe, "Le culte de Dea Caelestis," ANRW II.17.4 (1984): 2203-2223. In specific regard to the Severan period, see also I. Mundle, "Dea Caelestis in der Religionspolitik des Septimius Severus und der Julia Domna." Historia 10 (1961): 228-237. Tanit is succinctly discussed in Rives 1995, 164-166 but see also A. M. Bisi, Le stele puniche (Rome: Istituto di Studi del Vicino Oriente, 1967).

(14.) F. Bertrandy, "Les representations du 'signe du Tanit' sur les steles votives de Constantine," Rivista di studi fenici 21.1 (1993): 3-28, A. M'Charek, "Maghrawa, lieu de provenance des steles punico-numides dites de la Ghorfa," MEFRA 100 (1988.2): 731-760, and M. Ghaki, "Steles libyques et neopunique de Tunisie," in L'Africa Romana: lo spazio marittimo del Mediterraneo occidentale: geografia, storica ed economia, eds. M. Khanoussi, P. Ruggeri and C. Visimara (Rome: Carocci, 2002): 1661-1678.

(15.) Golvin and Khanoussi 2005, 161-163. The portal inscriptions, all uniform, read CAELESTI AVG(ustae) SACR(um) / Q(uintus) GABINIVS RVFVS FELIX BEATIANVS LIBERALITATES PARENTVM MVLTIPLICAVIT EXCOLVIT DEDICAVIT.

(16.) Golvin and Khanoussi 2005, 183-208.

(17.) For the importance of water in African temples, see A. Bel Faida, "Eau et sacre en Afrique romaine," in L'Africa Romana: lo spazio marittimo del Mediterraneo occidentale: geografia, storica ed economia, eds. M. Khanoussi, P. Ruggeri and C. Visimara (Rome: Carocci, 2002: 17151727.

(18.) Cagnat and Gauckler 1898, 28. Golvin and Khanoussi do not mention this inscription, although they frequently refer to the many losses between the 1890's excavation and today.

(19.) Poinssot 1954, 71-72. The pediment inscription is presented in Golvin and Khanoussi 2005, 124-126.

(20.) Golvin and Khanoussi 2005, 135-156.

(21.) Trees are suggested in Cagnat and Gauckler 1898, 27. In agreement are G. Dareggi, "Le sanctuaire de Coelestis a Thugga: un temoignage de l'ideologie imperiale de l'epoque des Severes," Histoire et archeologie de l'Afrique du Nord: Actes du IVe Colloque Internationale I (1990): 199-213 and H. Eingartner, "Fora, Capitolia und Heiligtumer im westlichen Nordafrika," Die romische Stadt im 2 Jahrhunderts n. Chr. Der Funktionswandel des offentlichen Raumes (Cologne: Rheinland-Verlag, 1992). Golvin and Khanoussi 2005, 173 report the irrecoverability of certainty, but did find that the unpaved area was designed to be level with the paved area.

(22.) In their reconstruction of the inscription, Golvin and Khanoussi 2005, 208 argue that the divinity here had a dual nature--local and imperial--and that there were in fact two silver statues of the goddess in the cult chamber to reflect her nature. The inscription begins by mentioning the goddesses Caelestes, but later reverts to a unified singular.

(23.) Golvin and Khanoussi 2005, 202.

(24.) Cagnat and Gauckler 1898, 29 for the original six. Poinssot 1958, 42 introduced Carthage without comment, and Dareggi 1990, 201 cites Hispania. P. Liverani, "'Nationes' e 'civitates' nella propaganda imperiale," RM 102 (1995): 243 mentions that the longest surviving fragment measures 130 cm. and omits Hispania from the list without comment. Golvin and Khannoussi 2005, 141-143 provide an extremely useful chart of all the entablature blocks, including a photograph of "Mesopotamia." They report that Syria and Hispania are now missing. They also (144) suggest the inclusion of the province of Africa and the city of Rome. I was unable to find or read any of these on a visit to the site in March 2000.

(25.) The fragments were found with an Asclepius torso and a bust of Antoninus Pius. Cagnat and Gauckler 1989, 30 suggested the fragments may have been a statue of Caelestis, although if so it must have been a secondary one as the entablature inscription makes clear that the main statue(s) would have been in silver. Golvin and Khanoussi 2005, 104 report the list and the opinion of Cagnat and Gauckler.

(26.) Poinssot 1958, 42. Eingartner 1992, 225-6 leaves the door open for this possible interpretation, mainly on the basis of the crescent-moon shape of the precinct, which he considered might have architectural parallels.

(27.) A theme in Rives 1995. For the temple to Caelestis in Carthage see H. R. Hurst, The Sanctuary of Tanit at Carthage in the Roman Period: A Reinterpretation (Providence: Journal of Roman Archaeology Supplementary Series 30, 1999).

(28.) The Severan period inscriptions to Klarian Apollo are listed in T. L. Robinson, "The Theological Oracles of Didyma and Klaros" (PhD diss., Harvard University. 1981).

(29.) Dareggi 1990 and Eingartner 1992. See also G. Dareggi, "Severo Alessandro, Romanus Alexander, e il complesso santuariale di Thugga," Latomus 53 (1994): 848-858.

(30.) For the emperor's attention to Judaism, see the SAH: Vita Sev. Alex 39. For Laodicea, see Herodian 3.3.

(31.) M. Khanoussi and L. Maurin, Mourir a Dougga: recueil des inscriptions funeraires. (Bordeaux and Tunis: Ausonius, 2002) gives a complete inscriptional record of the family, while Rives 1995, 125-132 provides a satisfying narrative account.

(32.) V. Brouquier-Redde and S. Saint-Amans, "Epigraphie et architecture religieuse a Dougga: l'exemple des templa Concordiae, Frugiferi, Liberi Patris, Neptuni," in Dougga (Thugga) Etudes Epigraphiques, eds. M. Khanoussi and L. Maurin (Paris: Ausonius, 1997): 175-200.

(33.) Concordia et al: Poinssot 1958 #16, 52-54; Forum: Poinssot 1958 #8, 38-40; Caracalla Temple: Rives 1995, 126; Cisterns: Poinssot 1958 #33 and #35, 70-71.

(34.) Liverani 1995 lists all Roman and provincial series known to date. The Caelestis series, the latest known, is treated on pages 243-4. He suggests that if anything connects all the series, it is a link to the imperial cult. However, no explicit reference to such a cult is known for the Caelestis temple (even though Dougga elsewhere has many monuments where the imperial cult was certainly practiced).

(35.) For the altar at Gaul, see Strabo 4.3.2. and the discussion of D. Fishwick, "The Temple of the Three Gauls," JRS 62 (1972): 46-52 and D. Fishwick, The Imperial Cult in the Latin West: Studies in the Ruler Cult of the Western Provinces of the Roman Empire (Leiden, E. J. Brill, 1987-1992). For the Sebasteion at Aphrodisias, see R. R. R. Smith "Simulacra Gentium: The Ethne from the Sebasteion at Aphrodisias," JRS 78 (1988): 55-138.

(36.) The Theater of Sabratha is beautifully photographed: A. di Vita and R. Polidori. Libya: The Lost Cities of the Roman Empire (Bonn: Konemann, 1999): 175-181. The same source illustrates the Oea relief on page 9.

(37.) F. Bejaoui, "Iles et villes de la Mediterrrannee sur un mosaique d'Ammaedara (Haidra, Tunisie)," Comptes rendues des seances de l'Academie des Inscriptions et belles-lettres, 1997: 825-858.

(38.) Bejouai presents them in this order: the southwest cluster includes the islands of Scyros and Cyprus with the Cypriot city of Idalium; the southeast cluster (partly destroyed) puts Cnidos with two lost companion "islands"; the northeast cluster shows Rhodes, Cytherae and the Cypriot city of Paphos; the northwest cluster has the island of Lemnos, the city of Eryx, and a missing third; while the central cluster depicts the island of Naxos and the cities of Egusa (on Sicily) and Cnossos (on Crete).

(39.) Anderson 1998, 42.

(40.) See the revised edition B. Anderson's Imagined Communities: 1991, 5-7.

(41.) Anderson 1998, 29-45.

(42.) For the further development of Anderson's theories and a set of responses to his work, see J. Culler and P. Cheah, Grounds of Comparison: Around the Work of Benedict Anderson (London: Routledge, 2003).

(43.) Anderson suggests that sensual reality has less influence as a prompt to imagined community than do social structures when he states that the common use of spices in a region of Malaysia is "less serious" than matrilineal kinship patterns: Anderson 1998, 10 ftn. 18. The material conditions of lived experience, it seems to me, are better understood as the tools with which social ties (kinship or otherwise) are communicated as ideas and performed. They can hardly be considered less serious, nor can visual culture.

(44.) M. Gleason, Making Men: Sophists and Self-Presentation in Ancient Rome (Princeton: Princeton University Press, 1995) provides a fascinating treatment of the performativity of Roman bodies on public display.

(45.) For Hadrianic coins, see K. W. Harl, Civic Coins and Civic Politics in the Roman East, AD 180-275 (Berkeley: University of California Press, 1987); for festivals in Carthage see Rives 1995; and for legionary names see H. M. D. Parker, The Roman Legions (Cambridge: Cambridge University Press, 1958).

(46.) For the Panhellenion, see M. T. Boatwright, Hadrian and the Cities of the Roman Empire (Princeton: Princeton University Press, 2000), C. P. Jones, "The Panhellenion," Chiron 26 (1996): 29-56, A. Spawforth, "The Panhellenion Again," Chiron 29 (1999): 339-352, A. J. Spawforth and S. Walker, "The World of the Panhellenion I. Athens and Eleusis," JRS 7 (1985): 78-104, and A. J. Spawforth and S. Walker, "The World of the Panhellenion II. Three Dorian Cities," JRS 76 (1986): 88105. For romance novels, see B. P. Reardon, The Collected Ancient Greek Novels (Berkeley: University of California Press, 1989).

(47.) Illustrated in Bisi 1967, plate 31.

(48.) Bisi 1967, plate 30.2.

(49.) F. Dohna, "Gestaltung offentlichen Raumes und imperiale Ideologie am Beispiel des Kapitols von Thugga," RM 104 (1997): 465476 and Poinssot 1956, 34-38.

(50.) Poinssot 1955, 66.
COPYRIGHT 2008 Southeastern College Art Conference Review
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Frakes, James F.D.
Publication:Southeastern College Art Conference Review
Article Type:Critical essay
Geographic Code:1USA
Date:Jan 1, 2008
Previous Article:Pin-Up Grrrls: Feminism, Sexuality, Popular Culture.
Next Article:Two Simple Cups: The Transcendence beyond the Enlightenment into Romantic Religiosity in Francisco Goya's Self-Portrait with Dr. Arrieta, 1820.

Related Articles
Victorian Writers and the Image of Empire: The Rose-Colored Vision. (Reviews: modern Britain).
Napoleon and Europe. (Reviews).
Imperial Co-Histories: National Identities and the British and Colonial Press.
The Gospel of Matthew in its Roman imperial context.
The Roman Predicament: How the Rules of International Order Create the Politics of Empire.
Roman Imperial Policy from Julian to Theodosius.
Empire and After: Englishness in Postcolonial Perspective.
Between the Eternal City and the Holy City: Rome, Jerusalem, and the imperial ideal in Britain.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters