3.1 Prolegomena to Porticated Streets

Albeit with notable exceptions, modern scholars have tended to underestimate the importance of colonnaded (or porticated) streets in the urban fabric of late antique cities, particularly in the West.¹ Even with regard to the East, where the majestic colonnaded avenues still visible in cities such as Palmyra, Apamea and Ephesus bring the monumental presence of these prepossessing features powerfully to life, there is nonetheless a continued inclination (again with notable exceptions) to propose their de-monumentalization by the later sixth century at the latest, following what remains a hallowed locus communis, the metamorphosis of the grand colonnaded thoroughfare of antiquity into the pullulating chaos of the medieval suq. Jean Sauvaget’s schematized diagram representing the chronological (d)evolution of the colonnaded street in his study of Laodicea of 1934,² still often cited and reproduced, has become emblematic of the idea that even before the Arab conquest in the seventh century, colonnaded streets had seen their pristine monumentality compromised by the encroachment of shops and residences haphazardly built into the spaces between the columns and thence outward onto the paving of the streets themselves. Hugh Kennedy’s influential article of 1985 translated Sauvaget’s diagram into a paradigm more generally applicable to the cities of the Levant, which in his formulation were well on their way to becoming the teeming bazaars characteristic of the medieval period by the later sixth century;³ and Helen Saradi has recently further extended the presumed range of the phenomenon to the ‘Byzantine city’ of the sixth century in general.⁴

As we shall see, this perspective requires substantial modification. At least as late as the fifth century in the western Mediterranean and the sixth in the East, porticated streets were either refurbished or built anew with remarkable frequency, joining city walls and churches as much the most prolific monumental architectural forms of the era. Further, the partition walls and structures in generally irregular masonry or more perishable materials that ultimately did come to encroach on the intercolumnar spaces and sometimes – more to the point – the roadbeds of so many colonnaded streets are often extremely difficult to date even for methodologically advanced archaeologists. As the majority of the extant remains were uncovered before the widespread use of rigorously stratigraphic excavation techniques in the past few decades, the essential question of chronology remains very much open. Much of the encroachment once dated to the sixth century largely on the basis of a priori notions about the inexorable decline of the classical townscape already in that period may well belong to the ninth century or later.⁵ There are indeed a number of tantalizing indicators of the continued vitality of some streets, in both East and West, into the seventh and eighth centuries, when they apparently continued to frame the unfolding of urban ceremony in ways that suggest that their accessibility and their imposing architectural presence had not been fatally compromised either by unchecked decay or by the installation of shops and other commercial establishments, which had indeed been a ubiquitous presence in porticated streets throughout late antiquity, as we shall see, and for that matter under the high empire as well.⁶

Before proceeding, there is a brief matter of terminology and definitions to clear up. Strictly speaking, a colonnade is composed of columns, bearing either a flat (trabeated) entablature, or a series of arches (an arcade). In the West more often than the East (where the proximity of good sources of marble permitted a ready supply of monolithic columns), the trabeated and arcaded structures flanking streets were often supported by pillars of masonry or even wood, whence they cannot properly be called colonnades.⁷ Hence, I will hereafter prefer ‘porticated streets’ as a catchall designator for all such structures, and ‘colonnaded streets’ in the more restricted sense of the term. There is the further issue of what exactly constitutes a porticated street: my discussion of these structures will focus on streets lined by lengthy porticoes of relatively uniform aspect, as opposed to streets flanked by buildings with their own irregularly sized and spaced porticoes giving onto the roadbed, which will have presented a more heterogeneous and discontinuous appearance. A porticated street constitutes a cohesive architectural scheme that transcends the character of the individual structures situated behind the porticoes, imbuing the street as a whole with an architectonic identity essentially independent of the adjacent buildings, for all that these buildings are often directly connected to the porticoes.⁸

It should also be stressed that such streets started to proliferate long before late antiquity, beginning in the first century AD. The longest and perhaps the earliest of all is the cardo at Antioch, repaved and widened by King Herod at the turn of the millennium and lined on both sides, for its entire length of 2,275m, with continuous files of columns by the mid-first century AD.⁹ By the later second century, extensive colonnades lined both sides of one or more principal streets at cities across the eastern Mediterranean and North Africa, as well as at Rome itself and at a number of sites in western Europe (e.g., Lincoln, Reims, Italica), though the Eastern exemplars usually receive considerably more attention, as they tend to be both more monumental and better preserved.¹⁰ It was only with the imperially sponsored urban or quasi-urban foundations of the third century, however, that the colonnaded or porticated street came into its own as an indispensable element in the architectural language of imperial power, East and West, from Philippopolis in the 240s to the palace-city complexes erected by the Tetrarchic emperors, as we saw in the preceding chapter.

Thereafter, porticated streets became (with the exception of city walls) the most ubiquitous, expensive and ambitious form of secular monumental architecture erected from the later third century through the sixth, leaving an indelible stamp on the greatest cities of the late Roman – and post-Roman – world, and many others besides. The majority of the most extensive and costly street porticoes either restored, extended or constructed ex novo were located in prominent centers of civic and ecclesiastical administration, provincial and imperial capitals above all,¹¹ which increasingly tended to gravitate around the colonnades that framed and connected the palaces and churches where rulers and bishops lived and performed their public duties.

Hence, an inquiry into the question of why the civic and, to a lesser extent, ecclesiastical authorities responsible for the most ambitious interventions in the urban topography of late antique cities devoted such a considerable portion of their limited resources to porticated streets should have much to reveal about the ways power and authority were enacted, publicized and given architectural form in the late Roman world. It should also help to explain the motives behind the development of the porticated street into the ‘imperial’ architectural feature par excellence in the third century, and the surprising persistence of these streets, and – more to the point – the ceremonial and commemorative praxis associated with them, through the sixth century and beyond in both East and West. Hence, the function of these grand avenues will concern us at least as much as their form. Our focus will be on the period ca. 300–600, which is rich enough in literary, epigraphic, archaeological and iconographic evidence to permit a relatively detailed perspective on the activities and ideological constructs that animated these streets and made them such indispensable features of the greatest cities of the epoch.

3.2 Imperial Capitals in Italy: Rome and Milan

In a fine recent article on colonnaded streets in Constantinople in late antiquity, it is stated that the city of Rome differed markedly from Constantinople in that it had no colonnaded streets worth speaking of in late antiquity.¹² In reality, Rome had long featured extensive street porticoes. The most ideologically charged route in the city, the Triumphal Way followed by returning conquerors since the early days of the Republic, came to be lined by covered stone arcades over much or all of its urban tract, from the Bridge of Nero in the Campus Martius, through the Circus Flaminius and the Forum Boarium, and on past the Circus Maximus to the Forum Romanum.¹³ Elsewhere, following the catastrophic fire of AD 64, a number of principal thoroughfares in the devastated city center were reconstructed as wide, porticated avenues, which imbued the imperial capital with a stately veneer reminiscent of the royal capitals of the Hellenistic eastern Mediterranean.¹⁴ Nero’s gargantuan new palace, the domus aurea, gravitated around a triple portico said by Suetonius to be a mile long;¹⁵ and Nero’s architects encased the nearby Sacra Via – the final stretch of the Triumphal Way running through the forum to the Capitoline Hill – within lofty, arcaded porticoes at the same time.¹⁶

But the most extensive evidence for the erection of monumental street porticoes at Rome comes from late antiquity, beginning in the later fourth century, when several of the busiest and most symbolically charged axes of communication in the city were embellished with extensive porticated façades. Indeed, one of the most significant architectural interventions witnessed in Rome in the second half of the fourth century involved the erection of new porticoes along what was then becoming one of the very most important roads in the city, the street leading north through the Campus Martius to the Pons Aelius, the bridge leading to the mausoleum of Hadrian, and from there onward to St. Peter’s. As the evidence for the monumentalization of this street has been presented elsewhere,¹⁷ a resume of the salient points will suffice.

At the time of or soon after the construction of the Aurelian Wall in the 270s, the Pons Neronianus, the old Neronian bridge that served as the principal connection between the northern Campus Martius and the west bank of the Tiber, went out of use.¹⁸ Thereafter, all traffic on the two main roads in the area, the so-called Via Recta (the modern Via dei Coronari) leading east through the Campus Martius past the Baths of Nero and on to the Via Lata, and the Via Tecta, the porticated road running northwest from the Circus Flaminius, effectively the urban continuation of the ancient Via Triumphalis (the modern Via dei Banchi Vecchi), was diverted northward to the bridge leading to the Mausoleum of Hadrian, which subsequently became the sole crossing point in the northern Campus Martius. The decision to privilege the previously little-frequented Pons Aelius over the Pons Neronianus was presumably conditioned by defensive considerations, as the mausoleum was admirably suited to function as a fortified bridgehead on the far bank of the Tiber, allowing access to the bridge to be strictly controlled in a way that would have been impossible at the Pons Neronianus without the addition of substantial new structures at its western approaches. Traffic on the ‘Via Recta’ and Via Tecta was subsequently diverted north to the Pons Aelius along the modern Via del Banco Santo Spirito, as can be deduced in part from the fact that almost no traces of the paving of the ‘Via Recta’ and Via Tecta have been discovered between the point where they converged on the road leading to the mausoleum and the Pons Neronianus;¹⁹ the paving of these streets may indeed have been reused to monumentally re-edify the tract of road leading to the Pons Aelius, as may the columns from the tract of the Via Tecta situated between the intersection with the Via del Banco Santo Spirito and the Pons Neronianus, which had become a blind alley with the closure of the bridge (Figure 3.1).²⁰

Figure 3.1 The western Campus Martius in Rome, with the route of the porticus maximae to the Pons Aelius and the arch of Valentinian, Valens and Gratian.

(After Dey Reference Dey2011, fig. 4.1.)

While this last point may be considered speculative, it is clear that the road leading to the Pons Aelius did indeed come to be flanked by impressive porticoes, which seem to have been installed, apparently for the first time, in the later fourth century. The key piece of evidence is the lost inscription from the triumphal arch of Valentinian, Valens and Gratian, completed ca. 380 athwart the southern approaches to the Pons Aelius, which survives in a ninth-century transcription: ‘Our lords, emperors and caesars Gratian, Valentinian and Theodosius, pious, fortunate and eternal augusti, commanded that (this) arch to conclude the whole project of the porticus maximae of their eternal name be built and adorned with their own money.’²¹ In addition to providing a firm date for the arch, the inscription strongly implies that the porticus maximae, which must refer to porticoes along the street leading to the arch, was realized in connection with, or shortly before, the arch itself, in what was clearly perceived as a unified architectural scheme, for which the arch served as the final and concluding element (arcum ad concludendum opus omne porticuum maximarum). The best conclusion is that the new porticoes extended the existing colonnades of the Via Tecta along the stretch of road leading to the only surviving river crossing in the area, which in turn provided access to St. Peter’s, one of the crucial nodes in Rome’s emerging Christian topography.²²

The centrality of St. Peter’s in the spiritual and ceremonial life of Rome was further accentuated by the construction of an additional covered colonnade on the far side of the Tiber, flanking the road leading from the mausoleum of Hadrian to the church. While it is first securely attested in the Gothic War of Procopius, written in the mid-sixth century, the colonnade was likely built rather earlier, perhaps indeed shortly after or in connection with the porticus maximae, to which it would effectively have formed the extra-urban continuation, creating a nearly continuous colonnaded panorama stretching all the way from the Via Tecta in the intramural heart of the city, via the new arch of Valentinian, Valens and Gratian and the Pons Aelius, all the way to the Vatican.²³

The other two most substantial street porticoes erected in late antique Rome led from the Porta Ostiensis in the Aurelian Wall along the Via Ostiensis to the church of St. Paul’s outside the walls; and from the Porta Tiburtina along the Via Tiburtina to the basilica of San Lorenzo. While they too are only known from later sources (of the mid-sixth and eighth centuries, respectively),²⁴ both structures may well date as early as the later fourth century. They likely belong to approximately the same period as the portico leading to St. Peter’s, as the similarities in form and function common to all three suggest that they were conceived as interrelated parts of a unitary architectural scheme designed to connect the suburban shrines of Rome’s three most venerated martyrs with the Aurelian Wall, and thus with the city center. In the case of the portico to St. Paul’s, a date in the 380s appears especially likely, as it was in precisely this period (beginning in ca. 383–84) that the Constantinian church on the site was replaced by a massive new basilica sponsored by Valentinian II,²⁵ the same Western emperor under whom, it should be remembered, the triumphal arch and the porticus maximae leading to the Pons Aelius were erected. It is surely as good a hypothesis as any that the imperial authorities, presumably working in concert with Pope Damasus (366–84), sought to monumentalize the route to the new St. Paul’s and link the church to the city center during or soon after its construction, by means of an imposing portico that mirrored the extra-mural extension of the porticus maximae leading to St. Peter’s. Damasus, of course, is best known for his tireless efforts to restore and popularize the suburban shrines of Rome’s leading martyrs,²⁶ whence the temptation grows powerful to imagine that he actively collaborated in a scheme to anchor the devotional circuit of the Roman periphery he did so much to promote on three porticated ‘access roads’ leading to the shrines of Rome’s three greatest martyrs.

By the sixth century, when the annual liturgical calendar of the Roman church was largely complete, the three churches of St. Peter’s, St. Paul’s and San Lorenzo remained the only extramural shrines regularly visited by the popes in the course of the stational processions that increasingly came to define the sacred topography of the city.²⁷ The porticated avenues thus provided a grandiose architectural framework for the ceremonial processions led by the popes to the three great extramural sanctuaries in the course of the annual liturgical cycle.²⁸ The prominence of these processional routes in the topographical horizons of the city and the ideological agendas of its bishops was such that they continued to be maintained for centuries, often at enormous cost. In the late eighth century, Pope Hadrian I (772–95) restored all three porticoes, in the case of the route to St. Peter’s allegedly by reusing 12,000 tufa blocks taken from the embankments of the Tiber to complete the project.²⁹

But while the popes may have been unusually successful in preserving the grand processional ways marked out in Rome in late antiquity into the early Middle Ages, similar porticoes once dominated the cityscapes of other Western capitals in late antiquity, notably those that superseded Rome as preferred imperial residences.

In Italy, Milan served as the primary seat of government from 286 until 402, during which time it too saw the erection of an architecturally prepossessing porticated avenue. As excavations undertaken in the 1980s during the construction of the MM3 underground line demonstrated, the new street took shape in ca. 375–80, and thus at almost exactly the moment when the porticus maximae at Rome were built, under the patronage of the same Italian emperor, Valentinian II. The Milanese porticoes were two stories high, and extended for nearly 600m along both sides of the road leading to Rome, beginning from the Porta Romana in the circuit-wall and terminating, again like the Roman exemplar, with a triumphal arch.³⁰ Moreover, just as at Rome, the colonnades were linked to an especially prominent extramural church, the new Basilica Apostolorum built by Bishop Ambrose between 382 and 386 at the midway point of the newly aggrandized street (Figure 3.2).³¹ In light of the new evidence for dating the porticoes scant years before the appearance of the church, there is now better reason than ever to imagine – as past scholars operating under the assumption that the rebuilt street dated to the third century were already tempted to do³² – that the proximity of the imperial triumphal way was a determining factor in Ambrose’s decision to situate the first and most prestigious of his three (or four) great extramural churches adjacent to its porticoes.³³ In so doing, Ambrose – surely consciously – reproduced the topographical relationship between St. Peter’s and the Via Triumphalis at Rome, and also, as we shall see, between a principal Constantinopolitan triumphal avenue and the city’s own church of the apostles, the Apostoleion, whose unusual cross-shaped plan directly inspired the cruciform layout of the Ambrosian Basilica Apostolorum.³⁴ So too at Trier, the east end of the cathedral – suggestively dedicated to St. Peter and rebuilt under Valentinian and Gratian (364–83) – featured a twelve-sided aedicula enclosed within a massive square precinct, another Apostoleion in miniature that was, moreover, directly accessible from the porticated cardo departing from the Porta Nigra.³⁵ In each case, we see civil and ecclesiastical authorities effectively collaborating in the creation of new, monumental urban itineraries, punctuated by imperially sponsored street porticoes, triumphal arches and city gates that provided a suitably grandiose setting for the formal processions that dominated the ceremonial repertoire of both church and state.³⁶

Figure 3.2 Milan: the via Romana colonnades and the Basilica Apostolorum.

(Francesco Corni/Civico Museo Archeologico di Milano.)

The dedication of the Milanese church of the Apostles on May 9, 386 culminated with the translation and installation under the high altar of the relics of John the Baptist and the Apostles Andrew and Thomas, doubtless conveyed to the church in a festive procession along the Via Romana arcades.³⁷ The event indeed left such an impression on the Christian faithful that when Ambrose later consecrated the Basilica Ambrosiana, the assembled crowd loudly demanded that he consecrate the new church with relics, as he had the basilica in Romana.³⁸ The Basilica Apostolorum, that is, derived its vulgar name from the porticated (via) Romana whence it was accessed, which in the popular imagination – as well as in material reality – was inextricably connected with the church itself, as was the memory of the relic procession that accompanied its dedication. Motivated by the prompting of the crowd, Ambrose exhumed the bones of the local martyrs Gervasius and Protasius, which he conveyed to the Ambrosiana in a triumphal procession directly inspired by the protocols governing secular adventus.³⁹ Nine years later, when Ambrose sought to exalt the remains of Nazarius, another local martyr, he sent them to the extramural church best architecturally equipped, by virtue of its monumental access road, to host yet another triumphal entry of relics: the basilica in Romana – Basilica Apostolorum, thereafter known also as S. Nazaro.⁴⁰

The chronology of the monumental porticoes built along the approaches to several of the leading churches in Rome and Milan in the 380s is particularly noteworthy given that evidence of imperial participation in processions held to mark the translation of relics begins with the reign of Theodosius I (379–95).⁴¹ According to Sozomen, in 391 Theodosius solemnly carried the head of John the Baptist to his new church dedicated in the martyr’s name, located outside the Golden Gate at the Hebdomon, the same place from which newly raised emperors departed on their ceremonial entrance into Constantinople.⁴² In 406, Theodosius’ son Arcadius joined the Patriarch of Constantinople and members of the Senate in transporting the relics of Samuel into the city.⁴³ In 411, these relics were deposited outside the new city walls at the church of St. John iucundianae at the Hebdomon, which took its name from the imperial palace of the same name located nearby.⁴⁴ During the reign of Theodosius II (408–50), the emperor and members of the imperial family regularly participated in translations of relics, as when, in 421, the remains of Stephen were paraded through the city, in a style befitting of an imperial adventus, on their way to their final resting place at the oratory of St. Stephen, newly built inside the imperial palace itself, at the eastern extremity of the Mese, the main processional artery in the city.⁴⁵ With the installation of the relics of both John the Baptist and Samuel at the Hebdomon, and those of Stephen at the palace inside the city, the starting and ending points of the imperial triumphal route along the Mese were bracketed by collections of newly installed relics, in the translation of which the imperial family had, moreover, actively participated. At San Nazaro, meanwhile, the relics of the eponymous martyr were ensconced in a chapel decorated with Libyan marble at the behest of the magister militum Stilicho’s wife, Serena, who prayed for her husband’s safe return from campaigning in the dedicatory inscription.⁴⁶ Thus, at Milan, too, the church that seems to have become the preferred stage for the display of court-sponsored patronage of the cult of relics was the one most closely connected with the monumental route (the very road, perhaps, along which Serena hoped to see her husband return in triumph?) followed by imperial processions.

Beginning in the later fourth century, then, the rulers of the empire were evidently increasingly eager to link their public personae with the triumphs of the Christian church, and they actively collaborated with church leaders in the celebration, veneration and architectural framing of the martyrs who, in triumphing over death, had preserved orthodoxy and placed the Christian Roman empire over which they presided under divine protection.⁴⁷ Ambrose himself, like many of his contemporaries, leaves no doubt that the arrival of relics was a triumphal occasion, best presented in the trappings of triumphal imperial adventus that would have been immediately familiar to the inhabitants of any imperial capital.⁴⁸ Enthroned on the four-wheeled cart (plaustrum) used to convey arriving emperors in the fourth century, preceded by torches and standards, and hailed by the acclamations of the masses, the bones of the illustrious Christian dead were greeted with reverence otherwise reserved for reigning emperors.⁴⁹ The transfer of relics to churches connected to the principal porticated streets of a late antique capital such as Milan could subsequently unfold in the same spaces used by the emperors for their own triumphal arrivals: a church procession through the porticoes of the Via Romana would inevitably have called to mind imperial processions along the same street. The connection between imperial and ecclesiastical ceremony thus received its definitive and lasting stamp in the realm of architectural space, in the form of a monumental processional way jointly exploited by civic luminaries and bishops. When the emperors themselves participated in translations of relics, the circle was closed: state and church triumphed together, as the representatives of each basked in the reflected glow of their counterparts. The imperial presence ennobled the ecclesiastical ceremony and proclaimed the unity of bishops and Christian sovereigns in the governance of the empire. The emperors themselves grew in the eyes of the Christian faithful, as the patrons and colleagues of the holy men tasked with stewarding the bones of the martyrs and mediating the access of the faithful to this most precious of all forms of spiritual currency.

3.3 Constantinople

The topographical nexus between imperial and ecclesiastical authority was nowhere more fully realized than in Constantinople, the ideal laboratory for the testing and elaboration of the urban paradigm envisioned by the first generations of Christian emperors. Officially inaugurated on May 11, 330, the birthday of its founder and namesake, the new city afforded Constantine and his successors in the fourth and fifth centuries a sort of architectural blank slate, onto which they were free to delineate an ideal vision of an imperial capital, largely unencumbered by the constraints of preexisting urban topography. In Constantine’s day, the new city center, built over and around the much smaller existing town of Byzantium, gravitated around an architectural core based on a single colonnaded main street, the Mese, that comprised the final tract of the principal land route leading to the city from Greece and the West, the Via Egnatia (Figure 3.3).⁵⁰ After traversing the Golden Gate in the new land walls, the visitor approaching from the west arrived first at the Capitol – prime symbol of the ‘old’ Rome and so too of Constantine’s ‘new’ Rome⁵¹ – and the nearby precinct of the Philadelphion, a colonnaded square filled with statues and commemorative monuments to the emperors (among them the porphyry Tetrarchs now in Venice).⁵² From the Philadelphion, the Mese continued to the east, lined with majestic two-story colonnades, as far as the circular forum of Constantine, itself surrounded by colonnades two stories high, and graced at its center by a porphyry column topped by a gilded statue of Constantine.⁵³

Figure 3.3 Constantinople. A: Forum of Arcadius; B: Apostoleion; C: Forum of the Ox/forum bovis; D: Philadelphion; E: Forum/column of Marcian; F: Forum of Theodosius; G: tetrapylon; H: Forum of Constantine; I: Baths of Zeuxippos; J. Hagia Sophia; K: hippodrome.

(Wikimedia Commons/Cplakidas, modified.)

The two-story colonnades of the Mese thence continued east toward the milion, the Constantinopolitan equivalent of the umbilicus in the Roman Forum, marked by a monumental arch, likely a quadrifrons, spanning the street, just behind which stood a statue of an elephant, the very symbol of imperial triumph that adorned the quadrifrons arch at Antioch.⁵⁴ To the north and south, the colonnades of a second avenue running perpendicular to the Mese converged on the archway, which thus stood, as at Antioch, Thessaloniki and Split, at the intersection of four radiating colonnaded axes, the shortest of which – again – led straight to the palace. This final stretch of the Mese, now wider and grander still and called – again – the Regia, stretched eastward from the milion to the main gate of the palace, known by the fifth century as the Chalke.⁵⁵ The form of the gate appears to have been rectangular, with a domed central chamber straddling the axis of the Regia and two lower flanking bays, very much like the Arch of Galerius at Thessaloniki.⁵⁶ North of the Regia stood the Augusteion, a colonnaded square flanked by a second senate house, joined under Constantius II (337–61) by the cathedral of Hagia Sophia and the palace of the patriarchs along its northern extremity.⁵⁷ The Constantinian cathedral of the city, Hagia Eirene, lay just beyond to the northeast.⁵⁸ To the south, the southern colonnade of the Regia gave onto the Baths of Zeuxippos and the carceres of the hippodrome, both rebuilt by Constantine and directly connected to the imperial palace itself, which sprawled south and east toward the shores of the Sea of Marmara (Figure 3.4).⁵⁹

Figure 3.4 The area of the Great Palace in Constantinople. A: Hagia Eirene; B: Hagia Sophia; C: Augusteion; D: Mese; E: Milion; F: Chalke Gate; G: Baths of Zeuxippos; H: hippodrome; I: pulvinar; J: Peristyle Court; K: Yerebatan cistern; L: Palace of Lausos.

(After Ćurčić 2010, fig. 190.)

While the archaeological evidence is scanty and leaves infinite scope for haggling over the details, the basic contours of the palatial quarter in Constantinople as it took shape beginning under Constantine are clear enough.⁶⁰ The complex gravitated around the Regia, itself framed at the west by the milion and at the east by the principal gate of the palace, the Chalke. The hippodrome and the greatest baths in the city (of Zeuxippos) stood just off the Regia, in the immediate vicinity of the entrance to the palace, and both in fact communicated directly with the palace to the rear, while their public entrances lay just off the colonnades of the street.⁶¹ By the fifth century (and probably earlier), the palace was also bodily attached to the cathedral of Hagia Sophia, via – yet another – colonnaded passageway, an elevated loggia by the sixth century, which connected the northern bay of the Chalke with the south flank of the church.⁶² Baths, hippodrome, arches, the imperial palace and the cathedral, all linked by the armature of a single, extraordinarily grandiose colonnaded avenue and further colonnaded extensions thereof: it is the culmination of the ‘imperial’ architectural paradigm that took shape in the third century, a cohesive ensemble that need not have exactly ‘copied’ any existing foundation to nonetheless reprise the essential features of all the imperial capitals and residences discussed in the preceding chapter, insofar as these can be known or surmised on the basis of the available evidence.

Later sources are surely right to suggest that already at the time of the dedication of the new capital in 330, the Mese with its associated monuments was conceived as a grand triumphal route. According to the eighth-century Parastaseis Syntomoi Chronikai, the founding ceremony on May 11 witnessed the transfer of a gilded statue of Constantine from the Capitol to the forum of Constantine, where it was laboriously raised, in front of the assembled people and dignitaries of the new city, up to its final resting place atop the porphyry column.⁶³ In the words of Franz Alto Bauer, ‘The transfer of the statue to its new location appears to have been staged as a triumphal entry of the emperor into the city. An escort wearing white chlamydes, holding candles, accompanied the statue, which stood erect in a chariot, from the Philadelphion to the Forum of Constantine. Thus was staged the adventus of the emperor who, present in his statue, proceeded along the main street of Constantinople and finally reached “his” forum.’⁶⁴

But while the transfer of the statue to the column was a one-time event, Constantine also made provision for a more lasting commemorative adventus, to be staged every year on the occasion of his birthday and the founding of the city. Again, the protagonist was (yet another) gilded statue of the emperor, ordinarily kept in a repository on the north side of the forum of Constantine, which every May 11 left the forum mounted on a triumphal chariot and processed down the Mese and into the hippodrome, where it made a lap of the track and came to rest in front of the imperial box (kathisma) to receive the obeisance of the reigning emperor.⁶⁵ ‘By means of this ritual, therefore, the memory of the founding emperor was awakened in the consciousness of the populace; gilded, like the statue on the porphyry column ... He received the acclamation of the populace, the aristocracy and the emperor, before returning to the Forum of Constantine.’⁶⁶

For more than a century following the death of Constantine in 337, succeeding emperors made it their task to extend the monumental framework of the Mese with a series of imperial forums and triumphal monuments arrayed along the main axis of the street. Theodosius I (379–95) built his forum between the forum of Constantine and the Capitol; the forum of Arcadius (395–408) took shape still farther to the east; and that of Theodosius II (408–50) rose between the old Constantinian walls and the new land walls of the city, built between 405 and 413.⁶⁷ There, the road exited the city via the Porta Aurea, much the most magnificent gate in the new circuit, which took the shape of a triumphal arch dedicated to Theodosius II.⁶⁸ New colonnades rose along the Mese in correspondence with the new forums, linking them all together in an unbroken progression of symbolically charged spaces designed to exalt the memory of their individual founders and propagate the timeless ideology of imperial victory upon which their claims to legitimacy and unchallenged sovereignty rested.⁶⁹

At the Philadelphion, the main tract of the Mese on its way to the Porta Aurea diverged from a second colonnaded avenue, which angled northwest toward the Charisios (or Adrianople) Gate in the Theodosian walls. The centrality of this northern branch of the Mese in the landscape of the new city had been proclaimed already under Constantine himself, who erected his mausoleum just to the north of the road shortly before it traversed the Constantinian walls.⁷⁰ The mausoleum, final resting place of the emperors through the reign of Anastastius I (d. 518),⁷¹ was flanked by the cruciform church of the apostles, the Apostoleion, probably begun in the 350s by Constantius and consecrated in 370.⁷² The fame of this church was such that Ambrose based the cruciform plan of his own Basilica Apostolorum at Milan on the Constantinopolitan exemplar;⁷³ and the location of Ambrose’s church at the midway point of the Porta Romana colonnade strongly implies that the Milanese bishop was attempting more broadly to replicate the architectonic context of the Apostoleion and to ensure that his new foundation was similarly privileged by virtue of its physical proximity to the main ceremonial thoroughfare of the city.

In the mid-fifth century, when all of the prime locations along the southern branch of the Mese had been occupied by imperial monuments, the emperor Marcian (450–57) was compelled to establish his own forum along the northern extension of the Mese, between the Philadelphion and the Apostoleion.⁷⁴ Thus, by the time of Marcian’s death, the architectonic imprint of the capital was largely complete. The most magnificent public buildings and forums in the city, as well as the churches of Hagia Sophia, Hagia Eirene and the Apostoleion, were either bisected by or proximate to the trunk of the Mese in the east and its two principal continuations in the west.⁷⁵ These expansive, colonnaded thoroughfares were the spinal cord of the city as a whole, the architectural profile that imprinted itself on the consciousness of visitors and residents alike, and the repositories of the institutional memory encoded in their flanking monuments. The Mese anchored the ceremonial processions that gave tangible form to the ascendancy of Constantinople’s civic and ecclesiastical authorities, and at the same time channeled the flow of quotidian life and commerce through the most ideologically charged spaces in the city, the forums and commemorative columns bedecked with statues, reliefs and inscriptions that inevitably called to mind the mighty who had commissioned them, and who animated them on important occasions with their superhuman presence.⁷⁶

When newly crowned emperors made their triumphal entrance into the city, as Leo I did in 457, or returned from extended absences, as Justinian did in 559, the entire population of the city, arranged sequentially and hierarchically according to rank and station, lined these streets and chanted their acclamations to the passing imperial cortege. On the most important days in the annual liturgical calendar, ecclesiastical processions followed the same route on their way to the leading shrines in the city.⁷⁷ When relics arrived, such as those of Samuel in 406, Stephen in 421, and John Chrysostom in 438, these too made their adventus into the city along the Mese and its northern and southern extensions, conveyed in regal splendor by patriarchs and reigning emperors, whose joint participation in these processions ostentatiously proclaimed the ideal unity of church and state in a polity in which these two institutions grew ever more inseparable.⁷⁸

Already in ca. 400, the presence of the emperor Arcadius and his wife Eudoxia featured prominently in the sermon John Chrysostom delivered upon the arrival of the relics of the Pontic martyr Phocas to Constantinople, in which, moreover, the physical essence of the city was distilled into a vision of its grand colonnades: ‘The city became splendid yesterday, splendid and illustrious, not because it has columns, but because it hosted the parade of the arriving martyr, who came to us from Pontus.’⁷⁹ The newly arrived relics may have glorified the city as the colonnades alone never could have, but it was nonetheless the image of the colonnaded street that provided, in the mind of the bishop, the essential architectural backdrop for the solemn festivities that bound Arcadius and Eudoxia together with the clergy and the urban masses in joint veneration of Phocas’ mortal remains.⁸⁰

3.4 Porticated Streets and the Literary Image of Late Antique Cityscapes

One of the great strengths of porticated streets was their capacity to channel the flow of both quotidian and ceremonial life within their confines: to direct movement along a limited number of privileged itineraries designed to highlight key civic monuments and connect them with grand colonnaded façades in a seamless topographical ensemble. The scattered civic monuments of the high imperial period, many of them falling into disrepair, and the teeming residential neighborhoods of the lower classes could be selectively filtered from view behind the lofty screens of porticoes lining the main streets of the late antique city. The testimony of contemporaries writing in the fourth and fifth centuries indicates that the urban experience was increasingly coming to be distilled into the visual and spatial parameters delineated by porticated avenues, which had become the essence of urban grandeur, the token by which a great city could be recognized and most succinctly defined.

One such indicator comes from the Notitia Urbis Constantinopolitanae, the extensive catalogue of urban topography in Constantinople compiled by an anonymous author writing ca. 425. While street colonnades are punctiliously listed in all of the fourteen chapters devoted to the various regions of the city, particularly evocative is the description of the seventh region, located in the ceremonial heart of the city, extending north from the Mese, between the forum of Theodosius and the forum of Constantine, as far as the shores of the Golden Horn. The summary description of the region that precedes the detailed listing of its component structures reads thus in its entirety:

For all that the seventh region was populous and densely built, it is the colonnaded sweep of the Mese between the forums of Constantine and Theodosius I, with its colonnaded cross-streets, that dominates the topographical panorama sketched by the author, and shapes the visual profile of the region as a whole. Visitors and residents alike might traverse the entirety of this central district of the capital without experiencing it as anything other than a continuous succession of stately colonnaded façades; the glorious illusion would give way to a more jumbled and heterogeneous reality only for those whose business carried them beyond the main roads.⁸²

In his list of the twenty greatest cities of the empire, written ca. 380, Ausonius includes a similarly succinct description of the city of Milan, which concludes with reference to ‘her colonnades (peristyla) all adorned with marble statuary, her walls piled like an earthen rampart around the city’s edge.’⁸³ The juxtaposition between city wall and colonnades is particularly striking, as it was the gates in the wall that framed the principal streets in the city, including the Via Romana with its grand new arcades, themselves under construction or very recently completed at the time when Ausonius was writing. For the visitor approaching from the south, Milan will indeed have been experienced precisely as the intersection of the porticated street and the city wall, which between them channeled the flow of traffic onto a single axis and screened from view all of the less monumental quarters of the city.

The predominance of porticated vistas in the imagination of late antique writers appears to herald a real shift in prevailing views, or mental images, of urban topography, for all that a similar inclination to condense the essence of urban civilization into colonnaded streets occasionally occurs earlier, notably in Achilles Tatius’ second-century description of Alexandria.⁸⁴ In his lament for the martyrs of the Diocletianic persecution in Palestine in 309, Eusebius turns the morning dew on the colonnades of Caesarea into tears of sorrow wept by the city for its fallen: ‘The air was clear and bright and the appearance of the sky most serene, when suddenly throughout the city from the pillars which supported the public colonnades (stoai) many drops fell like tears; and the marketplaces and streets (agorai te kai plateiai), though there was no mist in the air, were moistened with sprinkled water.’⁸⁵ For Eusebius, the columns of the street colonnades were evidently the most pervasive and characteristic feature of the cityscape, the architectural element best suited to express the mourning of the entire city: they were the eyes and the wounded heart of Caesarea personified. So too in the mind of the anonymous mid-fourth-century author of the Expositio totius mundi et gentium, the grandeur of Carthage is apparent in its regular grid of streets, which in their ordered symmetry called to his mind the image of an orchard.⁸⁶ The columns lining the streets make the metaphor: in the urban orchard of Carthage, the rows of columns take the place of trees as the dominant visual element. Likewise in the imagination of John Chrysostom, as we have seen, the columns of Constantinople became a synecdoche for the architectural grandeur of the city as a whole.⁸⁷

But by far the longest and most explicit surviving account of the place of colonnaded streets in the physical, social and mental geographies of the late antique metropolis comes in Libanius’ Oration 11, the Antiochikos.⁸⁸ The text concludes with a topographical excursus on the city that runs to some ten pages in Downey’s edition, nearly half of which centers on the form, the beauty, and the function of the city’s colonnaded avenues, its grand central axis above all.

For Libanius, Antioch is a great city above all because of its colonnaded streets, which are both beautiful and the essence of civic life.⁹⁰ The main street is quite literally the architectural centerpiece of the city, the conduit that in turn leads to its grandest public buildings, all directly accessible from its flanking colonnades.

Nor was Libanius alone in judging Antioch’s wide, colonnaded streets the epitome of its urban glory. When John Chrysostom was still a priest at Antioch, he delivered a sermon during the ‘affair of the statues’ in 387, while the Antiochenes were anticipating with dread Theodosius’ response to their destruction of imperial images near the palace on the Orontes. In what was almost certainly a conscious evocation of his despised former teacher Libanius’ Antiochikos, Chrysostom pointed out to his congregation that the true beauty of their city, whatever became of it as a result of the emperor’s wrath, lay not in its wide, colonnaded streets, but rather in the virtue and piety of its (Christian) inhabitants.⁹¹ As with his later sermon on the arrival of Phokas’ relics at Constantinople, what Chrysostom’s effort to subordinate colonnaded streets to the greater glories of the Christian faith really shows is just how synonymous these streets had become, by the later fourth century, with the concept, the idea and the reality, of a leading Mediterranean metropolis.

But to return to Libanius, his lengthy encomium is also noteworthy for being the only extant account that makes a concerted effort to explain why colonnaded streets were so central to the configuration of the ideal cityscape:

Rather than treating them only as essential elements of urban décor, that is, Libanius considered the stoai of Antioch in terms of their functionality; of the uses to which they were put, and the activities that unfolded within and around them.

In addition to keeping the winter rains off the Antiochenes and allowing all members of the urban collective to mingle and conduct business even in the depths of winter, Libanius’ colonnades were the teeming commercial heart of the city, packed with industrious craftsmen and vendors who filled the spaces between the columns and spilled out onto the street:

His words are a valuable reminder that colonnaded streets were lived space, fulcrums of everyday living, as much as settings for grand displays on special occasions. Antioch’s main roads were both beautiful and bustling, grand architectonic vistas and the center of daily life. Here is the death knell for attempts to apply Sauvaget’s vision of the ‘devolution’ of the colonnaded street into the suq to the end of antiquity:⁹⁴ long before the alleged disintegration of effective civic government and the ‘privatization’ of public spaces in the sixth century, the colonnaded thoroughfares of the great cities of the eastern Mediterranean already teemed with shops and commercial activity.⁹⁵ Indeed, it was their mundane role as much as their ceremonial profile that made such streets so desirable in late antiquity, and led them to become more prominent than ever before on the urban landscape.

3.5 Commerce, Commemoration and Ceremony in the Colonnades of the Eastern Mediterranean

Even if Libanius’ hyperbolic assertion that Antioch was unique for the amount of activity in its colonnades were true, his depiction of the throng of shops filling the intercolumniations of Antioch’s main streets would be nonetheless noteworthy. Antioch was the capital of the diocese of Oriens and seat of both the Praetorian Prefect of the East and the comes orientis, a place where the reach of the imperial administration and the ceremonial regime governing the public deportment of high officials were particularly robust.⁹⁶ A perusal of Libanius’ voluminous corpus suggests that work on street colonnades at Antioch was almost the exclusive preserve of ranking members of the imperial administration, chiefly serving governors or ex-governors of proconsular rank resident in the city, who in fact seem to have made colonnaded streets their first priority in the realm of public building.⁹⁷ The crucial point is that, their immense prestige value aside,⁹⁸ these colonnades were envisioned from their inception as grand commercial arcades, useful for producing revenues for the imperial servants who built them as well as ennobling the spaces where they paraded about. When the ex-governor Florentius widened a street and lined it with a colonnade in 392, he expected to recoup his costs (and far more, according to the hostile Libanius) from the rents of the shops located behind the columnar façade.⁹⁹ When the provincial governor Tissamenes had a street colonnade repainted (!), he rewarded the painters and defrayed his own costs by compelling the commercial tenants of the structure to have their shop signs done by the same painters, presumably at substantial cost.¹⁰⁰ Far from a symptom of the collapse of effective civic authority, then, the ‘encroachment’ of commercial activities on the public colonnades of Antioch is better construed as an indicator not only of economic vitality, but also of the immense power and local influence of the imperial representatives who built and maintained the colonnades. It is also a sure sign that there was no profound incompatibility between the dual roles of Antioch’s stoas as lived space on the one hand, and ceremonial space on the other.

Further, it turns out that the situation in Antioch was far less anomalous than Libanius would have us believe. Both archaeological and textual evidence makes it clear that in other provincial and imperial capitals in the eastern Mediterranean, the colonnades of the main streets were similarly packed with vendors’ stalls and workshops, and continued nonetheless to preserve the monumental profile upon which their role as privileged theaters of ceremonial life depended.¹⁰¹ In the fifth century, we find the central administration in Constantinople legislating with the express intent of reconciling the exigencies of commerce with the equally pressing need to maintain the architectural decorum of the Mese: a decree addressed by the emperor Zeno to Adamantius, praefectus urbi of Constantinople in 474–79, prescribes in minute detail the proper configuration of shops along the colonnades of the Mese, from the milion at its eastern extremity as far as the Capitolium.¹⁰² The façades of the intercolumnar stalls were to be no more than six feet wide and seven high, and to be revetted with marble on their exterior facings, ‘that they may be an adornment for the city and a source of pleasure for passersby.’¹⁰³ Zeno’s special concern for the principal ceremonial avenue in the city, the route indelibly associated with his public appearances, is manifest in the further specification that the disposition of shops in all other city colonnades fell under the purview of the city prefect.¹⁰⁴ The décor of the city’s ‘imperial’ street alone required the direct supervision of the emperor himself. Archaeology reveals that similar aesthetic concerns prevailed at Ephesus, where the shop fronts installed in the intercolumnations of the main processional way (the Embolos) were covered with marble veneer on their external facings in the fifth and sixth centuries.¹⁰⁵

Ephesus in fact provides, I would suggest, the closest thing possible to a canonical example of the intertwining of social and economic imperatives with ceremonial agendas in the topographical evolution of a late antique metropolis in the eastern Mediterranean. Capital of the diocese of Asia after Diocletian and later seat of a metropolitan bishop, it is one of the best excavated and most thoroughly studied of the empire’s leading cities. Its extant remains furnish an unusually detailed perspective on the evolving patterns of human activity, and even of the changed mentalités underlying them, that so profoundly conditioned the physical parameters of the urban environment and imbued them with their characteristically late antique features, chief among them a colonnaded street and its associated monuments.

Between the fourth century and the seventh, the city gravitated ever more closely around a single main street, composed of three intersecting segments on different orientations, flanked along nearly all of its length by colonnades built in the fourth and fifth centuries (Figure 3.5). Throughout this period, the majority of investment in public architecture and urban infrastructure occurred in the environs of this central axis, beginning with the paving and colonnades of the road itself, and extending to the public monuments – squares, triumphal arches and columnar monuments, nymphaea, the theater – and the opulent private houses that lined its course.¹⁰⁶ Those approaching the city center from the direction of the harbor (as most distinguished visitors arriving from afar will have done) first embarked on the Arkadiane, the wide, straight promenade between the harbor and the theater, which was monumentally re-edified ca. 400 with 600 meters of continuous colonnades along both sides of the street, sponsored by Emperor Arcadius, from whom it took its name.¹⁰⁷ A sharp right at the theater led to the ‘Marble Street,’ running past the scaena of the theater and the lower agora and on to the old library of Celsus, rebuilt as a grand nymphaeum, also ca. 400; there, the road turned sharply again to become the Embolos (or ‘Curetes Street’), leading upward from the façade of the nymphaeum toward the upper agora, beyond which lay the principal gate in the land walls, the Magnesian Gate (Figure 3.6).¹⁰⁸ As with the Arkadiane, the Marble Street and the Embolos were maintained in grand style into the early seventh century: the solid pavers of the roadbed were kept whole and unobstructed, the colonnades assiduously restored whenever necessary, the floors of the covered sidewalks repaved with new mosaics and sumptuous marble panels, the shop fronts revetted with marble.¹⁰⁹

Figure 3.5 Central Ephesus in late antiquity. A: Harbor-Gate; B: Baths of Constantius; C: Church of Mary; D: palace; E: Arkadiane; F: Theater Gymnasium; G: Temple of Serapis; H: Lower Agora; J: Library of Celsus; K: Embolos; L: Gate of Herakles; M: Upper Agora; N: to Magnesian Gate. Dotted line marks course of the Arkadiane-Embolos route.

(After Bauer Reference Bauer1996, modified.)

Figure 3.6 The Embolos today.

(Photo: ArtStor IAP/Mellink Archive, Bryn Mawr College.)

This continuous monumental armature comprising the three sections of the road and its associated structures attained unprecedented heights of architectural grandeur, beginning in the fourth century, because the governors who thenceforth controlled the funds available for building chose to make it the architectural showpiece of the city as a whole, and to recognize its de facto role as the epicenter of civic life. As Franz-Alto Bauer has shown, by the fifth century, the Arkadiane–Marble Street–Embolos axis had largely supplanted traditional civic spaces, such as the upper and lower agorai, as the venue of choice for the display of statues and inscriptions honoring the leading patrons and benefactors of the city and the late Roman state.¹¹⁰ City councilors and private euergetes are almost nonexistent: the dedicatees of the dozens of fourth-to-sixth-century statues and inscribed bases found among the colonnades and along the street overwhelmingly represent the emperors and members of the imperial family on the one hand, and governors and other representatives of the central administration on the other.¹¹¹ This is of course to be expected, given that imperial agents were chiefly responsible for the upkeep of public architecture and infrastructure from the fourth century on.¹¹² What is remarkable is the extent to which the colonnades and other monuments flanking the road were preferred to all other buildings in the city for the display of honorary monuments. That they were so privileged strongly suggests that there was no better place to be honored and immortalized for posterity, and thus that the city’s main processional avenue had become the heart of the urban collective, the place where the images of the emperors and their local delegates were most effectively brought before the eyes of the urban populace as a whole; where the ubiquitous presence of the imperial establishment was driven home in the constant succession of honorary inscriptions and stone faces gazing out from the shadow of the colonnades.

Beginning in the fourth century, these porticoes also became the venue in which the will of the state was made manifest in a more immediate and explicit way, in the form of the laws and edicts promulgated by the emperors and their representatives, as Denis Feissel has shown. Until the third century, the inscribed texts of such decrees were regularly displayed at the locations where the popular assembly and city council met: the theater, the odeon, the bouleuterion and the upper agora.¹¹³ After the third century, there is not a single example of an official decree posted at any of these places, a clear indication that the traditional organs of civic government were no longer responsible for their publication.¹¹⁴ Of the twenty-five inscribed edicts datable between the mid-fourth century (the earliest belongs to the reign of Constantius II) and the end of the sixth whose original location is well established, fourteen adorned the columns, balustrades and walls of the colonnades lining the Embolos and the Marble Street.¹¹⁵ Feissel’s meticulous study even suffices to trace the migration of newly inscribed acts over time: by the sixth century, when the available surfaces of the Embolos were evidently jumbled with official documents, the inscribers moved on to the adjacent stretch of the Marble Street, covering its porticoes too with a dense patchwork of documents.¹¹⁶ In the process, the colonnades were transformed into an indelible testament to the pull of the imperial will on the lives of the Ephesians, an archive of centuries’ worth of official pronouncements imposed on the consciousness of the masses who traversed the high street of the city on a daily basis.¹¹⁷

The commemorative monuments and legal texts clustered along the Marble Street and Embolos, taken in their entirety, testify eloquently to the centrality of this axis in the minds of the authorities responsible for commissioning them. The profusion of shops in the colonnades indicates that, as at so many other late antique cities, the main street was bustling with commercial activity.¹¹⁸ The numerous game boards inscribed in the pavements of the colonnades at Ephesus and numerous other cities reveal them as places of leisure and social gathering as well.¹¹⁹ The colonnades of the main street at Ephesus, in short, had become the most frequented space in the city, a development that also helps to explain why they were the preferred location for conspicuous forms of epigraphic and artistic display.

But why? Why, when emperors and their representatives became the prime movers behind the shaping of urban infrastructure at Ephesus in the fourth century, did they so assiduously maintain and embellish the architectural décor of a single colonnaded thoroughfare, which progressively subsumed so many of the commercial, commemorative and administrative functions previously more widely diffused throughout the markets, agorai, assembly halls and entertainment venues of the high imperial city?

As at the imperial capitals – Trier, Rome, Milan, Thessaloniki, Constantinople – discussed earlier, so too at Ephesus, I think the answer lies to a considerable extent in the exigencies of public ceremony. The colonnaded processional way was the conduit through which the authoritarian juggernaut of the late Roman state could most visibly and efficiently permeate the heart of the urban center. Not merely grand architectural statements punctuated by honorary monuments to the ruling establishment (which they certainly were), these streets became the main stage for the living tableaux placed before the urban populace, dutifully assembled and arrayed along the porticoes on, for example, the several occasions each year when an incoming governor either made his first adventus into the city, or returned from regular visits to the rest of the province.¹²⁰ It is for these ephemeral activities, so memorable for participants and so hard for modern scholars to trace in the material record, that the rich corpus of epigraphic evidence from Ephesus is so unusually revealing.

Dozens of extant inscriptions datable from the fifth century into the seventh contain verbatim transcripts of acclamations shouted out by the crowds lining the processional route on special occasions. Like the inscribed edicts discussed earlier, they cluster almost exclusively along the Marble Street and the adjacent section of the lower Embolos. ‘Many years for Christian emperors and Greens!’ ‘Many years for pious emperors!’ ‘Many years for Heraclius and Heraclius, our god-protected lords, and for the Greens!’ ‘Heraclius and Heraclius, our god-protected lords, the new Constantines!’ ‘[Lord] help Phokas, crowned by God, and the Blues!’¹²¹ As Charlotte Roueché has said, many of these inscriptions must reproduce the wording of real chants, and they should often be understood to mark the very spot where the recorded words were uttered on one or more occasions, in addition to the identity of the speakers:¹²² a group of supporters of the Green faction very likely filled the colonnades of the Marble Street where their acclamations are incised, for example, just as the Blues who supported Phokas must have done along the lower Embolos where their chant appears.

A particularly eloquent evocation of a particular time and place in the ceremonial life of the city comes from an inscribed block reused in the paving of a platform leading to the northwest entrance to the theater, recently published by Roueché, which preserves the roar of the crowd as it hailed the adventus of the new proconsular governor of Asia, Phlegethius, into the city around the year 500: ‘Enter, Lord Phlegethius, into your city!’¹²³ The position of the reused block near the intersection of the Arkadiane and the Marble Street suggests that it originated along one of the two streets, and thus on the route the entering governor would have taken on his way into the city from the harbor at the foot of the Arkadiane, where those who uttered the acclamation stood.

Another inscription preserved in situ within the colonnades of the lower embolos, just past the nymphaeum at the library of Celsus, graphically illustrates the rootedness of specific individuals in the ceremonial topography of the processional way: ‘(this is) the place (topos) of the meleparchos Andreas’ (Figure 3.7).¹²⁴ Similar topos-inscriptions are attested in late antique urban contexts elsewhere in the eastern Mediterranean, where they are generally interpreted as place markers for shopkeepers and other purveyors of services, some of whom are clearly identified as such in the texts.¹²⁵ In this case, however, the mention of Andreas with his official government title (meleparchos) suggests another interpretation: this is the spot where Andreas, in his capacity as a representative of the civic administration, took his place on ceremonial occasions, when the ranks of spectators arrayed themselves along the parade route according to their rank and station. Whatever the precise duties of the ‘meleparch’ (a hapax, as far as I know), it is evidently the title of a representative of the urban administration, whose official quarters are unlikely to have been sandwiched in among the shops adjacent to the place along the Marble Street where his inscription occurs. We might speculate further: the urban prefects of Constantinople were later known as eparchs; their responsibilities included oversight of all the logistical arrangements necessary for triumphal processions;¹²⁶ and the tenth-century De caerimoniis aulae byzantinae places the eparch and his staff (the taxis eparchou) along the route of Justinian’s adventus of 559.¹²⁷ It is thus tempting to imagine that in sixth-century Ephesus, the meleparch was involved in the organization of urban processions, whence it would be particularly appropriate that his ‘place marker’ lies along the heart of the processional route.

Figure 3.7 Topos-inscription of the meleparch Andreas.

(Courtesy of F. A. Bauer.)

The wording of the inscription is suggestively echoed in a sermon delivered by Bishop Proclus of Constantinople (434–46), whose passing reference to the protocols surrounding imperial adventus constitutes the most explicit extant testimony to the existence of what we might call stational chants:

For Proclus, particular locations along the processional route were associated with specific chants, each of which, like the meleparchos Andreas himself at Ephesus, had its proper place (kata topous). The further implication is that these chants were voiced by discrete subgroups among the urban populace, each of which had its assigned place along the Mese on those occasions when official processions passed along its colonnades.

So much is in fact clear from the literary sources, which show that at Constantinople and elsewhere, the citizenry assembled along the processional route, subdivided by rank and profession, according to a prescribed sequence. According to Libanius, when a new governor made his adventus to Antioch, his route as far as the city gate was lined by senators and ex-governors, current members of the governor’s staff (officiales), local city councilors, and then lawyers and teachers, in that order. Upon traversing the gate, he received the acclamations of the masses, doubtless including the proprietors of the shops lining the street, who thronged the colonnades along his route to the palace on the Orontes.¹²⁹ When Justinian made his triumphal entrance into Constantinople after the retreat of the marauding Kotrigurs in 559, the Mese from the Capitol all the way to the palace was lined with various corps of government officials, followed by ‘the silversmiths and all the artisans, and every guild.’¹³⁰ In the tenth century, shopkeepers and craftsmen still clustered according to profession along the colonnades of the Mese, which they presumably helped to decorate on ceremonial occasions.¹³¹

The example of Constantinople thus suggests also a substantial degree of interpenetration between ceremonial and commercial topography: the ranks of shops, revetted in marble and decorated by their proprietors on festive occasions, must also – at least in some cases – have been the very place where their occupants stood and voiced their acclamations to passing dignitaries. The section of the Mese called the ‘portico of the silversmiths,’¹³² then, would not merely have been where the silversmiths plied their trade, but also the place where they participated, as a corporate entity, in events such as the arrival of Justinian in 559, where they feature so prominently in the description preserved in the De caerimoniis.

Serious consideration might then be given to the idea that some of the many topos-inscriptions identifying individuals by name, with or without further information about their rank and profession, may relate to their placement on ceremonial occasions. The phylarch Eugraphius and the ‘most eloquent John,’ for example, whose topos-inscriptions appear on the north portico of the South Agora at Aphrodisias,¹³³ seem more likely to have laid claim to this space in their capacity as distinguished representatives of their city than as permanent occupants of the colonnade. Similar concerns may also better explain the function of the many inscriptions located in close proximity to circles inscribed on pavements at both Ephesus and Aphrodisias, which Roueché has already been inclined to link to the positioning of spectators in attendance at public ceremonies.¹³⁴ The circles, after all, have no appreciable connection to any sort of commercial activity, but would serve admirably to delineate the space to be occupied by a single, standing individual.¹³⁵ A remarkable graffito found near the tetrapylon at the crossing of two main roads at Aphrodisias indeed appears to show a figure with one foot in such a circle, arms raised in what might be a gesture of acclamation.¹³⁶

In any case, while the function(s) of such topos-inscriptions remains open to question, the large corpus of inscribed acclamations alone suffices to demonstrate the transformation of the Arkadiane-Marble Street-Embolos route at Ephesus into a living archive, a repository of institutional memory that ensured that the echo of the ephemeral chants voiced from its colonnades never died away. The memory of the choruses that accompanied the arrivals and other public appearances of the mighty was encoded in the fabric of the street itself, which proclaimed itself, in the voices of the people who assembled there to salute their rulers, a triumphal monument to the imperial establishment. Numerous inscribed acclamations documented at other sites, such as those present on the tetrapylon and adjacent colonnades at Aphrodisias, for example, and others by the propylon at Magnesia-on-Meander, and still others at Phrygian Hierapolis indicate that the main streets of regional centers across the eastern Mediterranean (and almost certainly beyond) experienced a similar transformation in late antiquity, for all that most of the evidence is either irretrievably lost, or awaiting identification by researchers more sensitive to its presence.¹³⁷ Traces of much more ephemeral painted texts of acclamations, moreover, indicate that the extant inscriptions represent merely the tip of an immensely larger iceberg: by the sixth century, the porticoes along principal urban thoroughfares must have teemed with writing to an extent that is today almost unimaginable.¹³⁸

The processional avenues that bisected the heart of the leading political and administrative centers of the late empire became, in short, the single most potent architectural manifestation of an ideal consensus between rulers and ruled. They were the stage that framed and reified the social order of the urban collective, its constituent parts hierarchically arranged – and perhaps even prescriptively oriented by inscribed place markers – festively attired, chanting pledges of allegiance to the emperor, the governor, Christian orthodoxy, Christ, Mary, their bishop, their circus-faction and/or the fortune of their city.¹³⁹ When the residents of a late antique metropolis lined the colonnades along the main processional way to witness the passage of their leaders, they intermingled with the statues of generations of emperors and imperial officials, whose edicts and proclamations peppered the columns, interspersed among the acclamations voiced by the citizenry in honor of these officials on innumerable past occasions, the cumulative legacy of which was evoked by, and integrally connected with, the procession unfolding before their eyes. On all the other days of the year, the profusion of written texts and honorary monuments served as a constant reminder for all comers of those special days when the leaders of the late Roman state put the spectacle of their might and the grandeur of the institutions of government they represented most visibly on display to their assembled subjects, by day and even – perhaps to a still greater extent – by night. The brilliant lighting of the main streets at (inter alia) Antioch, Ephesus and Constantinople must only have increased their relative prominence in the cityscape, and enhanced their propensity to attract and direct the flow of passersby.¹⁴⁰

The closely linked imperatives of commemoration, display and ceremonial praxis, in short, deserve pride of place in any attempt to explain why colonnaded streets were so privileged relative to other types of public building in late antique cities, and why, beginning in the Tetrarchic period, imperial and provincial capitals witnessed so much more building activity than the cities that remained more peripheral to the administrative apparatus of the state. Grand processional avenues proliferated in large part for their unique capacity to transform urban landscapes into scenic backdrops, majestic tableaux expressly intended to enhance the public appearances of emperors and their provincial representatives who commissioned and – directly or indirectly – funded them, and to distill the topographical profile of the places they inhabited into a narrowly circumscribed monumental itinerary.

With the growth of the Christian church, the wealthiest and most influential bishops, who themselves tended to be based in the secular capitals in which the metropolitan structure of the church was rooted,¹⁴¹ came almost inevitably to treat these porticated armatures as the preferred venue for the expanding battery of liturgical ceremony over which they presided, and often to annex the churches and episcopal residences they commissioned to the same streets.¹⁴² In the fifth and sixth centuries, when bishops took an increasingly active interest in the infrastructure and architectural patrimony of their cities, colonnaded streets indeed featured prominently in the list of construction projects they sponsored. Theodoret, for example, when listing the most significant architectural commissions he undertook as bishop of Cyrrhus in the second quarter of the fifth century, twice cited colonnaded streets in the first place, followed by mention of two bridges, baths and an aqueduct, all underwritten with money from church coffers.¹⁴³ While the colonnades will undoubtedly have pleased the shopkeepers and passersby who frequented them, and proclaimed the bishop’s stature as a leading patron of his city, one wonders whether Theodoret did not also hope to endow the rather mediocre capital of his remote provincial see with a more distinguished architectural profile, one better suited to magnify the public appearances of a leading figure in the ecclesiastical politics and theological controversies of the age.¹⁴⁴ Similar motives may also help to explain why other fifth- and sixth-century bishops chose to embellish their cities with street colonnades; their evident predilection for such structures surely suggests that church dignitaries, like their counterparts in the imperial administration, considered them key features of the urban landscapes they inhabited.¹⁴⁵

The extant image that perhaps best captures the role of colonnaded streets as the vibrant center of public life in the late antique city in fact depicts a religious procession, with two bishops for protagonists. This is the ‘Trier Ivory,’ a relief panel from a reliquary produced in Constantinople, variously dated between the fifth century and the ninth (Figure 3.8).¹⁴⁶ The event depicted on the panel has been interpreted most plausibly to represent the translation of the relics of the protomartyr Stephen, discovered near Jerusalem in 415 and triumphantly carried into Constantinople in 421, where they were installed in a new church built inside the imperial palace and dedicated to the saint whose remains it was to house.¹⁴⁷ Two bishops mounted on a chariot at the left of the panel convey the reliquary casket toward a church that workers are scurrying to complete, before the doors of which a bejeweled woman of imperial rank waits to receive them.¹⁴⁸ The center of the scene, however, is dominated by the profile of a two-story colonnade, packed with ranks of spectators gazing at the procession unfolding before them. Whether the colonnades represent the final section of the Mese, or the interior of the palace (or perhaps a conflation of the two),¹⁴⁹ they are manifestly the glue that holds the image together: for the creator(s) of the casket, the physical fabric of the city condensed itself into the image of a processional way; its inhabitants to the multitude that filled both levels and the roof of the colonnade.¹⁵⁰ In both architectural and human terms, ‘Constantinople’ is epitomized by the vision of a colonnaded street filled to bursting with beholders, without whose presence the elaborate spectacle engineered by the leaders of church and state would have been an exercise in futility.

Figure 3.8 The ‘Trier Ivory.’

(Photo: author.)

3.6 The Long Shadow of Constantinople, I: New Cities and Old Traditions in the Sixth-Century East

A particularly telling indicator of the enduring prominence of colonnaded streets in the minds of the leading architects of early Byzantine urban topography – again, chiefly the emperors and their provincial representatives – comes from the various cities founded ex novo as late as the sixth century, which, like Constantinople, afforded their designers extensive freedom to map out the contours of an ideal cityscape in a single, cohesively planned and executed initiative. Among the dozens(!) of examples of new cities founded in the eastern Mediterranean between the reigns of Anastasius and Justinian,¹⁵¹ Zenobia stands out for the quality of its extant remains, which furnish a reasonably complete picture of its basic configuration. Located along the Euphrates, on the strategically vital frontier with Sassanian Persia, the city took shape in the 530s and 540s under Justinian, though work may have begun already under Anastasius.¹⁵² The topographical imprint of the site leaves no doubt about the vision of urban infrastructure shared by the planners tasked by the central administration in Constantinople with designing it, whom Procopius names as John of Byzantium and Isidore of Miletus the younger.¹⁵³ The city gravitates around two continuous, colonnaded streets that lead directly from the main gates in the city walls to a central crossing point, punctuated by a tetrapylon, where they meet at right angles (Figure 3.9).¹⁵⁴ The principal churches and the praetorium are located in close proximity to these streets. The result was yet another urban armature made up of a continuous sequence of monumental architecture centered around colonnaded thoroughfares. All comers approaching from the outside experienced Zenobia first as an uninterrupted panorama of walls and towers, converging on the massive fortified gate through which they were required to pass; upon traversing the gates, they were funneled onto one of two wide, colonnaded streets that in turn led directly to the intramural foci of civic and ecclesiastical life. The most distinguished residents and visitors presumably need not have seen much of anything else, either during the course of their arrival/adventus or at any point thereafter.¹⁵⁵

Figure 3.9 The ‘new city’ of Zenobia on the Euphrates.

(Author.)

Another example of sixth-century urban planning that merits particular attention is the site of Justiniana Prima, the new city erected by Justinian on the site of his birthplace at Caričin Grad, in modern-day Serbia. The tiny size of the intramural nucleus – it is some 600 meters in length, barely more than 100 meters wide over much of that extent, and encloses an area of about 7ha – belies its immense institutional prestige: by 535, Justinian had made it the capital of an independent archbishopric covering the northern half of the prefecture of Illyricum,¹⁵⁶ and probably the seat of a praetorian prefect as well.¹⁵⁷ The city proper is formed of three interconnected, walled nuclei perched along a narrow spit of elevated terrain: a roughly circular citadel enclosing the cathedral, baptistery, episcopal audience hall and perhaps residence; an upper city that contains at least three churches and a number of administrative buildings; and a lower city with several more churches, a bath complex and a tiny residential quarter (Figure 3.10).¹⁵⁸ The full length and breadth of the intramural space was traversed by a total of two streets, lined along their full extent with arcaded porticoes, which converged at an oblique angle on a circular plaza in the upper city.¹⁵⁹ The enclosed perimeter was so narrow that there was generally room for only one rank of sizeable structures in the space left between the circuit-wall and the rear of the porticoes, whence all of the principal buildings fronted one of the two streets.

Figure 3.10 Justiniana Prima (Caričin Grad).

(After Ćurčić 2010, fig. 217.)

The walled center of the city, in other words, primarily consisted of two porticated streets, along with the single files of large, solidly built complexes – almost exclusively pertaining to government and church – visible and directly accessible from those streets.¹⁶⁰ It was a processional itinerary with a stately façade, artfully arranged to appear larger and more ‘urban’ than it really was. Both streets made a slight bend about midway along their course, making it impossible to see how short they really were, and providing the illusion of endlessly receding arcades. From the outside, the walled profile of the city likewise appeared to extend indefinitely for anyone approaching from the south, along the principal road that connected it with the outside world: on both sides of the narrow façade of the lower city wall, comers saw only the swelling profile of the upper city and citadel, their bristling walls sweeping upward and outward along the crest of the hill, giving no sign that the enclosed space ended just past the point where the circuit reached its greatest breadth. The presence of three distinct circuits and, more to the point, three sets of gates along the main axis of traffic is also remarkable, the more so if the wall of the lower city was an original part of the city plan, as Bavant has proposed, and not a later addition.¹⁶¹ Regardless, anyone bound for the citadel occupied by the newly created metropolitan archbishop of northern Illyricum would have traversed three bristling city gates in the space of some 500 meters. The considerable ceremonial potential of so many gates aside, the effect was surely to make the city seem far larger, grander and more impregnable than it was, a fiction that will have been far more convincing when experienced from street level than on the modern observer’s scaled city plan.

‘Small, fortified, Christian and imperial,’¹⁶² Justiniana Prima embodies the essential characteristics of the new cities founded on orders from Constantinople in the sixth century, though it is dominated to an unparalleled extent by the profile of its porticated armature, to which the entirety of the built space inside the walls is little more than an appendage. It is unlikely to be pure coincidence that the one sixth-century city destined from its inception to attain the highest rank in the administrative hierarchy of empire and church was also the one designed to be little more – and no less – than a parade route situated in the midst of what, at first glance, might plausibly have been a great city.¹⁶³ The population that filled the neighborhoods flanking the Mese at Constantinople – the epitome of a great city – was almost nonexistent within the walled confines of Justiniana Prima,¹⁶⁴ yet from street level, the difference will hardly have been apparent; the absence of the masses was as effectively concealed by the porticoes of Justiniana Prima as was their teeming presence by those of Constantinople. In reality, the inhabitants of Justinian’s magniloquent urban gesture were almost all required to live outside the walls, where other churches, baths and traces of a defensive palisade hint at the sweep of a sprawling suburban neighborhood substantially larger than the area enclosed within the walls.¹⁶⁵ ‘The people’ were thus very literally peripheral to the urban showpiece of the walled center, for all that they were indispensable on those occasions when the archbishops and praetorian prefects required living, chanting spectators to transform the colonnades along the processional route into a plausible facsimile of the cosmopolitan city that the dignity of their offices required them, as much as ever, to inhabit.

The ruling establishment’s continued fascination with colonnaded streets in the sixth century is equally manifest at Antioch, which Justinian was compelled to rebuild practically from scratch after it had first been devastated by a series of earthquakes and fires in the 520s, and then disastrously sacked by the Persians in 540.¹⁶⁶ The Persians had no sooner retreated than the emperor launched a massive reconstruction effort, focused first and foremost on reconstructing the city wall and restoring the colonnades along the main intramural thoroughfares.¹⁶⁷ Over a layer of rubble more than two meters thick in places, Justinian’s architects rebuilt the colonnaded backbone of the city’s main cardo to nearly the size and width of its predecessor, lining it with thousands of new columns along the two kilometers and more of its intramural sweep.¹⁶⁸ Looming over the blasted ruins of a city that would never fully recover, the stately ensemble of walls and colonnades screened the devastation from view behind a plausible monumental façade, ready outwardly to proclaim the resurgence of the greatest metropolis of the Levant, and to dignify the public appearances of its governors as they struggled to reassert the authority of the central administration in the devastated province.

3.7 The Long Shadow of Constantinople, II: Ravenna

The fifth century is now quite universally recognized as a period of relative peace and prosperity, indeed of economic and demographic expansion, across most of the eastern half of the empire. Institutional change came slowly, and generally flourishing cityscapes did not transform radically, though naturally urban topography is never static.¹⁶⁹ In the West, however, the disintegration of the empire and its partitioning into ‘barbarian’ successor kingdoms provoked more serious shifts in the ordering of government and society. The causes of the western empire’s dissolution in the fifth century continue to provoke lusty debate, and need not be rehashed here, except insofar as they relate to the condition of cities and towns.

An important point of convergence in much of the best recent work on the breakup of the western empire is, precisely, fragmentation. The western provinces, always a multicultural, multiethnic hodgepodge, embarked on diverging regional, and even micro-regional, trajectories with surprising alacrity, centrifugally spinning off from the center in Italy with a momentum that the advent of barbarian rulers does not entirely suffice to explain. To be sure, the arrival of large numbers of unsubdued peoples inside the empire, beginning with Adrianople in 378 and continuing through the collapse of the Rhine frontier ca. 406, was an historical watershed. But as (for example) Peter Brown and Guy Halsall have stressed, the really remarkable thing is how quickly Roman provincials, beginning with the landed aristocracy, reached an accommodation with Gothic, Burgundian, Frankish and Vandal rulers, recognizing in their local courts the best and most immediate source of protection and patronage available, and eagerly vying to serve there.¹⁷⁰ While the fourth century had witnessed the rise of a mobile service aristocracy of supra-regional scope and ambitions, empowered and enriched by their imperial connections, their fifth-century descendants in North Africa, Spain, much of Gaul and eventually Italy too continued to work for local rulers in similar capacities, but in (relative) mutual isolation.¹⁷¹ Their horizons contracted dramatically, in a manner reminiscent of the mid-third century, when the empire fragmented into Teilreiche, in part because in difficult times provincials preferred to serve, and gain advancement from, a nearby court rather than a remote and distant emperor.

The result was a constellation of what Brown has called ‘little Romes,’ regional kingdoms under Germanic rulers that modeled themselves on imperial precedent, adopted Latin as the language of law, administration and diplomacy, and were largely staffed by native ‘Romans.’¹⁷² The royal seats of these early successor kingdoms were consistently city based, resulting in a profusion of regional capitals: Vandal Carthage; Burgundian Geneva and Lyon; Visigothic Bordeaux, Toulouse and Barcelona; sub-Roman Soissons; Frankish Cologne, Trier and Tournai, and so on.¹⁷³ Further, as Halsall has suggested apropos of Spain and Gaul, the breakdown of Roman central authority and the generally unsettled conditions of the fifth century frequently made even non-‘regal’ civitates more central than ever to the ordering of local society, as the sole remaining stage upon which ambitious aristocrats and clergy could hope to act.¹⁷⁴ As imperial control wavered and receded, in other words, the civitas was frequently all that remained. While the West was undoubtedly poorer in absolute terms than the East in the fifth century (and after), the working hypothesis that we will seek to develop, first in relation to Ravenna, and in the next chapter to a number of other places in the West, is that a substantial group of leading urban centers continued to absorb and retain a relative preponderance of the resources available, as the cream of local society and the clergy joined their new rulers in working, spending and displaying themselves there.

The continuing capacity of the urban paradigm incarnated by (inter alia) Constantinople to reach even beyond the regions under the direct control of the eastern emperors is best exemplified at Ravenna, the last imperial capital in the West, later the first city in the realm of that most Romanized (or Byzantinized) of post-Roman sovereigns, Theoderic. As the most ambitious example of urban remodeling undertaken in the Latin West in the fifth and sixth centuries, Ravenna deserves consideration at some length. It offers crucial insights into the architectural vernacular shared by its rulers and archbishops, a vocabulary of power in which central porticated thoroughfares seem to have featured as prominently as they did in the greatest cities of the eastern Mediterranean.

Following Honorius’ transfer of the imperial court from Milan to Ravenna in 402, the western scions of the Theodosian dynasty clung to Constantinople both as the buttress of their faltering hold on power, and as the urban template for the transformation of their new capital into an imperial residence worthy of comparison with the New Rome on the Bosporus.¹⁷⁵ Honorius himself spent much of his youth in Constantinople. His sister Galla Placidia and nephew Valentinian III (western emperor 425–55) came directly to Ravenna from Constantinople in 425, following the deposition of the usurper John, effected with the help of troops dispatched from Constantinople by the young Valentinian’s cousin, Theodosius II.¹⁷⁶ Theoderic, who perhaps did more, over the course of his long reign (493–527), to shape the monumental infrastructure of Ravenna than any other resident ruler, had likewise spent some ten years of his youth as a hostage at the imperial palace in Constantinople, an experience that manifestly shaped his vision of the magnificent capital – a sort of ‘New Constantinople,’ as it were – he wished Ravenna to be.¹⁷⁷ Following the Byzantine conquest of the city in 540, links between Ravenna and Constantinople remained strong for the following two centuries of Byzantine rule, though building activity trailed off considerably from the later sixth century.¹⁷⁸

With the exception of the surviving churches of fifth- and sixth-century date, our knowledge of the topographical contours of the imperial (and later regal and exarchal) capital that Ravenna became after 402 is unfortunately rather vague. Excavations, limited by the dense fabric of the modern city, offer only tantalizing glimpses of the ancient city, widely scattered and of limited extent, amongst which the most significant is the section of Theoderic’s palace excavated between 1908 and 1914, to the east of the modern Via Roma and behind the church of Sant’Apollinare Nuovo.¹⁷⁹ The textual record is similarly lacunose, and the Liber Pontificalis of Andreas Agnellus, much the longest and most detailed source available, was compiled only in the mid-ninth century. While Agnellus’ text is replete with topographical references and descriptions of important monuments, many of the edifices he mentions were ruinous or wholly vanished in his day, and his sense of fifth- and sixth-century chronology is vague at best, and often badly confused.¹⁸⁰

The broad outlines of Ravenna’s urban plan as it evolved in the two centuries after 402 can nonetheless be reconstructed with some confidence. The monumental framework of the new imperial capital arose to the east of the rectangular perimeter of the old city center, in a neighborhood that came to be extensively reconfigured beginning soon after the arrival of the imperial court.¹⁸¹ The fulcrum of the new ‘imperial’ quarter, the regio Caesarum of later sources, was the platea maior, the urban tract of the road that led through the suburb of Caesarea and on to the port of Classe, which entered the city through the Porta San Lorenzo/S. Laurentii in the new circuit of walls, themselves apparently erected during the reign of Valentinian III, though work may have begun already under Honorius and continued into the latter part of the fifth century.¹⁸²

During the fifth century, a contiguous series of palace complexes arose on both sides of the road. A brief synopsis of centuries’ worth of debated issues might run something like this: when Honorius moved to Ravenna, he occupied an existing, previously extramural residential complex, possibly the residence of the praefectus classis ravennatis, located on the site of the excavated portions of Theoderic’s palace, where the existing structures of the first century AD were restored and embellished with new mosaics and opus sectile pavements around the beginning of the fifth century.¹⁸³ Valentinian III is the first emperor credited in the textual record with the construction of a palace ad Lauro/ad Laureta, located to the south of the later palace of Theoderic, stretching along both sides of the platea maior toward the Porta S. Lorenzo. Agnellus, clearly relying on an earlier source, puts the murder of Odoacer at the hands of Theoderic in 493 in the palace ad lauro;¹⁸⁴ he goes on to describe the complex at some length in a passage that contains a number of noteworthy observations: Valentinian III built a ‘lofty palace’ in a place called ad Laureta, a toponym that Agnellus associates with past triumphal victory parades based on the triumphal associations of the laurel; he goes on to say that the emperor constructed palace buildings on both sides of the platea (a point of exceptional importance, too often overlooked by modern scholars),¹⁸⁵ and marvels at the metal clamps, evidently visible in his day, used to reinforce the masonry in these buildings, before passing on to the city walls (which ran quite close to the palace ad Laureta), which he attributes to the same emperor.¹⁸⁶

In the early sixth century, Theoderic made extensive repairs and additions to the existing imperial palace(s), traces of which appeared in Ghirardini’s excavations of the complex north of the palace ad Laureta. Colonnades were added to the internal courtyard, new mosaics installed in the covered walkways of the quadriporticus, and an audience hall opening onto the northern flank of the colonnaded courtyard was enlarged and adorned with new floor mosaics.¹⁸⁷ To the west, just off the platea maior, Theoderic constructed a new, Arian palatine chapel dedicated, like the fourth-century church in the palace at Constantinople, to Christ the Savior (today’s Sant’Apollinare Nuovo).¹⁸⁸ The principal entrance to the palace lay off the platea maior in the immediate vicinity of the church, in a spot called ad Calchi (vel sim.) in later sources, a clear reference to the monumental entrance vestibule of the palace at Constantinople, the Chalke.¹⁸⁹ Porticoes were added to the exterior façade of the palace along the platea maior, flanking the entrance, as well as to the remaining three sides of the palace.¹⁹⁰ Thus, just as the colonnades along the final tract of the Mese at Constantinople led to the Chalke gate of the palace, so too at Ravenna, by Theoderic’s time at the latest, the principal entrance to the palace opened off the porticoes of the platea maior, which likely continued as far south as the walls at the Porta S. Lorenzo.¹⁹¹ With the construction of the Arian cathedral, baptistery and bishop’s palace along the west side of the platea maior, just north of the palace, the principal axis of Theoderic’s royal capital was complete: visitors arriving from the port at Classe, upon entering the Porta S. Lorenzo, found themselves invited to traverse a receding vista of porticoes, leading directly to the Calchi gate of the palace, the palatine chapel of the Savior/Sant’Apollinare Nuovo, and the Arian episcopal complex.

A second porticated avenue very probably intersected the platea maior at right angles just north of Sant’Apollinare Nuovo, along the line of the modern Via Mariani.¹⁹² This street would have connected the area of the palace with the one topographical focal point of the late antique city established before the arrival of the imperial court, the orthodox episcopal complex with its cathedral, baptistery and palace, which began to take shape under Bishop Ursus at the end of the fourth century.¹⁹³ It may indeed have been the first street to be monumentalized after 402, as the connector between the twin poles of secular and ecclesiastical government, an architectonic extension of the intertwining of civic and ecclesiastical authority so essential to the vision of a Christian Roman empire championed by Theodosius I and his heirs. This road, the principal east-west axis in the late antique city, was the urban extension of the old highway leading to Rome, the Via Popilia, which entered the city through a monumental gateway erected under the emperor Claudius, itself included in the fifth-century circuit of walls and renamed – in all probability in the fifth century – the Porta Aurea, in memory of the gate in the new Theodosian land walls of Constantinople.¹⁹⁴ From the episcopal complex, the road proceeded across the River Po on a bridge identified in later medieval sources as the ‘covered’ bridge (pons copertus),¹⁹⁵ past the imperial mint for gold coinage (the moneta aurea) located just north of the point where it crossed the platea maior,¹⁹⁶ on past the orthodox palatine chapel of Saint John the Evangelist, constructed during the reign of Valentinian III,¹⁹⁷ and finally on – very possibly in a direct line – to the gate in the eastern flank of the late antique circuit called the porta palatii.¹⁹⁸ Its porticoes, or the memory thereof, would endure for close to a millennium: in the tenth and eleventh centuries, when the road again connected the episcopal complex with the seat of civic government at the palazzo comunale, the two churches of S. Georgio de porticibus and S. Giustina in capite porticus, situated between the episcopal complex and the platea maior, still derived their popular designations from the porticoes lining the street, which remained the center of civic life in the medieval city (Figure 3.11).¹⁹⁹

Figure 3.11 Ravenna in the sixth century. A: Porta Aurea; B: cathedral; C: orthodox baptistery and episcopal complex; D: San Vitale; E: Santa Croce; F: Arian cathedral and baptistery; G: mint; H: Sant’Apollinare Nuovo; J: S. Giovanni Evangelista; K: Theoderic’s palace (1917 excavations); L: Porta Serrata; M: Porta S. Lorenzo; N: Porta Palatii?

(Author.)

By the time of Theoderic’s death in 527, then, the ‘topography of power’ assembled over the 125 years since the arrival of the imperial court and the resulting elevation of Ravenna to a position of ecclesiastical and administrative preeminence was largely complete. The orthodox and Arian episcopal complexes, the sprawling palatial quarter with its associated palatine churches, royal mint and monumental entrance vestibule ad Calchi, were all directly accessible from, and linked to each other by, the converging axes of two principal streets. These streets, architectonically elevated above the rest by their flanking porticoes and connected to the principal gates in the circuit-wall, structured and defined the urban experience of Ravenna, channeling the flow of traffic and majestically framing the approaches to those buildings that most visibly and effectively proclaimed the ascendancy of their inhabitants and the institutions of church (orthodox and Arian) and government (western imperial, Gothic, and Byzantine) they championed. As at Constantinople, which the architects of Ravenna’s transformation strove so diligently to emulate, the monumental core of the city was engineered to compress the experience of a much larger, more chaotic and undoubtedly less architecturally distinguished urban conglomeration into an architectonic mise-en-scène, a city within a city that translated the language of temporal and spiritual sovereignty into the realm of built space.

And as at all of the other late antique capitals hitherto discussed, so too at Ravenna, that language of power was never more ostentatiously proclaimed, nor more integrally connected with its architectural context, than during those moments of ceremonial high drama when bishops and rulers paraded in majesty through the main streets of the city. While we know relatively little of the particulars of the ceremonies that unfolded at Ravenna, both regarding recurring events such as liturgical processions, and exceptional occasions such as adventus and victory celebrations, it is beyond doubt that Ravenna’s porticated avenues regularly hosted processions similar to those attested at, for example, Rome and Constantinople. Much of what is known comes from Agnellus, whose repeated evocations of the gates and main streets, bedecked with flowers and tapestries to receive the protagonists at the heart of defining moments in Ravenna’s storied past, reveal, if nothing else, the extent to which the decaying monumental contours of the ninth-century city continued to resound with the echoes of celebrations long past.

In his account of the negotiations that led to Attila’s decision to spare the city from destruction in ca. 450, Agnellus has Attila depart amidst general rejoicing, hailed for his clemency by the crowds lining his route: ‘Therefore the king went out from the city; [the people of Ravenna] went before him with great acclamations, with all the streets adorned and the city decorated with all manner of flowers.’²⁰⁰

When Maximian entered the city in 546 to be consecrated archbishop, following bitter disputes over the episcopal succession, ‘all set out in unanimity, as though one person, and opening the gates of the city, with crosses and signs and banners and praises they honorably conducted him inside this city of Ravenna; and they kissed his feet and adorned the streets of the city with various ornaments. All the buildings were crowned [bedecked with flowers], and there was public rejoicing among the contending parties.’²⁰¹ Two centuries later, when a Byzantine attack was repulsed on the feast day of John and Paul, the people of Ravenna began to mark the anniversary ‘almost as though it were the holy festival of Easter, adorning the streets of the city with many tapestries, and setting out in a liturgical procession (lethaneis) to their [sc. John and Paul’s] church.’²⁰² It might also be recalled that the site of Valentinian’s palace ad Laureta was, for Agnellus, entwined with the memory of a ‘triumphal victory once celebrated in that place,’²⁰³ an association that evokes shades of the triumphal processions that subsequently unfolded along the stretch of the platea maior – the Mese or Regia, as it were, of late antique Ravenna – in the midst of the palace.

Whatever the relationship between Agnellus’ visions of past urban ceremony and the reality they purport to represent (and he likely did draw some of his information from earlier sources that have since perished), his sense of the city’s monumental infrastructure as the stage upon which marvelous pageants were enacted is noteworthy. Agnellus’ continued inclination to perceive his surroundings as processional space, and to imaginatively layer crowns of flowers and teeming multitudes over a backdrop of city gates, streets and crumbling palaces, suggests something of how successful Ravenna’s fifth- and sixth-century rulers were in creating a physical landscape infused with the memory of their public appearances, a space capable of invoking an aura of the very epiphanies it was designed to facilitate. Agnellus, in other words, seems to have perceived the cityscape and the semiotics of power encapsulated therein in much the way Theoderic, for example – whom we know best of all Ravenna’s late antique rulers – intended it to be perceived.

Agnellus’ famous description of Theoderic’s palace indicates that the pediment of the main entrance to the palace ad Calchi was decorated with a mosaic depicting the king on horseback, armed with lance and shield, surrounded by personifications of Rome and Ravenna.²⁰⁴ A second equestrian portrait, this the famous statue later taken by Charlemagne to Aachen, stood in front of the entrance.²⁰⁵ The equally famous mosaic depiction of Ravenna at Sant’Apollinare Nuovo in fact shows what must be the Calchi, in the form of a tetrastyle entrance porch surmounted by a triangular pediment (bearing the designation PALATIUM) that once contained an equestrian portrait, surely an echo of the mosaic described by Agnellus (Plate I).²⁰⁶ The entrance is flanked by lower, arcaded colonnades, which may have been meant to represent – or better, to call to mind – the colonnaded façade of the palace along the plateia maior, in the midst of which the Calchi was indeed located, and which Theoderic is said to have built.²⁰⁷ Yet the presence of a city gate, bearing the legend CIVITAS RAVENN(AS), shown directly abutting the outermost column of the arcade on the right side of the image, demonstrates that the vignette as a whole is highly schematic.²⁰⁸ While its bearing on the topography of the ‘real’ city can be (and has been) endlessly debated,²⁰⁹ the image is sufficiently illuminating when taken purely on its own terms, as a visual compression of the elements considered most essential to the presentation of Theoderic’s capital. Seen in this light, the message seems quite clear: the palatial quarter of the city, the regio Caesarum, was to be understood, epitomized, symbolized as a city gate, opening onto an arcade – a porticated street – that in turn led directly to the entrance of the palace itself, the seat and embodiment of Theoderic’s rule, the ‘fair face of imperium,’ as Cassiodorus put it.²¹⁰

Plate I Sant’Apollinare Nuovo: nave-mosaic showing Ravenna and Theoderic’s palace.

(Photo: author.)

The figures originally present in the intercolumniations of the palace façade, hands raised to acclaim Theoderic as he processed to or (more likely) from the palace,²¹¹ provide a crucial interpretive key to the mosaic’s selective distillation of urban topography: the gates, colonnaded platea and palace are the essence of the city, insofar as they constituted the monumental backdrop that underpinned the ceremonial processions of the king, and proclaimed his capital a locus of rarified temporal power. Such is precisely the sequence of monuments that Theoderic would have traversed when approaching or leaving the main entrance of his palace, via the porticated expanse of the platea maior and the gates that bracketed its northern and southern extremities, the Porta Serrata and the Porta S. Lorenzo, respectively. The placement of Theoderic’s equestrian statue, and the closely related mosaic on the façade of the Calchi, at the juncture where the arcades of the platea opened onto the entrance to the palace, marked the culmination of the triumphal itinerary, the very spot shown at the center of the urban vignette at Sant’Apollinare Nuovo, where the reproduction of the equestrian portrait on the Calchi stood in suggestive visual counterpoint to the depiction of the king in the act of processing through the city. Like Justinian’s equestrian statue in the Augusteion by the entrance to the palace at Constantinople,²¹² these images permanently embedded the presence of the sovereign in the heart of the most symbolically charged real estate in the city, as a constant reminder of those moments when he appeared in person among his subjects, processing amongst them on horseback in full martial regalia.

Under Theoderic, then, topography and iconography combined to transform Ravenna into a vessel for the triumphal epiphanies of the king. In constructing (or reconstructing) the exterior colonnades of the palace and thereby framing the Calchi gate and integrating its approaches into the porticoes of the platea maior, Theoderic reified the nexus between city gates, porticated street and royal dwelling that defined the architectural language of temporal rule in the early sixth century, and remade Ravenna in the image of the greatest capital of the age, Constantinople. The ultimate confirmation of the cohesiveness of the urban template envisioned by Theoderic relates not to Ravenna, however, but to Verona, the second city (along with Pavia) of the realm. Immediately after describing the porticoes erected around the palace at Ravenna, the author of the Anonymous Valesianus, Theoderic’s younger contemporary, turned to Verona, where the king ‘Likewise (my emphasis) at Verona built baths and a palace, and constructed a portico from the city-gate as far as the palace.’²¹³ To transform Verona from a provincial center into a seat of power, what was needed was a palace, connected to a city gate by a porticated street: an urban armature, that is, fit for the arrival of a king, particularly a king as steeped in the traditions of late Roman triumphal ideology as Theoderic manifestly was.²¹⁴

3.8 The Intersection of ‘Real’ and ‘Ideal’: Cities Depicted

Nowhere is the capacity of these porticated armatures to streamline and condense the physical and ideological components of the urban experience more apparent than in the extant depictions of cities produced in both the eastern and western Mediterranean in the fifth to eighth centuries. Rather than seeking to produce comprehensive plans of cities, late antique artists strove to capture the idea of ‘city’; to imply the whole through a limited canon of representational tropes centered on the features most readily associated in the minds of viewers (or patrons, or artists) with the concept of polis or civitas.²¹⁵ It is thus the more noteworthy that after city walls, porticated streets are perhaps the single most distinctive feature of the cityscapes depicted in extant manuscripts and mosaics commissioned under the aegis of church and state, which is to say by the same class of patrons responsible for reshaping real cities into visual panoramas largely defined by walls and colonnades.

The most salient examples in the manuscript tradition appear in the Notitia Dignitatum, a detailed catalogue of the administrative subdivisions of the empire compiled around the beginning of the fifth century, preserved in a sixteenth-century copy that seems to reproduce quite faithfully the illustrations – at one remove – of a late antique exemplar.²¹⁶ Two sections of the ‘western’ half of the document, both devoted to the commands of officials stationed in Italy, contain especially valuable illustrations. The image accompanying the list of forces at the disposition of the Comes Italiae depicts the fortified passes of the Alps, at the base of which stands a city, labeled ‘Italia,’ that stands in for the region as a whole (Plate II).²¹⁷ An almost identical cityscape represents the provinces of Apulia and Calabria, joined under the governorship of the Corrector Apuliae et Calabriae.²¹⁸ Both images are dominated by the profile of a hexagonal city wall, the three visible faces of which occupy approximately half of the urban vignettes as a whole. A centrally placed gate in the façade of the wall opens onto a single dominant intramural feature, an arcaded street colonnade, roofed with vivid red tiles, that departs from the gate at an oblique angle, bends sharply off to the right and continues as far as the receding profile of the wall, presumably to connect with another city gate.

Plate II ‘Italia’ in the Notitia Dignitatum.

(Bayerische Staatsbibliothek BSB-Hss Clm 10291, fol. 214v.)

This remarkable compression of the urban image into a schematized ensemble of city wall, gate and colonnaded street finds its closest parallels in a number of mosaics datable between the fifth century and the seventh, all of them, interestingly enough, from Italy. The earliest are the representations of Jerusalem and Bethlehem on the spandrels of the triumphal arch at Santa Maria Maggiore at Rome, built under Pope Sixtus III (432–40). Here, too, the contours of the cityscape are defined by the profile of a hexagonal city wall, in this case bedecked with jewels in a fashion recalling the description of Celestial Jerusalem in the Book of the Apocalypse (Apoc. 10: 10–21),²¹⁹ above which only the roofs of a few intramural structures are visible. The gate in the wall opens onto the receding profile of a monumental street colonnade, which joins the wall as the most visually arresting component of the scene (Plate III). The iconographical signposts used to convey all the physical and spiritual grandeur of Jerusalem and Bethlehem, in their quality as superlative embodiments of the (Christian) metropolis, boil down to an interlinked armature of city wall and colonnaded street. It is precisely the vision that would have confronted real visitors as they approached the walls and passed through the gates of the urban showpieces of late antiquity, from Constantinople, to Justiniana Prima, to – as we shall see – Jerusalem itself.

Plate III Jerusalem at Santa Maria Maggiore, Rome, ca. 440 AD.

(Photo: author.)

More than two centuries later, in the oratory of S. Vincenzo, annexed to the Lateran Baptistery in the 640s under popes John IV and Theodore I, Jerusalem would again be signified by the image of a wall, a gate and a two-story colonnade (Plate IV). While the form of the cityscape – which appears in the same position on the spandrel of the triumphal arch as the example at Sta. Maria Maggiore – was doubtless inspired by earlier precedents, the fact remains that in the seventh century as much as the fifth, the essence of the Christian city was best conveyed to a Roman audience by the vision of a wall and a gate opening onto a colonnaded street.

Plate IV Jerusalem in the oratory of San Vincenzo, Rome, 640s AD.

(Photo: author.)

Elsewhere in Italy, there is the representation of the port city of Classe, located above the nave arcade in Theoderic’s new palace-church of Sant’Apollinare Nuovo, built adjacent to the royal palace in Ravenna in ca. 500.²²⁰ Once again, the predominant feature is the city wall, flanked on the right by a gate and on the left by a view of the port itself. Above the wall, the view of the interior of the city centers on the profile of two large, round buildings, never convincingly identified, which are connected by the unmistakable silhouette of a porticated street, the arcades and tile roof of which closely resemble the examples depicted in the Notitia Dignitatum (Plate V).²²¹ Whatever the two buildings represent, the intramural topography of Classe is epitomized in the form of a single monumental itinerary, a colonnaded avenue linking the most imposing structures in the city, the rightmost of which may in turn be connected to the city gate by the suggestion of an additional, trabeated colonnade.

Plate V Sant’Apollinare Nuovo: nave-mosaic showing the port of Classe.

(Photo: author.)

In the more famous vignette of Ravenna, located en pendant just across the nave, the colonnaded façade of Theoderic’s palatium abuts a city gate, almost certainly the Porta S. Laurentii, through which the road leading through the suburb of Caesarea and on to Classe entered the city (Plate I). Thus, the principal ceremonial axis of the city, from the Porta S. Laurentii, along the – colonnaded – platea maior and thence to the royal palace, becomes a binary pairing of gate and palace, joined by a colonnade that likely recalled the intervening street as much as the façade of the palace itself.²²² Just as Theoderic’s royal capital at Ravenna was to be experienced in a stately progression of monuments linked by a colonnaded avenue, so was the adjacent port of Classe, both in mosaic and, in all probability, in reality: excavations undertaken in Classe in the 1970s and 1980s in fact uncovered a lengthy tract (running for ca. 500m) of what was almost certainly the main street leading away from the harbor; the street was repaved, provided with new sewers, and lined on both sides with porticoes supported by brick pilasters at the end of the fifth century, scant years before the execution of the mosaics at Sant’Apollinare Nuovo.²²³

The markedly ceremonial character of the topography represented in the images of Ravenna and Classe was further accentuated by the presence of King Theoderic and his entourage, who almost certainly appeared in both cityscapes, filling the intercolumniations of the palace façade in the depiction of Ravenna (much like the onlookers in the Constantinopolitan ‘Trier Ivory’), and silhouetted against the walls of Classe.²²⁴ The fundamental nexus between (urban) architecture, ceremony and the construction of regal legitimacy would have gone unrealized without the presence of the king, who animated the monumental armature of Ravenna and Classe with his public appearances, and maintained its architectural decorum with his patronage. Hence, following the Byzantine reconquest of the city in 540, all trace of human presence was expunged from both cityscapes, when the formerly Arian palace-church was rededicated to St. Martin of Tours under Archbishop Agnellus (ca. 557–70).²²⁵ Bereft of the ceremonial pageantry that had proclaimed the regal status of the city and the glory of its (Arian) Gothic ruler, the architectural background remained, innocuous, generic, shorn of royal pretensions, befitting the administrative seat of a peripheral and humbled province of the Byzantine empire.²²⁶

Turning to the eastern Mediterranean, the best-known depiction of a late antique cityscape in the region is undoubtedly the image of Jerusalem placed at the center of the Madaba Map Mosaic, a tesselated floor-pavement from the church of St. George in Madaba, Jordan, that depicts the geography of the Levant, from Egypt in the south to the borders of Asia Minor in the north (Plate VI).²²⁷ The centerpiece of the whole composition, the city of Jerusalem is shown larger and in considerably more detail than any of the other cities on the map. The ovoid periphery of the city wall, rendered in bird’s-eye perspective, circumscribes an ensemble of intramural topography dominated by the contours of two colonnaded streets, both of which depart from a roughly semicircular plaza located just inside the northern gate of the city, today’s Damascus Gate. The two most monumental structures depicted both connect directly to the colonnades of the central cardo: the Church of the Holy Sepulcher, placed farther south than its actual location to lie in the very center of the city, and thus the map as a whole, and Justinian’s new church of the Virgin Mary, the nea ecclesia, at the southern extremity of the same street (its dedication in 542/43 provides the terminus post for the mosaic). A third monumental structure, probably the Church of Holy Zion, lies just beyond and to the west of the point where the street terminates.²²⁸

Plate VI The ‘Madaba Map Mosaic,’ detail of Jerusalem.

(Photo: author.)

While the focus is clearly on the religious topography of the city, the monumental armature remains firmly in the mold of the late antique capital. A central core of colonnaded streets connects the principal gates in the city wall with the most prominent focal points of intramural life, presented in the form of three preeminent churches.²²⁹ The mosaicist’s vision indeed closely mirrors that of Justinian himself, whose most salient intervention in the topography of the real city aimed precisely at accentuating the architectonic profile of the cardo maximus, the colonnades of which were extended for hundreds of meters south of the tetrapylon at the crossing of the cardo and decumanus, as far as the façade of the newly completed Nea church.²³⁰ The principal colonnaded thoroughfare of the city, nearly doubled in length, was thenceforth bracketed by the Church of the Holy Sepulcher in the north, and Justinian’s massive new church in the south.²³¹ These twin poles, connected by an unbroken line of colonnades, anchored the ceremonial topography of a city that, for the first time since its elevation to patriarchal rank at Chalcedon in 451, boasted an architectural centerpiece worthy of its lofty status, a space in which the patriarchs of Jerusalem could be seen to rival their colleagues in Rome, Constantinople, Antioch and Alexandria.²³²

The visual perspective of the mosaic is thus clearly rooted in contemporary reality, for all that the representation of Jerusalem is hardly a faithful rendering of its architecture and topography. The irregular profile of the city wall is depicted as a perfect oval; side streets are nonexistent, as is the Temple Mount; and the majority of the buildings that fill the interstices between the wall and the colonnaded streets are schematically rendered and difficult to identify. Yet the diverging profile of the two colonnaded streets shown on the map squares remarkably well with their actual configuration, still preserved in the contours of the two modern streets that faithfully follow their course, the Tariq Khan ez-Seit and the Tariq el-Wad (Figure 3.12).²³³ A comparative glance at Plate VI and Figure 3.12 will make this point best: by far the closest correspondence between the mosaic and the real city occurs in the profile of the two colonnaded streets, the (real) spaces that surely not coincidentally did most to define the experience of the city for those who saw it firsthand, from – naturally – street level.

Figure 3.12 Jerusalem in late antiquity. A: Neapolis Gate (just inside modern Damascus Gate); B: open plaza and column; C: arch; D: Eastern Gate/Lion Gate; E: Holy Sepulcher complex; F: Western Gate; G: tetrapylon; H: Nea Church; J: Nea Gate; K: Church of Holy Zion.

(Author.)

In condensing the symbolic hallmarks of the city into an ensemble of colonnaded streets and grand churches, which corresponded in its essentials with the disposition of those monuments on the ground, the makers of the mosaic succeeded in creating a vision of Jerusalem that would have been as edifying for the residents of Madaba who saw it on the floor of the church as it would have been plausible for those who regularly traversed the city center, or participated in the processions between the three churches that anchored the liturgical topography of the city and proclaimed its unique place at the epicenter of the Christian cosmos. The monumental corridor in the midst of the city, expanded and embellished by Justinian, was simultaneously a reality susceptible to (near-) cartographic reproduction, and an ideal landscape designed to convey the essence of Jerusalem as the Christian capital par excellence.

That the archaeological map of Jerusalem on the one hand, and its features as shown on the Madaba Map on the other, correspond most closely in the positioning of the two colonnaded streets and the principal edifices attached to the central cardo is compelling testimony to what all of the ‘western’ images discussed earlier already strongly suggest: the story of the late antique metropolis is not – not always, in any case – characterized by a profound divergence between ‘image’ and ‘reality,’ a widening chasm between rosy, anachronistic literary and artistic tropes applied to ever-more squalid cityscapes, but rather by the methodical privileging of a very selective reality.²³⁴ The Madaba Map’s Jerusalem translated the material essence of the city in its ‘Justinianic’ form into an iconic panorama, a visual itinerary that closely mirrored the ensemble of gates, colonnaded avenues and churches consciously designed to define the experience of the real city. The images of Ravenna and Classe at Sant’Apollinare Nuovo functioned in much the same way, adding only the palace that proclaimed Ravenna the seat of an earthly monarch. In sum: the colonnaded streets, with their associated gates, churches and palaces, that loomed so large in the civic and religious life – and ceremonial – of the late antique capital were as real and imposing as ever in the sixth century, presenting the same exemplary urban façade that artists captured in their depictions of cities, both real and, in the case of Celestial Jerusalem, imagined.

3.9 Conclusions

Porticated streets, then, were the places where the dominant personalities in late antique society most effectively displayed themselves to the urban masses, where they acted out their status as civic and/or religious leaders, and drove home the majesty of their offices to the largest possible number of people. The unfolding of ceremonial processions brought together all members of the urban populace, framed by stately lines of receding columns, in a seamless blend of architecture and bodies that came to stand for, and in a sense to be, the city as a whole. As long as there was a processional way and people to fill it, there was in a real sense a city, regardless of the condition of the neighborhoods behind the porticoes, some still teeming with life in the sixth century (Constantinople, Ephesus, Ravenna); some ruined (Antioch); some almost nonexistent (Justiniana Prima). In condensing the architectural, commercial and symbolic essence of the city into the profile of a porticated street and its associated monuments, emperors, governors and bishops achieved a kind of architectural shorthand, a relatively economical means of preserving the appearance of capital cities and ensuring that the urban spectacle lived on.

Yet perhaps the strongest testament to just how powerful the physical, conceptual and functional legacy of these central urban itineraries ultimately became is the signs of their continuing relevance even in the seventh century and later, in a definitively post-Roman world in which the social, political and economic structures that had underpinned the flourishing of ‘classical’ urbanism had changed almost beyond recognition, as had the cities themselves, when they survived at all. It is to this changed world of the seventh to ninth centuries, the darkest of the ‘Dark Ages,’ that we now turn, in an effort to trace further the nexus between urban topography and what we might call spatial praxis – the use and experience of lived space – in the crucial transitional centuries between late antiquity and the early Middle Ages.