Since 1953, four centuries after the event, every year on May 29, Istanbul celebrates the conquest of Constantinople by Mehmet II Fatih.Footnote ¹ In 2024, another of the city’s anniversaries did not warrant such a public commemoration. Instead, it was left to the Pera Museum, a private institution, to arrange a well-attended two-day conference – “324,” to mark 1,700 years since the city’s foundation by Constantine the Great. The year has been marked by the publication of two significant contributions to the city’s Late Roman and Byzantine past, the first by the eminent Byzantinist Paul Magdalino,Footnote ² a subtle study of how the past was renegotiated over the centuries of the city’s Byzantine existence, and the second by a renowned historian of the Late Roman empire, John Matthews (M.), which is the subject of the current review. The core of M’s study is a translation and commentary of the Notitia urbis Constantinopolitanae. The significance of this 5th-c. account of the urban districts within the Constantinian walls, comparable to Rome’s regionary catalogues, has been known since Pierre Gilles’s pioneering account of the 16th-c. Ottoman city. The translation, albeit with minor changes, is not new,Footnote ³ but there is a much more extensive commentary set in the broad context of the earlier Greco-Roman city and the Constantinian re-foundation. In the introduction M. identifies many of the key previous studies of Constantinople’s early years, especially the seminal studies by Cyril Mango, although surprisingly Gilbert Dagron’s fundamental Naissance d’une capitale is not singled out.Footnote ⁴ While there is much to commend, in what follows in this review, I will attempt to identify key recent works which have not been included or recently published, since as the author observes, his study was completed in a time when library access was limited.

The first chapter is a bracing account of Constantine’s trajectory to power from distant Britannia, campaigning against the Picts, to the victory over Licinius and sole ruler of the Roman world on the seaway separating Europe and Asia. Here, M. adroitly summarizes Constantine’s path to power stressing his “greatness of mind” (ingens animus from the inscription on the Arch of Constantine in Rome), which could achieve the foundation of his new and extended city on November 8, 324, only 51 days after the defeat of his rival. The next chapter moves on to consider the pre-Constantinian city. M. stresses the range of sources for both the Greek and the Roman city. Besieged by Philip II, attacked by Gauls, the opportunities for commerce and defense its remarkable setting presented were offset by threats to the city’s security. In assessing Byzantium’s strengths and weaknesses, M. accepts Tacitus’s statement of its prosperous fish resources but questions his comment on its fertile lands (Annals 12.63). M. goes on to discuss how the new city of Constantine required the grain ships from Egypt to sustain its increased population, although it should be noted that the near-equivalent sized 11th-c. Constantinople could be resourced from adjacent regions, which for the smaller city of Byzantium helps support Tacitus’s statement.Footnote ⁵ Similarly M. follows several recent studies, quoting amongst other sources the Admiralty Pilot, concerning the difficulties of passage in the Bosporus when faced with contrary headwinds and adverse currents. However, whatever the difficulties encountered by traffic sailing north through the straits, both archaeological and historical sources report Justinian’s long-distance logistical support of the Quaestor Exercitus for the Danube frontier, derived from southwest Asia Minor and the Greek islands.

The discussion of the Imperial Roman-period evidence is particularly welcome. In the late-1st c., Pliny’s letters show Byzantium was part of Bithynia, although whether this continued up to the time of the Severan siege is less certain: Perinthos as metropolitan city of provincia Thracia was not far distant.Footnote ⁶ M. is right to question those who have underestimated Byzantium’s size, although here one confronts the challenge of an almost total absence of material remains of the earlier city and a reliance on the various testimonies relating to the Severan siege and subsequent city. At the time of the siege the city was walled, and we can envisage a cut-stone curtain similar to the one at Assos in the southern Troad, or the land walls of Sinope on the Black Sea.Footnote ⁷ These massive walls, dating from the Hellenistic age, were demolished by Severus after the siege. Janin hypothetically shows the seven towers noted by Dio flanking the west side of the acropolis and Topkapı Sarayı in his map of “Byzance antique.”Footnote ⁸ But nothing survives or has been documented in the past (and no convincing remains were reported at the recent “324” conference). Two features determine reconstructions of the pre-Constantinian city. The first is the statement that the new forum of Constantine, whose location is still marked by the eponymous porphyry column, lay outside the boundary of the old city. Previous studies, notably by Janin, Müller-Wiener, and Mango, make reference to the distribution of funerary remains: stelae, sarcophagi and hypogaea, based notably on the publications of Dr Nezih Fıratlı, director of the Istanbul Archaeological Museums until the later 1970s.Footnote ⁹ These distribution maps of Greco-Roman necropoli represent differing phases of extra-mural burials, with concentrations along modern urban developments, such as implementation of 20th-c. sewer systems especially along Ordu Caddesi (close to the line of the later Mese and the main road to Thrace). Significantly, on the maps in Müller-Wiener the funerary remains extend up to the north end of the Hippodrome.Footnote ¹⁰

The second urban feature that influences any reconstruction of the extent of the Roman city is the presence of the Hippodrome, jutting southwestwards from the high ground now occupied by Hagia Sophia and the Augusteion. Still prominent in the cityscape is the massive curved terraced wall of its sphendone. In M’s map, the line of the Severan wall is projected directly across the peninsula within the site of Constantine’s forum and easily incorporates the Hippodrome (M. fig. 2.1). Other reconstructions follow a similar course up from the Golden Horn (thus including the earlier harbors), incorporate the sphendone of the Hippodrome, but include a smaller area east to the shore of the Bosporus (see Fig. 1). The main study of Circuses and Hippodromes in the Roman empire, by John Humphrey, explicitly omits a discussion of the Constantinople Hippodrome, recognizing the complexities of the site and structure,Footnote ¹¹ and its history, referencing a 1956 Harvard PhD thesis by W. L. McDonald, later a Roman architectural historian of note, although the thesis is currently unlisted in the Hollis Library Catalogue. The most recent study of the Hippodrome appears in a number of significant contributions to the two-volume catalogue accompanying the exhibition at the Pera Museum in 2010 titled the Hippodrome/Atmeydanı: A Stage for Istanbul’s History. The first volume concerns the Roman and Byzantine Hippodrome and the second, its Ottoman legacy.Footnote ¹² In the first are contributions by Dagron, Mango, and Jonathan Bardill. All three are consistent in doubting the 6th-c. sources’ claims for Hippodrome’s Severan origin. Bardill, the leading expert on Constantinopolitan bricks and brick stamps, observes that none of the known bricks from the Hippodrome are earlier than Constantine and stresses the Constantinian and later date of the structure.Footnote ¹³ By way of contrast, in the second volume, concerned with the Ottoman legacy of the Hippodrome, the veteran architectural historian of the city Doğan Kuban and also Özdemir Nuktu both follow the ingrained urban myth of the Severan foundation. Current scholarship in this and other publications on the origins of the city has come to recognize a more nuanced and close reading of the Roman and Byzantine sources, cogently summarized in a recent article by Albrecht Berger, a scholar often cited in M’s discussion:

Fig. 1.

An excerpt from Alexander Van Millingen’s map of Constantinople showing the Acropolis, the projected line of the wall of Byzantium, and the outline of the Notitia’s regions I–XI. (Van Millingen Reference Van Millingen1899, facing 19.)

Humphrey, writing 40 years ago, accepted the conventional Severan origin for the Hippodrome but later in his study highlighted the Constantinopolitan circus as part of the ceremonial and ideological toolkit for Tetrarchic palaces from Antioch to Trier. Significantly, he also included the possibility of a hippodrome at nearby Perinthos, renamed Heracleia in Late Antiquity.Footnote ¹⁵ By insisting on the supposed Severan origin for these structures, M. detracts from a major part of Constantine’s achievement, the very vision of the ingens animus he identified from the emperor’s arch in Rome. After the 6th c., Constantinople was to become a holy city; how this emerged remains uncertain, certainly in the time of the “First Christian emperor,” and M. provides a concise and acute survey of the textual evidence for the limited presence of the city’s first churches.Footnote ¹⁶ A third chapter considers the Sources and Materials and recognizes the multi-layered character of not just the rich but destructive stratigraphy of Istanbul’s built environment, but also the written accounts such as the Patria and Parastaseis, redefining past narratives and monuments into contemporary fables. M. does well to provide the reader with a taste of both the eccentricity and potential of these sources. But from literary urban imagery, the author moves on to an especially valuable consideration of how early cartography spanning the 15th and 16th c. contributes to an understanding of past monuments and topography; in particular, he highlights Ian Manners’s important studies of Buondelmonti and other sources.

The key contribution of this book is not just the translation of the Notitia urbem Constinopolitana, alluded to in the preceding chapters, but the extensive commentary that follows. The manuscript, with a single illustration, survives as a late medieval copy from Speyer, and the pendant Roman part is lost except from its frontispiece, although earlier regionary catalogues of the imperial city are known, as well as a short Syriac Notitia from Alexandria. The Constantinopolitan text dates to the early 5th c. and describes the 12 regions within the Constantinian Walls with two outside. The new 5th-c. land walls completed by Theodosius II are mentioned in the preface, which helps to define the text’s chronology,Footnote ¹⁷ but the text provides a precise measurement of the city’s length up to the Constantinian Porta Aurea of 14,075 Roman feet, which pretty closely conforms with the distance from the eastern promontory to the estimated site of the Constantinian Golden Gate and helps to define the 12 intra-mural regions within the Constantinian walls. Many of the elements in the catalogue require little interpretation, such as bakeries, baths, nymphaea, or palaces, but some items, especially relating to the residences of the urban population, remain enigmatic. M. notes how the porticus are described variously as either “continuous” (Region I), “grand” (Region II–VII), or “greater” (Region VIII–XIV) and observes how some of the 65 colonnades are at right angles to the city’s main east–west armatures. Local streets are termed as vicus, also with the meaning of neighborhood, or angiportus – alleys.Footnote ¹⁸ More challenging is the term domus, which M. sensibly accepts as meaning a freestanding generic residence, of various sizes and potentially multiple occupancies; unlike the Roman catalogues, there is no mention of insulae. Footnote ¹⁹

The following chapter 5 presents a commentary on the text of the 14 regions, dividing them into three groups (I–VI, VII–IX, X–XII) followed by the extra-mural regions. The graphical representation of the lists is most welcome, unlike that of all other commentators, who present the regions like large jigsaw pieces in the framework of the peninsula (see Fig. 1). M. recognizes the limitation of this approach and sets the key elements of the text onto three sketch maps of the city, thus reflecting what is known and unknown (M. figs. 5.1, 5.2, 5.3). Much of region I included the Great Palace, and to the north in region II were the city’s two main churches and beyond them, the Acropolis of Byzantium with a theater on its eastern flank.Footnote ²⁰ Region III lay west of the palace and included the hippodrome extending to the new harbor on the south shore of the peninsula. Region IV lay to the north and included the Augusteum and Basilica (defined by the Yerebatan Saray) and ran down the slope west of the Acropolis to the heart of Byzantium and the shore of the Golden Horn. To the west along the shoreline were regions V and VI with warehouses, harbors, and landing stages (scalae), but at the south of VI was the forum of Constantine and the porphyry column. The regions to the west (VII–XII) occupied the new city up to the Constantinian land walls. Here M. adopts the maximalist approach for the line of the circuit known to abut the later 5th-c. open cisterns.Footnote ²¹ To the north and adjacent to VI was region VII, amongst the most populous in the city, including the forum of Theodosius and to the north the district of Oxeia.Footnote ²² South of the Mese (the medial colonnaded street) was Region VIII, one of only two regions that did not touch the sea, and to the south, IX, including new granaries and harbor of Theodosius.Footnote ²³ Finally, within the walls from north to south were the regions X, XI, and XII, including the bifurcation of the city’s main colonnaded streets and a second forum of Theodosius (later Arcadius). Beyond the walls, region XIII lay across the Golden Horn at Sycae (later Galata) and for the enigmatic region XIV, here M. follows Cyril Mango’s later solution at Rhegion, situated 12 Roman miles west of the city. Other commentators continue to speculate about its location.Footnote ²⁴

Having established the urban framework based on the Notitia text, M. then returns in the following three chapters (6, 7, 8) to trace the urban development from Greco-Roman Byzantium to the Theodosian city, identifying cumulative layers of the city’s growth. Although in part this covers ground already described in the introduction, there are many valuable observations, particularly concerning the Constantinian foundation and later phases, focusing in particular on the urban monuments. Altogether this is a welcome and critical overview of recent scholarship informed by the author’s deep knowledge of Late Roman history, although as expressed earlier, I found it unfortunate that there was emphasis on the so-called Severan city with comparanda to examples such as Lepcis Magna. By contrast, a recent study of the Basilica in region IV in the context of the earlier Greek city identifies the site of the later Basilica as an upper agora with an earlier Hellenistic-age stoa, comparable to Lindos on Rhodes, which at the same time privileges the topography of the Greek city over its Roman successor.Footnote ²⁵ From the perspective of the 6th-c. Christian city of Constantinople, one of the conundrums of the Notitia is that it lists only 14 churches, and in the concluding section of the final chapter in this sequence M. presents a judicious and convincing analysis of the city’s evolving religious topography into the Theodosian age.

In the discussion of the listed churches, it is noted that in two of the most densely populated regions, VII (beside the Golden Horn) and IX (the new harbor of Theodosius), the churches “catered for the large Christian community among the workers of the city” (179). Such insights derive from the detailed and innovative analysis in chapter 10, “Housing and Population.” Here, the inventory of the Notitia is not simply a guide to the urban topography and its key monuments but, in the absence of any known archaeologically documented domestic architecture apart from palaces,Footnote ²⁶ offers insights into the lived-urban environment. It is the houses (domus) – here, M. questions Mango’s assertion that that these are only substantial houses (199) – the streets and alleys, the public and private baths and bakeries, the gradus, steps for the distribution of the corn dole, and the collegiati, bands of men deployed against fire and urban hazard for each of the city’s regions. M. offers three cautions: the Notitia inventories are not statistics (and there are inconsistencies); Roman numerals are vulnerable to distortion in the transmission of manuscripts; the regions themselves are administrative districts not socio-economic boundaries.Footnote ²⁷ By careful tabulation of the evidence M. is able to tease out and compare the character of the regions through an assessment of the proportions of residences to streets to baths and so on. The final chapter goes beyond the Notitia to return to aspects of M’s earlier studies, such as the Theodosian code. But here he concludes that “the Notitia is more like a key to the understanding of the city of Constantinople than a true representation of it. Its interpretation lies in the connection of this limited, highly structured text with the open-structured volatile material that is open to us. We are looking, and the Notitia provides the framework at the emergence of an international city from the fabric of the Classical world” (218). Whilst there are parts of this study which some students might question, this is a contribution to history of the City we cannot ignore.