I would like to thank Mary Beard for help and suggestions in the preparation of this paper.

1. As does Krautheimer, R., Rome: profile of a city, 312–1308 (1980)Google Scholar

2. For an account of the excavation with illustrations and bibliography of the site, see Erim, K. T., Aphrodisias: city of Venus Aphrodite (1986)Google Scholar.

3. See especially Reynolds, J., Aphrodisias and Rome (1982)Google Scholar.

4. Roueché, C., Aphrodisias in late antiquity (1989)Google Scholar; hereafter cited as Roueché.

5. The art historian who wants to study the statues which belonged to the inscribed bases will need to supplement this publication: for example inscription 62 publishes the statue base of Flavius Palmatus – without the statue (see 102–4 (and also inscription 63) and pl. xvi). The commentary here has its statutory reference to decline, but the broader implications of artistic ‘decline’ are not considered.

6. The classic formulation is by Squarciapino, M., La scuola di Afrodisia (1943)Google Scholar. See also Erim, K. T. and Roueché, C., ‘Sculptors from Aphrodisias: some new inscriptions’, PBSR 50 (1982) 101–15Google Scholar; and for material from Asia Minor, including Aphrodisias, see Inan, J. and Rosenbaum, E., Röman and early Byzantine portrait sculpture in Asia Minor (1966)Google Scholar and Inan, J. and Rosenbaum, E., Römische und frühbyzantinische Porträtplastik aus der Türkei: neue Funde (1979)Google Scholar.

7. Roueché, esp. xx–xxvii.

8. See for example Foss, C., History and archaeology of Byzantine Asia Minor (1990)Google Scholar, and Mango, C., Byzantium. The empire of New Rome (1980)Google Scholar.

9. Mango (n. 8), 66ff.

10. There are of course no population figures; but the estimate of David Jacoby that the population never exceeded 400,000 is most commonly quoted; the mortality estimate of Mango and others has been seen as exaggerated, see Durliat, J., ‘La peste du VIe siècle. Pour un nouvel examen des sources byzantines’, in Hommes et richesses dans l'empire byzantin I (1989) 107–25Google Scholar.

11. Mango (n.8), 68–9.

12. Roueché, 151.

13. The mirror image of the empirical Mango–Roueché view appears in Treadgold, W., The Byzantine revival 780–842 (1988)Google Scholar. He accepts the idea of decline and a subsequent Dark Age in the seventh and eighth centuries as a contingent part of his belief in revival after it. He sees his framework as a reality, something to extract with accuracy from primary sources: he assures us that the study of Byzantine Greek and Byzantine texts ‘cannot be replaced by any amount of comparison with modern societies’. Yet his terms of explanation do not of course come from his texts.

14. The Greek language and literature of the Middle Ages was repellent to one German scholar who asked a hundred years ago how can anyone study a culture where apo is used with the accusative; and more recently Romilly Jenkins has warned that there is not one single text in the whole of Byzantine literature which can be read for pleasure alone. One wonders what he would have made of the recent identification of intentional wit and humour in Byzantine literature!

15. One example among several is in the work of Liebeschuetz, J. H. G. W. whose reprinted papers have just appeared under the title, From Diocletian to the Arab conquest: change in the late Roman Empire (1990)Google Scholar. See also Barnish, S. J. B., ‘The transformation of Classical cities and the Pirenne debate’, Journal of Roman Archaeology 2 (1989) 385–400CrossRefGoogle Scholar.

16. Even E. H. Gombrich found it difficult to condemn the sixth-century mosaics of San Vitale in Art and illusion, though he is obviously ambivalent in front of them as he recollects the ‘Greek Revolution’ which they succeed: see Beard, M., ‘Reflections on “Reflections on the Greek Revolution”’, Archaeological Review from Cambridge 4:2 (1985) 207–13Google Scholar. Roberts, M., The jewelled style (1989) 4ff.Google Scholar, attributes the promotion of the recognition of the qualities of Late Antique culture to art historians – the pioneers being Alois Riegl and Gerhardt Rodenwaldt – who were followed by Peter Brown and Henri-Irénée Marrou (particularly in his Retractio of 1949 to the reissue of his book of 1938 on Augustin et la fin de la culture antique). Roberts, 148, sees the achievement of Late Antiquity to be the creation of a Romano-Christian synthesis. Chapter three discusses this in relation to ‘Poetry and the Visual Arts’. This is an interesting formal analysis of parallelism between literature and art, esp. 118ff. – where the period is characterised as one of the development of display and elaboration, which would coincide with Roueché's characterisation of the epigraphy.

17. See Clover, F. M. and Humphreys, R. S., ‘Towards a definition of Late Antiquity’ in their Tradition and innovation in late antiquity (1989) 3–26Google Scholar. The debate which this raises is how to recognise and define what constitutes change or continuities. The obvious criticism of their discussion is that they have concentrated too much on formal continuities and not recognised the ideological changes beneath. One contribution to their book which has influenced their views is Allen, T., ‘The Arabesque, the Beveled style, and the mirage of an early Islamic art’ (209–44)Google Scholar, who by emphasising the formal links with early Christian and Byzantine art reached the paradoxical conclusion that there is nothing ‘Islamic’ about Islamic art. The interest of this debate perhaps lies less in the ‘answer’ than the assumptions made in reaching one.

18. Tainter, J., The collapse of complex societies (1988)Google Scholar who quotes Isaac that ‘It goes without saying that the collapse of ancient civilisation is the most outstanding event in history’ and quotes Wilamowitz on the Roman Empire: ‘Civilisation can die, because it has already died once’.

19. Tainter (n. 18) 187.

20. Harvey, A., Economic expansion in the Byzantine empire 900–1200 (1989)Google Scholar.

21. Perkins, B. Ward, From Classical Antiquity to Middle Ages (1984)Google Scholar.

22. His approach is blocked out in a review of Treadgold, W. (n.13) in The International History Review 11 (1989) 313–19Google Scholar, and set out in his forthcoming book on the seventh century.

23. B. Berenson for one had given considerable prominence to the history of Ancient sculpture as a location for the identification of decline: see his Arch of Constantine: the decline of form (1954). This discourse is undercut by Peirce, P., ‘The Arch of Constantine: propaganda and ideology in late Roman art’, Art History 12 (1989) 387–418CrossRefGoogle Scholar.

24. See G. Jacobi, ‘Gli scavi della Missione Archaeologica Italiana ad Afrodisiade’, and Crema, L., ‘I monumenti architettonia afrodisiensi’, in Monumenti antichi pubblicati per cura della reale accademia dei lincei. Roma 38 (1939–1940)Google Scholar; and most recently Roueché, C. and Erim, K. T. (eds), Aphrodisian papers: recent work on architecture and sculpture (Journal of Roman Archaeology, suppl. ser. 1) (1990)Google Scholar, especially part 2: the Temple of Aphrodite. This volume includes my fuller study on the temple as cathedral. See also Cormack, R., The Byzantine eye (1989) ch. 7Google Scholar, ‘The classical tradition in the Byzantine provincial city: the evidence of Thessalonike and Aphrodisias’.

25. For the fear of pagan statuary inhabited by demons, see Mango, C., ‘Antique statuary and the Byzantine beholder’, reprint V in his Byzantium and its image (1984)Google Scholar, and Freedberg, D., The power of images: studies in the history and theory of response (1989)Google Scholar.

26. Roueché, 149–51, dates the change of name to the middle of the seventh century.

27. The fire may have occurred during an invasion in 1188–9 according to Roueché, 155, if the church of the Archangel Michael burnt by Theodore Mangaphas can be identified as this cathedral (which is not without its difficulties). Between the time of conversion and the fire, the church was developed and altered in the course of several restorations. Unfortunately a precise chronology of the Christian period is difficult, in part because more study is needed on the Christian floors (lifted during excavation), on the sarcophagi from the aisles (which give evidence of the changing ideology of death and the location of the tomb monument in the Middle Ages), on the reconstruction of the curved inscribed aedicula of the tenth or eleventh century, and the eleventh-century frescoes found in the synthronon.

28. This part of the site was investigated to identify ruins which rose above ground level. When the function of the structures had been determined, the excavation here was discontinued.

29. See Cormack (n.24).

30. Roueché, 159–61, and pl. xxvi. One piece was found in the church in 1962, the others were stray finds.

31. A case could be formulated that earthquake damage had caused buildings to collapse onto the street and this stretch was abandoned rather than cleared of heavy fallen masonry. But empirical considerations are unlikely to be adequate as a full explanation of the circumstances.

32. It would be possible to develop here a ‘heritage’ argument based on twentieth-century attitudes towards the reuse of the old. The conservation and constructive reuse of redundant buildings is not regarded as a necessary indication of twentieth-century decline. We adapt churches because of our changes in architectural needs; and we criticise the developers of the 1950s and 1960s for tearing down the ‘Past’ instead of incorporating it in new architecture. A notorious case in London was the demolition of the Euston Arch by British Rail in preference to its incorporation in the new station – justified at the time on the grounds that removal was cheaper than preservation in a new setting. Yet in terms of architectural quality, it is by no means clear that the new is preferable to the old. If we can understand this kind of ‘heritage’ argument now, we undercut the condemnation of the Byzantines for renovating the tetrakionon as a new high cupola church. To accept or reject the reuse of old buildings and materials must be seen as an issue of cultural attitudes, see Lowenthal, D., The past is a foreign country (1985)Google Scholar; an indication of the current attention given to architectural conservation is the substantial publication of the Council of Europe Report on religious redundant buildings Doc. 6032 (1989).

33. The sculpture of the scenae frons was still lying where it fell when the theatre was uncovered by excavation.

34. Most attention has been given on the acropolis to the examination of the earlier layers of the site, see Joukowski, M. S., Prehistoric Aphrodisias: an account of the excavations and artifact studies (1983)Google Scholar.

35. As does the whole question of individual masters from Aphrodisias and their identification. Roueché, 27–9, now speaks of proprietors of workshops as the individuals named on inscriptions in preference to seeing names as the signatures of sculptors. But this is a difficult solution, with its implication that sculptors did not run their own establishments.

36. The base of this statue is discussed by Roueché, 102–4, and the statue published by Erim, K. in Porträtplastik (n.6), 236–8Google Scholar. For a broader review of the issues, see Smith, R. R. R., ‘Roman portraits: honours, empresses, and late emperors’, JRS 75 (1985) 209–21Google Scholar.