* This is published as a lecture, but a few references with selected bibliography have been appended.

1 Robert, Hillenbrand, ‘Studying Islamic Architecture: challenges and perspectives’, Architectural History, 46 (2003), pp. 1–18.Google Scholar

2 The horseshoe profile on the interior of the entrance arch into the Pantheon is one such unusual example, as Eric Fernie pointed out to me after the lecture.

3 Report on St Paul’s in Wren Society, ed. A. T. Bolton, 11, pp. 15-20.

4 Paul Frankl, The Gothic: Literary Sources and Interpretations through Eight Centuries (Princeton, 1960), p. 376.

5 Jean Bony in his French Gothic Architecture of the Twelfth and Thirteenth Centuries (Berkeley, 1983), pp. 16-17 and 466-68, notes 8, 13 and 14, argued that, ‘The pointed arch was probably a schematization of the parabolic curve used by Sasanian builders since the early third century for the construction without centring of domes and barrel vaults (e.g. the huge iwan of the Taq-i-Kisra at Ctesiphon)’.

6 Paol, Cuneo, ‘La Basilique de Tsitsernakavank (Cicernakavank) dans le Karabagh’, Revue des Etudes Armeniennes, 4 (1967), pp. 203–16.Google Scholar

7 Hamilton, R. W., Khirbat al Mafjar. An Arabian Mansion in the Jordan Valley (Oxford, 1959).Google Scholar

8 For succinct and accessible discussion of these issues, see Richard, Ettinghausen, Oleg Grabar and Marilyn Jenkins-Medina, Islamic Art and Architecture 650-1250, Pelican History of Art (New Haven and London, 2001), pp. 3–8; Hillenbrand, Robert, Islamic Architecture: Form, Function and Meaning. (Edinburgh, 1994), pp. 31–36; and Grabar, Oleg, The Formation of Islamic Art (New Haven and London, 1973, rev. 1987).Google Scholar

9 The literature on the Dome of the Rock is extensive, but for important recent discussions, see Grabar, Oleg, The Shape of the Holy: Early Islamic Jerusalem (Princeton, 1996); Bayt al-Maqdis: Jerusalem and Early Islam, ed. Johns, Jeremy (Oxford, 1999); and Hillenbrand, Robert, ‘The Legacy of the Dome of the Rock’, Studies in? Medieval Islamic Architecture, I (London, 2000), pp. 1–l20.Google Scholar

10 An observation curiously omitted by Creswell in his list of early examples: Cresswell, K. A. C., A Short Account of Early Muslim Architecture, ed. Allen, James W. (Oxford, 1989), p. 116,Google Scholar and not commented upon by Grabar in The Shape of the Holy.

11 Allen, H. R., ‘Observations on the Original Appearance of the Dome of the Rock’, in Bayt al-Maqdis, pp. 197–213.Google Scholar

12 Northedge, Alastair, Studies on Roman and Islamic Amman I: History, Site and Architecture (Oxford, 1992).Google Scholar

13 A Medieval Islamic City Reconsidered: An Interdisciplinary Approach to Samarra, ed. Chase F. Robinson, Oxford Studies in Islamic Art xiv (Oxford, 1999), pp. 18-19 ^and 96. For the method of brick construction of these vaults, see K. A. C. Creswell, Early Muslim Architecture, n (1st edn, Oxford, 1940), p. 61.

14 Bloom, Jonathan, ‘On the Transmission of Designs in Early Islamic Architecture’, Muqarnas, 10 (1993), pp. 21–28 (p. 23)Google Scholar, notes that later medieval Egyptian writers attempted to account for these manifestly imported features in other ways, suggesting that the brick piers were intended as a precaution against fire and flood; and that columns had not been used (as would be expected in Cairo) because they had been tainted by prior use in a Christian church, a good example of symbolic value being attached to an architectural feature.

15 Cresswell, , Short Account of Early Muslim Architecture, p. 116.Google Scholar

16 This argument was put to the test by John Warren, ‘Cresswell’s Use of the theory of Dating by the Acuteness of the Pointed Arches in Early Muslim Architecture’, Muqarnas, 9 (1991), pp. 59-65. Despite questioning one or two of Cresswell’s examples and noting a number of omissions, Warren concluded that Cresswell’s hypothesis could be vindicated. The idea of development was followed in Andrew Petersen, Dictionary of Islamic Architecture (London, 1996), p. 24, where he notes the change from the slightly pointed arches of the Dome of the Rock (691) to the more decisive pointed profile in the cisterns at Ramla (759) where the centres of the arcs are one-fifth of the span apart, a construction which became fairly standard in early Islamic buildings.

17 On this, see Hillenbrand, Robert in Bayt al-Maqdis, pp. 304–05.Google Scholar

18 A relief in the Burgos Museum (room 3, no. 118) and on a Roman stele in Léon, Sant Marc museum, illus. in Puig y Cadafalch, L'Arquiitectura Romana a Catalunya (1934), p. 378, fig. 511.1 am indebted to Eric Fernie for bringing these examples to my attention.

19 Camilla and David Edwards, ‘The Evolution of the Shouldered Arch in Medieval Islamic Architecture’, Architectural History, 42 (1999), pp. 68-95.

20 Painted terracotta fragments of what could have been a muqarnas vault indicate that the form could date from as early as the tenth century. See Ettinghausen, Grabar, and Jenkins-Medina, Islamic Art and Architecture, p. 115 and figs 176, 178. See also Laleh, H., ‘Muqarnas: remarques sur la conception géométrique et les origins architecturales’, in Art and Archaeology of Ancient Persia. New Light on the Parthian and Sasanian Empires, ed. Vesta Sarkosh Curtis, Robert Hillenbrand and J. M. Rogers (London, 1998), pp. 150–61; Bloom, J. M., ‘The Introduction of Muqarnas into Egypt’, Muqarnas, 5 (1988), pp. 21–28.Google Scholar

21 Hillenbrand, , ‘The Legacy of the Dome of the Rock’. Studies in Medieval Islamic architecture, 1 (London, 2000), pp. 1–20.Google Scholar

22 Harvey, John, ‘The Origins of Gothic Architecture: Some Further Thoughts’, Antiquaries Journal, 48 (1968), pp. 87–99.Google Scholar See also Arthur Upham Pope, ‘Possible Iranian Contributions to the Beginning of Gothic Architecture’, Beitrage zur Kunstgeschichte Asians: in memoriam Ernest Diaz, ed. Oktay Aslanapa (Istanbul, 1963), pp. 1-30.

23 As to other craftsmen, there is also the dubious case of ‘Lalys the mason’ who appears in John Harvey’s English Medieval Architects: a Biographical Dictionary down to 1550, and is said to have been a Saracen captured on crusades and later to have become architect to Henry I, but the evidence is thin.

24 Conant, Kenneth J., Carolingian and Romanesque Architecture 800-1200, Pelican History of Art, 2nd edn (Harmondsworth, 1966), p. 118.Google Scholar

25 Ibid., p. 144. According to Bony, the first appearance of the pointed arch in a ‘properly Romanesque context’ was at Cluny in the 1090s, certainly by the 1120s, arguing that its importance is shown by its use not only for arches but for the profile of the barrel vault, which had the effect of reducing the (lateral) thrusts by about 20 per cent, ‘and this technical advantage explains its rapid diffusion in some areas of Romanesque Europe’; Bony, French Gothic Architecture, p. 18.

26 Stalley, Roger, Early Medieval Architecture. (Oxford, 1999), p. 139.Google Scholar