1 Armi, Edson, Design and Construction in Romanesque Architecture (Cambridge, 2004), p. 3.Google Scholar

2 See Ćurčić, Slobodan, ‘Some Reflections on the Flying Buttresses of Hagia Sophia in Istanbul’, Metin Ahunbay'a Armagan, Bizans Mimarisi Uzerine Yazilar (Istanbul, 2004), pp. 7–22 (pp. 7-8).Google Scholar

3 Theophanes the Confessor (c. 760-818) recorded two devastating earthquakes at Salamis in 331 and 34: Chronographia, AM 5824, AM 5834.

4 Theodoret, , Historia ecclesiastica, II, 26–30.Google Scholar

5 Malalas, John, Chronographia, XII, 48.Google Scholar For Constantius II's building campaign and his activity in Antioch, see Henck, Nick, ‘Constantius Ο Φιλοκτίστης?’, Dumbarton Oaks Papers, 55 (2001), pp. 279–304 (pp. 294-98).CrossRefGoogle Scholar

6 As many as 35% of the surviving milestones were erected in the time of Constantius II; see Bekker-Nielsen, Tønnes, The Roads of Ancient Cyprus (Copenhagen, 2004).Google Scholar

7 Pouilloux, J., ‘Les Trois Gymnases de Salamine de Chypre’, Revue archéologique, 2 (1966), pp. 337-40.Google Scholar

8 A.H.S. Megaw provided a series of excavation notes entitled ‘Archaeology in Cyprus …'in the journal Archaeological Reports from 1955 to 1958 (vols 2-5); Vassos Karageorghis later continued these reports as 'Chroniques des fouilles …’ in the Bulletin de correspondance hellénique from 1959 to 1973 (vols 84-97); also Karageorghis, V., ‘Aνασĸαφαί Σαλαμινoς?’, Report of the Department of Antiquities Cyprus (1966), pp. 13–19 Google Scholar. The French excavations at Salamis/Constantia between 1964 and 1974 are the subject of ongoing publications in a series called Salamine de Chypre published by E. De Boccard of Paris. It should be noted that archaeological excavation today at Constantia (Salamis) is prohibited by the Republic of Cyprus. My discussion here is based on previous excavation reports and first-hand observation of the above-ground monuments.

9 Karageorghis, V., ‘Chroniques des fouilles en 1965’, Bulletin de correspondance hellénique, 90 (1966), pp. 297–389.Google Scholar

10 Regarding the history of the Gothic flying buttress, see Prache, Anne, ‘Les Arcs-boutants au XIIe siècle’, Gesta, 15.1/2 (1976), pp. 31–42 CrossRefGoogle Scholar; James, J., ‘Evidence for Flying Buttresses before 1180’, Journal of the Society of Architectural Historians, 51.3 (1992), pp. 261-87CrossRefGoogle Scholar; Clark, William and Mark, Robert, ‘The First Flying Buttresses: A New Reconstruction of the Nave of Notre-Dame de Paris’, The Art Bulletin, 66.1 (1984), pp. 47–65.CrossRefGoogle Scholar

11 ’Chroniques des fouilles et découvertes archéologiques à Chypre en 1966’, Bulletin de correspondance hellénique, 91 (1967), pp. 363-70 (p. 354).Google Scholar

12 Lancaster, Lynne, Concrete Vaulted Construction in Imperial Rome (Cambridge, 2005), pp. 145-46CrossRefGoogle Scholar; see also Krautheimer, Richard, Early Christian and Byzantine Architecture (New Haven, 1986), p. 232.Google Scholar There are even earlier examples of pier-buttresses, necessary for vaulting large spaces, such as those of the third-century baths at Bir el Jabbana (Libya); see Rossiter, J., ‘A Roman Bath-House at Bir el Jebbana’, Carthage Papers (Portsmouth, RI, 1998), pp. 103-15.Google Scholar

13 Here I am contrasting Tetrarchan and Constantinian buildings with the earlier Colosseum and Pantheon. These earlier examples have classical order on their exteriors, although they are built of brick and concrete and utilize round arches — which have no necessary dependency on the classical orders as representative of post-and-lintel construction. Ward-Perkins describes this shift (from an exterior to interior aesthetic) as a ‘local’ Roman development: see Ward-Perkins, J.B., Roman Imperial Architecture (New Haven, 1981), pp. 98–99.Google Scholar I thank the anonymous reviewer who assisted me in locating this citation.

14 Lancaster, , Concrete Vaulted Construction, pp. 134-38Google Scholar; see also Ward-Perkins, , Roman Imperial Architecture, p.88, fig. 39Google Scholar; MacDonald, William L., The Architecture of the Roman Empire (New Haven, 1982), pp. 90–93 and 129-36.Google Scholar

15 Anderson, J., ‘The Date of the Thermae Traiani and the Topography of the Oppius Mons’, American Journal of Archaeology, 89.3 (1985), pp. 499–509 CrossRefGoogle Scholar; DeLaine, Janet, The Baths of Caracalla: a Study in the Design, Construction, and Economics of Large-Scale Building Projects in Imperial Rome (Portsmouth, RI, 1997), p. 46.Google Scholar

16 As explained below, the vaulted frigidarium at the Baths of Caracalla (24 m x 55 m, with an area of 1,320 sq.m) and that of the Baths of Diocletian (23 m x 59 m; 1,357 sq. m) required exterior buttressing. The area of the frigidarium of the Baths of Trajan was comparable (25 m x 60 m; 1,500 sq. m).

17 Aitchison, George, ‘What is Architecture, and How Can it be Advanced?’, Architecture and Building, 18 (25 March 1893), pp. 135-37 (p. 136)Google Scholar; see also Coldstream, Nicola, Masons and Sculptors (Toronto, 1991), p. 58.Google Scholar

18 DeLaine, , Baths of Caracalla, p. 46.Google Scholar

19 Roman vaulting, such as for the Pantheon dome, was designed with coffers sometimes decorated with stars to mimic the heavens; see MacDonald, , The Architecture of the Roman Empire, p. 117 Google Scholar; Semper, Gottfried, Style in the Technical and Tectonic Arts, or Practical Aesthetics, ed. and Mallgarve, tr. H. F. and Robinson, M. (Los Angeles, 2004; originally published in 1863), pp. 147–49.Google Scholar

20 The north-east buttresses are of similar design to those of the Markets of Trajan. We can consider the north east buttresses as indirect prototypes for the four gigantic sixth-century pier-buttresses at Hagia Sophia in Constantinople. Because the Baths of Diocletian have been heavily restored in subsequent centuries, our understanding of their initial design depends on early drawings as well as the surviving archaeology. For such drawings, see Sestieri, Anna and others, Museo nazionale romano: the Baths of Diocletian (Milan, 2005), pp. 7—31.Google Scholar

21 Compare my diagram (Fig. 7) with the elevation of the Baths of Diocletian in Rivoira, G.T., Roman Architecture and its Principles of Construction (New York, 1972), p. 208 Google Scholar; see also DeLaine, , Baths of Caracalla, p. 46 Google Scholar; Minoprio, Anthony, ‘A Restoration of the Basilica of Constantine, Rome’, Papers of the British School at Rome, 12 (1932), pp. 1–25 (p. 8)CrossRefGoogle Scholar; Ferretti, A., ‘The Structures of the Basilica’, in The Basilica of Maxentius: the Monument, its Materials, Construction, and Stability, ed. Giavarini, Carlo (Rome, 2005), pp. 161–224 (pp. 161-62).Google Scholar

22 Minoprio, , ‘Restoration of the Basilica’, pp. 6–8.Google Scholar Auguste Choisy, the pioneering architectural historian, credited Roman bath complexes and the Basilica Nova as developing the design concepts behind the contrefort, which is essentially the same concept of the arc-boutant in Gothic architecture; see Choisy, Auguste, L'Art debâtir chez les Romains (Paris, 1873), pp. 88–101.Google Scholar

23 Dikigoropoulos, Andreas, ‘Cyprus “betwixt Greeks and Saracens”: A.D. 647-965’ (doctoral thesis, Oxford University, 1961), pp. 22–23;Google Scholar Megaw, A.H.S., ‘Archaeology in Cyprus, 1958’, Archaeological Reports, 5 (1958), pp.25–34 (pp. 31-32).CrossRefGoogle Scholar Megaw had originally attributed this inscription to Emperor Justinian and Empress Theodora on the basis that the ‘Titulature Inscription’ from Chytroi has similar epigraphy, before suggesting the regional earthquake of 528 as the catalyst for this later building phase at Constantia; see Megaw, , ‘Archaeology in Cyprus, 10 1957’, Archaeological Reports, 4 (1957), p. 47.Google Scholar He would later favour a seventh-century date.

24 See Bruce-Mitford, Terrence and Nicolaou, Ino, The Greek and Latin Inscriptions from Salamis (Nicosia, 1974), p. 6.Google Scholar For a thorough overview of the Arab occupation of Cyprus, see Christides, Vassilios, The Image of Cyprus in the Arabic Sources (Nicosia, 2006).Google Scholar

25 As mentioned in n. 23, Megaw originally dated this inscription to the time of Justinian. Bruce-Mitford, and Nicolaou, continued to support that date (The Greek and Latin Inscriptions, pp. 69–70)Google Scholar; however, Sodini persuasively argued the inscription had to be later, tentatively suggesting the time of Tiberius II (reg. 578-82). A weakness of Sodini‘s proposal that he recognized was the fact that emperor’s name was chiselled out, which meant a damnatio memoriae was issued; see Sodini, J.-P., ‘Une titulature faussement attribuée à Justinien Ier. Remarque sur une inscription trouvée à Kythrea, Chypre’, Travaux et mémoires du Centre de Recherche d'Histoire et Civilisation de Byzance, 5 (1973), pp. 373-84.Google Scholar Phocas was the only Byzantine emperor (reg. 602-10) who unquestionably garnered such hostility; see Ostrogorsky, George, History of the Byzantine State (New Brunswick, NJ, 1969), p. 85 Google Scholar; Varner, Eric, Mutilation and Transformation: damnatio memoriae and Roman Imperial Portraiture (Leiden, 2004), p. 8.Google Scholar There is a slight possibility that the Emperor Maurice (reg. 582-602) was treated this way, although not in Cyprus or Egypt where he was popular; see Efthymiadis, Stephanos, Ashgate Research Companion to Byzantine Hagiography (Farnham, 2011), p. 53.Google Scholar

26 Stewart, C.A., ‘Flying Buttresses & Pointed Arches in Byzantine Cyprus’, in Masons at Work, ed. Ousterhout, Robert and others (Philadelphia, 2012), pp. 1–23 (pp. 5-6); see also n. 23.Google Scholar

27 For the important role that the pier-buttresses played at S. Lorenzo in Milan (fourth century), see Kleinbauer, W. E., ‘” Aedita in Turribus“: the Superstructure of the Early Christian Church of S. Lorenzo in Milan’,Gesta, 15.1/2 (1976), pp. 1–9 CrossRefGoogle Scholar; for the general discussion, see Krautheimer, , Early Christian and Byzantine Architecture, pp. 110–12, 131, 137-38 and 151.Google Scholar There is still no consensus about whether all aisled-tetraconch buildings, such as Seleucia Pieria, were vaulted in stone; see Kleinbauer, W.E., ‘The Origin and Functions of the Aisled Tetraconch Churches in Syria and Northern Mesopotamia’, Dumbarton Oaks Papers, 27 (1973), pp. 89–114 (p. 89)CrossRefGoogle Scholar; and idem, ‘The Double-Shell Tetraconch Building at Perge in Pamphylia and the Origin of the Architectural Genus’, Dumbarton Oaks Papers, 41 (1987), pp. 277-93 (p. 280)CrossRefGoogle Scholar. In my opinion it seems that the chief function of the ‘double-shell’ was to distribute the weight of roof both to the columns beneath and to the outer walls, indicating a substantial burden over the central space. The most monumental example of this type, S. Lorenzo in Milan, was originally designed to carry brick or some other lightweight material, as was S. Vitale at Ravenna.

28 We know much about Justinian's building campaigns from Procopios who recorded many imperial investments, no matter how minor, but who hardly mentioned Cyprus; see n. 37.

29 Dr Karageorghis has kindly affirmed that the flying buttresses could be of an Early Byzantine date (personal communication, 2 December 2011).

30 For a survey of construction methods on Cyprus, see Wright, G., Ancient Building in Cyprus (Leiden, 1992).Google Scholar Regarding the walls of Constantia, see Stewart, C.A., ‘Early Byzantine Military Architecture in Cyprus’, in The Archaeology of Late Antique and Byzantine Cyprus (4th-12th centuries AD), ed. Parani, Maria and Michaelides, Demetrios (Paris, forthcoming).Google Scholar

31 Chytroi was also an important bishopric in the Byzantine period: Gregoire, H., ‘Saint Demetrianos, évêque de Chytri (ile de Chypre)’, Byzantinische Zeitschrift, 16 (1907), pp. 209-12CrossRefGoogle Scholar; Jenkins, R.J., ‘The Mission of St. Demetrianos of Cyprus to Bagdad’, Annuaire de I'Institut de Philologie et d'Histoires Orientates et Slaves, 9 (1949), pp. 267-75.Google Scholar

32 Munro, A. et. al., ‘Excavations in Cyprus, 1890, Salamis’, Journal of Hellenic Studies, 12 (1891), pp. 59–198 (pp. 90-91 for map).Google Scholar

33 For the analysis of the water capacity of the aqueduct, see Baur, A., ‘Die Wasserversorgung der antike Stadt Salamis auf der Insel Zypern’, Schriftenreihe der Erontinus-Gesellschaft, 11 (1989), pp. 203-18 (pp. 208-11).Google Scholar

34 For the Neronian inscription, see Nicolaou, Ino, ‘Inscriptions Cypriae Alphabeticae II’, Report of the Department of Antiquities Cyprus (Nicosia, 1963), pp. 41–51 (p. 48)Google Scholar; Supplementum Epigraphicum Graecum, 23 (1968), n. 675.Google Scholar

35 Vessberg, O. and Westholm, A., The Swedish Cyprus Expedition: The Hellenistic and Roman Periods in Cyprus (Stockholm, 1956), p. 17 Google Scholar; Bruce-Mitford, T., ‘Roman Cyprus’, in Aufstieg und Niedergang der römischen Welt, ed. Temporini, H. and others (Berlin, 1980), pp. 1285-384 (pp. 1323-24).Google Scholar Later in the Byzantine period, Procopios mentioned the ‘poor-house of St Conôn’ and that the emperor ‘renewed the aqueduct of the same in Cyprus'; see Procopios, Buildings, V, 9,36. Some scholars have assumed this poor-house was located at Chytroi, although St Conôn was venerated 70 miles to the west on the Akamas Peninsula; see Hill, George, A History of Cyprus I: to the Conquest of Richard Lion Heart (Cambridge, 1940), p. 280. 24 Google Scholar

36 See nn. 23 and 25. Other inscriptions discovered within the baths in 1958 mention a hypatikos and consularis named Ioannes who restored the sudatorium, while also referring to the Emperor Justinian. Bruce-Mitford and Nicolaou date this to AD 542/3 (The Greek and Latin Inscriptions, p. 76).

37 According to Oberhummer, who conducted surveys around Chytroi, the aqueduct was ‘partly constructed above ground, and partly lying on the ground as a channel’; see Oberhummer, Eugen, Die Insel Cypern; eine Landeskunde auf historischer Grundlage (Munich, 1931), p. 232.Google Scholar Perhaps most of the older Roman aqueduct was at ground level and consequently less conspicuous and easier to dismantle over the centuries.

38 The second surviving inscription has been eroded and is difficult to make out. A final cross is visible and the remains of the word ‘archbishop’ (αϱχιεπσĸoóπoν); because of its size and epigraphy we can assume it followed a similar formula to the inscription quoted here.

39 Sodini, J.-P., ‘ Epigraphica: Notes sur quelques inscriptions de Chypre’, Travaux et mémoires du Centre de Recherche d'Histoire et Civilisation de Byzance, 4 (1970), pp. 477-86Google Scholar; idem, ‘Les inscriptions de l'aqueduc de Kythrea à Salamine de Chypre’, Eνψνxiα: Mélanges offerts à Hélène Ahrweiler (Paris, 1998), pp. 619-34 (pp. 626-28)Google Scholar; see also Feissel, Denis, in Salamine de Chypre XIII. Testimonia salaminia 2: Corpus épigraphique (Paris, 1987), pp. 83–85 Google Scholar; and Bruce-Mitford, T., ‘Some New Inscriptions from Early Christian Cyprus’, Byzantion, 20 (1950), pp. 105-75 (pp. 128-32).Google Scholar

40 Sodini, , ‘Les inscriptions’, p. 633.Google Scholar While no Byzantine source records the Archbishop Ploutarchos, his successor Archbishop Arcadios is mentioned by seventh-century authors, such as John, Bishop of Nikiu in Egypt (Chronicle, 120, 64) and the Syrian Bishop, George of Reshaina; see Charles, R.H., The Chronicle of John, Bishop of Nikiu (Oxford, 1916), p. 190 Google Scholar; Brock, Sebastian, ‘An Early Syriac Life of Maximus the Confessor’, Analecta bollandiana, 91 (1973), pp. 299–346 (pp. 315-17).CrossRefGoogle Scholar

41 Sodini, , ‘Les inscriptions’, p. 623.Google Scholar In comparison, the Trajanic inscriptions of the Caesarea Maritima Aqueduct in Israel are located in the same positions in the spandrels below the water channel. These record the tenth legion as its builders.

42 Although the Byzantine dating of the aqueduct is straightforward, some prominent publications have described the structure as Gothic; see Enlart, Camille, L'Art gothique et la Renaissance en Chypre (Paris, 1899), p. 514 Google Scholar; Pococke, Richard in Cobham, CD., Excerpta Cypria (Cambridge, 1908), p. 256 Google Scholar; Gunnis, Rupert, Historic Cyprus (London, 1936), p. 420 Google Scholar; Wright, G.R.H., Ancient Building in Cyprus (Leiden, 1992), p. 233.Google Scholar Most archaeologists and historians, both past and present, reject a medieval date, since Famagusta did not have an aqueduct; see Oberhummer, , Die Insel Cypern, p. 232 Google Scholar; Megaw, A.H.S., ‘Betwixt Greeks and Saracens’, Acts of the International Archaeological Symposium ‘Cyprus Between the Orient and the Occident’, ed. Karageorghis, V. (Nicosia, 1986), pp. 505-19 (p. 508 n. 17)Google Scholar; Jeffery, George, A Description of the Historic Monuments of Cyprus (Nicosia, 1918), p. 233.Google Scholar

43 This church has well-preserved frescos, which are reliably dated to the early twelfth century, so the structure was built sometime earlier; see Winfield, D., ‘Hagios Chrysostomos, Trikomo, Asinou: Byzantine Painters at Work’, in Πρακτικά Πρώτoν Διεθvovς Kυπρoλoγικov Συvεδρioυ, 3 vols (Nicosia, 1972), II, pp. 285-91.Google Scholar

44 During the reign of Justinian, the province of Cyprus was placed under the military protection of the quaestor exercitus, whose headquarters were in Odessus (modern Varna in Bulgaria); see Haldon, John, Byzantium in the Seventh Century: the Transformation of a Culture (Cambridge, 1990), p. 210.CrossRefGoogle Scholar

45 Hodge, A.T., Roman Aqueducts and Water Supply (London, 2002), pp. 93–99.Google Scholar

46 Bruce-Mitford, , ‘Some New Inscriptions’, p. 122 n. 2.Google Scholar Loutron means ‘bath’ and is what the locals called it in the seventeenth century. Although this was a mistaken designation, it is still used in archaeological literature.

47 Argoud, G. and others, ‘Le Temple de Zeus’, Report of the Department of Antiquities Cyprus (1975), pp. 122-41 (pp. 140-41).Google Scholar

48 Cobham, , Excerpta Cypria, p. 25.Google Scholar

49 Hill, , History of Cyprus, p. 242.Google Scholar The population of Constantia could have been much larger since there were other sources of water nearby such as the Pedeios River. This was the longest river on Cyprus (110 km) and its delta formed the southern boundary of Constantia.

50 Baur incorrectly reconstructed the Loutron with nine barrel-vaults running in the north-south direction, in a way that resembles the first-century Grand Cistern at Bordj Djedid at Carthage, Libya; see Baur, A., ‘Die Wasserversorgung’, p. 214.Google Scholar

51 Wright incorrectly claimed that it had ‘vaulted roofing carried on 39 piers and 12 responds’; see Wright, , Ancient Building, p. 154.Google Scholar

52 Ibid., p. 89.

53 Ibid., pp. 86 and 89.

54 The excavators concluded that the renovation belonged to the ‘later city’ of Constantia, and so implied a Byzantine date; ibid., pp. 89-91.

55 See the drawing illustrated in Munro and others, ‘Excavations in Cyprus’, p. 87.

56 Roman cisterns that come to mind are the Piscina Mirabilis (Bacoli, Italy), Aïn Mizeb and Aïn El Hammam (Thugga, Tunisia), and Aptera (Crete). Unfortunately, the groin-vaulted cisterns at Alexandria (Egypt), such as the El-Nabih, are currently inaccessible and cannot be reliably dated without comprehensive archaeological examinations.

57 Winogradov, Zalman ‘The Aqueduct of Tiberias’, in The Aqueducts of Israel, ed. Amit, D. and others (Portsmouth, RI, 2002), pp. 295–306 (p. 306).Google Scholar

58 These massive pier-buttresses were called ‘spurs’ by the excavators; see Munro, and others, ‘Excavations in Cyprus’, p. 86.Google Scholar

59 For Palestinians fleeing to Byzantine Cyprus, see Kyrris, Costas, ‘The “Three Hundred Alaman Saints” of Cyprus: Problems of Origin and Identity’, in The Sweet Land of Cyprus, ed. Bryer, A. and others (Nicosia, 1993), pp. 203-35Google Scholar; for Egyptian refugees, see Leontius of Neapolis, Vita S. johannis Ellemosynarii.

60 Regarding the decline of the Byzantine Empire, see Kountoura-Galake, Eleonora, ed., The Dark Centuries of Byzantium (7th-9th c.) (Athens, 2001).Google Scholar In contrast, for the flourishing of art and architecture on Cyprus, see Rautman, Marcus, A Cypriot Village of Late Antiquity (Portsmouth, RI, 2003)Google Scholar; and Stewart, C.A. and others, eds, Cyprus and the Balance of Empire: Art and Archaeology from Justinian to the Cœur de Lion (Boston, 2014).Google Scholar

61 Grierson, Philip, Catalogue of the Byzantine Coins in the Dumbarton Oaks Collection and in the Whittemore Collection: Phocas to Theodosius III, 602-717 (Washington, DC, 1992), p. 41 Google Scholar; Metcalf, D.M., Byzantine Cyprus 491-1191 (Nicosia, 2009), pp. 143-46.Google Scholar

62 Metcalf, , Byzantine Cyprus, pp. 151-55.Google Scholar

63 This was the view of Chrysos which I find persuasive given all the textual and archaeological evidence; see Chrysos, Evangelos, ‘O ‘Hϱακλειoς στην Kνπϱo (609-610)’, in Πρακτικα Συμπoςιoν Kνπριακης lστoριας (Nicosia, 1984), pp. 53–62 Google Scholar; for a counter-perspective, see Kaegi, Walter, Heraclius, Emperor of Byzantium (Cambridge, 2003), pp. 47–48.Google Scholar

64 Brock, , ‘An Early Syriac Life’, pp. 315-17.Google Scholar A letter dated AD 625 from Sergius, Patriarch of Constantinople, to Bishop Cyrus of Phasis (Georgia) is quoted in the twelfth session of the Third Council of Constantinople in 680 and mentions Arcadios's role in the implementation of Monotheletism; see Clavis patrum graecorum, ed. Geerard, M. and others (Turnhout, 1974-2003), col. 7604.Google Scholar

65 Bury, J.B., A History of the Late Roman Empire: From Arcadius to Irene, 2 vols (New York, 1889), II, p. 251.Google Scholar

66 Krueger, Derek, Symeon the Holy Fool: Leontios's Life and the Late Antique City (Berkeley, CA, 1996), pp. 12–14.Google Scholar

67 For general histories that describe economic reforms, see Haldon, Byzantium in the Seventh Century and Kaegi, Heraclius. For a survey of cultural renewal, see Reinink, Gerritt and others, The Reign of Heraclius (610-641) (Leuven, 2002)Google Scholar; for literary achievements, see Pertusi, A., Giorgio Pisidia, Poemi: I. Panegirici epici (Ettal, 1959)Google Scholar; for the flourishing of art, see Evans, Helen and others, Byzantium and Islam: Age of Transition, 7th-9th Century (New York, 2012).Google Scholar

68 Mark has argued that the massive pier-buttresses were ineffective, although his critique is directed at the materials — brick and thick mortar — rather than the ‘elegant’ design; see Mark, Robert, Light, Wind and Structure: the Mystery of the Master Builders (New York, 1990), pp. 85–89 Google Scholar; cf. Rivoira's insights (Roman Architecture, pp. 207-08).

69 Sedlmayr, Hans, ‘Spätantike Wandsysteme’, in Epochen und Werke (Vienna, 1959), pp. 87–99.Google Scholar The roots of this concept, once again, go back to Apollodorus of Damascus's design of the Baths of Trajan. Choisy identified this design theory as ‘self-bracing’ technology; see Etlin, Richard, ‘Auguste Choisy's Anatomy of Architecture’, in Auguste Choisy (1841-1909): L'Architecture et I'art de bâtir. Adas del simposio iniernacional celebrado en Madrid, ed. Girón, Javier and Huerta, Santiago (Madrid, 2009), pp. 151-81 (pp. 159-62).Google Scholar

70 Gerola, Giuseppe, ‘II sacello primitivo in San Vitale’, Felix Ravenna, 10 (1913), pp. 427-34; and 11 (1913), pp. 459-71Google Scholar; idem, ‘Le volte delle loggie e la decorazione delle pareti in San Vitale’, in Alii del Reale Istituto Veneto di Scienze, Lettere ed Arti, 75 (1915-16), pp. 827-39Google Scholar; see also Deichmann, F.W., Ravenna: Hauptstadt des spätantiken Abendlandes, 3 vols (Wiesbaden, 1976), II.2, p. 51.Google Scholar

71 Rivoira, , Roman Architecture, p. 208.Google Scholar

72 Mainstone, Rowland, Hagia Sophia (London, 1988), pp. 102–05.Google Scholar

73 Ćurčić, , ‘Some Reflections’, pp. 7–22.Google Scholar

74 Sodini, , ‘Les inscriptions’, p. 623 Google Scholar; Krautheimer, , Early Christian and Byzantine Architecture, p. 248 Google Scholar; Draper, Peter, ‘Islam and the West: the Early Use of the Pointed Arch Revisited’, Architectural History, 48 (2005), pp. 1–20 (pp. 3-4).Google Scholar

75 For Bethesda, see Rousée, J.-M, ‘Chronique archéologique. Jérusalem (Piscine probatique)’, Revue biblique, 69 (1962), pp. 107-09Google Scholar; for Emmaus, see Hirschfeld, Yizhar, ‘The Aqueducts of Emmaus-Nicopolis’, in Amit, and others, The Aqueducts of Israel, pp. 187-98 (pp. 187-88).Google Scholar

76 Hirschfeld, , ‘The Aqueducts’, p. 188.Google Scholar

77 Regarding pointed arches in Middle Byzantine architecture on Cyprus, see Stewart, C.A., ‘Architectural Development in Byzantine Cyprus’, in Cyprus and the Balance of Empires, ed. Stewart, and others, pp. 107-34 (pp. 124-26).Google Scholar

78 Hirschfeld, Yizhar, Excavations at Tiberias, 1989-1994 (Jerusalem, 2004), pp. 75–134.Google Scholar It is unclear from the excavation report whether these pointed arches are Byzantine or date from the Abbasid rebuilding. Additionally, the earliest surviving Islamic monuments with pointed arches were also waterworks, such as the baths at Hammam as-Sarakh (mid-seventh century) and the so-called ‘Romanesque’ decorative arches of the nilometer in Cairo (dated to 861); see Creswell, K.A.C., Early Muslim Architecture, 2 vols (Oxford, 1969), I, pp. 442-49, II, p. 200Google Scholar; see also the discussion in Draper, , ‘Islam and the West’ (pp. 4–11).Google Scholar

79 Creswell, K.A.C., A Short Account of Early Muslim Architecture (Harmondsworth, 1958), pp. 228-30.Google Scholar

80 Meyendorff, John, Byzantine Theology: Historical Trends and Doctrinal Themes (New York, 1979), pp. 19–53. and 134-37.Google Scholar

81 Maximos the Confessor (580—662) provided a symbolic interpretation of Byzantine architecture: see Maximos the Confessor, Mystagogia, ed. Cantarella, Raffaele (Florence, 1931).Google Scholar These ideas would influence the Patriarch of Constantinople Germanos I's writings on the divine liturgy in the early eighth century. For the relationship of Maximos and ‘Neoplatonic architecture', see Stewart, C.A., Domes of Heaven: the Domed Basilicas of Cyprus, published doctoral thesis, Indiana University (Ann Arbor, MI, 2008), pp. 173-78;Google Scholar and Ozoline, Nicholas, ‘La Symbolique Cosmique du temple chrétien selon la Mystagogie de Saint Maxime le Confessor’, Byzantinorossica, 1 (1995), pp. 30–38.Google Scholar

82 Bahat has persuasively laid out the arguments for a Heraclian date; see Bahat, Dan, ‘The Golden Gate and the Date of the Madaba Map’, The Madaba Map Centenary 1897-1997, ed. Piccirillo, M. and others (Jerusalem, 1999), pp. 254-56.Google Scholar The argument for an Umayyad date is provided by Rosen-Ayalon, Myriam, The Early Islamic Monuments ofal-Haram al-Sharif: an Iconographic Study (Jerusalem, 1989); see especially pp. 33–45.Google Scholar

83 Draper, , ‘Islam and the West’, pp. 4–11.Google Scholar