Skip to main content Accessibility help

Late Antique Knossos. Understanding the city: evidence of mosaics and religious architecture1

  • Rebecca J. Sweetman (a1)


Interpretation of the historical and epigraphical data can only provide a bare outline of the political and social environment of Knossos between the 5th and 7th centuries AD. Consequently, our understanding of late Antique Knossos comes primarily from the excavated remains of two of the Christian basilicas in the Valley; the Sanatorium Basilica and the KMF Basilica. Although excavations of the former have been published in full, concerning the KMF Basilica, only a brief summary of its architecture has been published. As a result, a detailed analysis of the excavations of the basilica itself will be made available here, including particulars of its architecture, mosaics, and summaries of the associated elements of the entire basilica complex (a detailed discussion of the whole Late Antique KMF excavations is forthcoming). As such, not only is a new and full interpretation of the KMF Basilica provided, but also for the first time the material culture of Late Antique Knossos is collated and subsequently analysed to provide a clear understanding of the nature of the city during this period. Despite a dearth of domestic evidence, a contextual study of the surviving architecture (principally religious) in conjunction with the mortuary remains, allows a broad perception of society and a clear comprehension of the development of the city. Thus the aim here is to examine the compelling evidence that, contrary to common belief, during the Late Antique period Knossos continued to be a vibrant and influential city as it had been in the Roman period.



Hide All

2 Eusebius, Historia Ecclesiastica. Discussed by Frend, 186, Sanders, 45 and Paton, S., ‘Roman Knossos and the Colonia Julia Nobilis Cnossus’, in Evely, D., Hughes-Brock, H., and Momigliano, N. (eds), Knossos: A Labyrinth of History. Papers in Honour of Sinclair Hood (London, 1994), 141–53, 148.

3 Sanders (45) claims that there may be evidence of Christian graves from this point, however in a footnote (n. 80) he states that it is unlikely that the graves are Christian. From the original report (Frend, 210–11) it seems that it is not possible to identify the graves securely as Christian. Material evidence of a Christian community in the 2nd c. A D is therefore weak.

4 KMF, Knossos Medical Faculty, refers to the code name of the excavation which revealed the Basilica and a complex diachronic cemetery.

5 For the purposes of the discussion presented, ‘late antique’ refers to the period between the 5th and 9th cc. AD. See section on terminology for further explanation of the use of the term.

6 Hood, S. and Smyth, D., Archaeological Survey of the Knossos Area (BSA Supp. 14; London, 1981), 26.

7 Sanders, 152.

8 Sweetman, R. J., ‘The changing nature of Knossos: Roman to Late Antique—some problems’, in Cadogan, G., Hatzaki, E., and Vasilakis, A. (eds), Knossos: Palace, City, State. Proceedings of the Conference in Herakleion organised by the British School at Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities of Herakleion, in November 2000, for the Centenary of Sir Arthur Evans's Excavations at Knossos (BSA Studies, 12; London, 2004), 481–8.

9 Ead., The Roman mosaics of the Knossos valley’, BSA 98 (2003), 517–47.

10 The KMF and Sanatorium Basilicas have been excavated and published, the latter with an attempt (Frend, 186) to shed light on Knossos through a comparison with other late antique cities. Such studies may present guidelines, but cities are not homogenous and do not necessarily follow a strict pattern of establishment and development.

11 Sweetman, 390.

12 Such a term is not without its problems; for example, as yet there is no universally accepted fixed chronological range for late antique.

13 Tsougarakis, 21.

14 Such inscriptions provide indirect evidence for the nature of the administration on the island: a Gortyn example, which records the repair of a wall during the time of a proconsul rather than a consularis, reveals change of governing structure (Sanders, 6).

17 Tsougarakis, 155.

18 Ibid., 198.

19 Sanders, 9.

20 Di Vita, ‘Scavi’; Di Vita, A. and Martin, A., Gortina II. Pretorio, Il materiale degli scavi Colini 1970–77 (Padua, 1997); Di Vita, A., Gortina V. Lo scavo del Pretorio (1989–95) (Padua, 2000); Themelis; Kalpaxis, T., Schnapp, A. and Viviers, D., ‘Itanos’, BCH 119 (1995), 713–36.

22 Sanders, 35. In the Roman period, Crete's southern coastal towns lay on the main corn trade route between Egypt and Italy. These towns benefited enormously and there is no evidence of any change in their situation until the 7th c.

24 Saltpans have been located atSouda, Spinalonga, and Elounda. Tsougarakis, 276.

25 Tsougarakis, 272.

26 Although a glass factory existed at Tarrah, G. Weinburg, ‘Excavations at Tarrah, 1959’ (Hesp. 29 (1960), 90117 at 100), it was probably unable to meet the needs of the whole island.

27 Tsougarakis, 277; Di Vita, ‘Scavi’; Di Vita and Martin (n. 20); Di Vita (n. 20); Themelis; Hayes.

28 Sanders, 30.

29 Sweetman, 343.

30 Indicating that the island was not greatly troubled by raids.

31 See Volanakis, I., “Τα παλαιοχριστιανιϰά μνημεία της Κρήτης”, Kr. Chron. 27 (1987), 235–61, for a reasonably up-to-date list of the basilicas.

32 Sweetman, work in progress.

33 The arguments are well summarized in Bowden, W. A. R., ‘Town and Country in Late-Antique Epirus Vetus’ (unpublished Ph.D. thesis, University of East Anglia, 2000), 150–3.

34 Sweetman, 22.

35 By the mid-7th c. so much of the Empire's funds had been diverted to defence that in order to finance a counterattack on the Persians the Emperor Heraclius had to borrow gold and silver from the church (Treadgold, 90).

36 Kalokyres, K. A., “Ανασϰαφὴ βυζαντινῆς βασιλιϰῆς ἐνΒυζαρίη̣ Κρήτης”, Πραϰὰ τῆς Ἀρχαιολογιϰῆς Ἑταιρείας,111 (1956), 250–61.

38 Sanders, 30–1.

39 For a reasonably up-to-date list of survey projects see Moody, J., Nixon, L., Price, S. and Rackham, O., ‘Surveying poleis and larger sites in Sphakia’, in Cavanagh, W. G. and Curtis, M. (eds), Post-Minoan Crete, Proceedings of the First Colloquium on Post-Minoan Crete 10–11 November 1995 (London, 1998), 8795 at 88.

40 Decreasing levels of security directly on the island and indirectly as a result of decline in the Empire.

41 Sweetman, 399–401.

42 Ibid., 435.

43 A mêlée of different styles and influences can be found in church architecture of late antique Greece but the standard style was the triple-aisled, simple apse (Krautheimer 1986, 119–21).

44 Bowden (n. 33), 145–7.

45 Sweetman, 439–41.

46 Tsougarakis, 200.

47 The inscription is helpful for dating the Basilica because Bishop Euphratas is listed as having attended the Council of Ephesus in AD 451 (Themelis, 273).

48 Thanks to the anonymous referee who brought this o my attention.

49 Ibid. (n. 46).

50 Sweetman, 380–1.

51 Mathews, T. F., The Early Churches of Constantinople: Architecture and Liturgy (London, 1971), 4.

52 Mango, C., Byzantine Architecture (New York, 1974), 73.

53 They were Bishops Zenobias, Gennadios, and Anastasios respectively (Tsougarakis, 393).

54 These records are currently housed in the BSA Archive. I should like to thank the Archivist, Amalia Kakissis, for facilitating the study of the excavation records.

55 M Sweetman (n. 9).

56 Megaw, 323.

57 Frend, 187.

58 The other is KS 202; see below.

59 Sanders, 84, 317.

60 Since this tomb has not been published in detail a brief description is included here.

61 The publication of the Christian tombs associated with the KMF Basilica is forthcoming.

62 Coldstream, N., Knossos: the Sanctuary of Demeter (BSA Supp. 8; London, 1973).

63 Sackett, L. H. (ed.), Knossos: From Greek City to Roman Colony. Excavations at the Unexplored Mansion (BSA Supp. 12; London, 1992).

64 These are briefly recorded in Platon, N., “Χϱονιχά”, Kr. Chron. 33 (1957), 336.

65 Coldstream (n. 62), 16–17.

66 Detailed study of the material associated with the well which is stored in the Stratigraphic Museum in Knossos, will be undertaken in the near future.

67 KN Logbook 106.

68 Frend, 186.

69 R. J. Sweetman, ‘The Romanization of Knossos: a study of mosaics and other material culture’, Proceedings of the 9th International Congress of Cretan Studies (forthcoming)

70 Sweetman (n. 8).

71 A more detailed account of the site is forthcoming and will complement the publication of the ceramic material by Hayes (n. 1) and Paton's publication of the so-called Martyr's tomb (Paton, S., ‘A Roman Corinthian building at Knossos’, BSA 86 (1991), 297318). This will include data from the associated ossuaries, Houses A and B, and the Courtyard Well, formerly known as the ‘Turko-Byz Pit’. Evidence from the excavations of the peribolos wall, inner colonnade, and portico are also treated in more detail in the forthcoming publication but a brief account is included here where relevant to the subject matter.

72 For example where the inspiration for the particular architectural style may have come from.

73 Megaw, 321–9.

74 Hayes, fig. 1.

75 The notebooks are housed in the BSA Archives; I should like to thank Amalia Kakissis for generously facilitating access to them.

76 The Basilica at Panormou on the north coast contains the remains of a similar cistern (Curuni, S. A. and Donati, L., Creta bizantina: rilievi e note critiche su ventisei edifici di culto in relazione all'opera di Giuseppe Gerola (Rome, 1987), fig. 14. 8).

77 343 KMF 6, 43.

78 Ibid., 18, 2–3.

79 This wall therefore divides the central apse from the square of the sanctuary area.

80 Megaw, 324.

81 Ibid., 323.

82 Varalis, 203.

83 343 KMF 18,2.

84 Ibid., 18, 4.

85 Ibid., 6, 21.

86 Megaw, 323–4.

87 343 KMF 6, 55–7.

88 As in the case of the Mitropolis Basilica, Gortyn: Vita, A. Di, ‘Atti della Scuola 1990–91’, ASA di Atene, 68–9, (19901991), 405500; id., ‘Scavi’.

89 Long soleas are found in basilicas such as that of Metropolis, Gortyn (Di Vita (n. 88), fig. 78) and Ayia Trias at Gialoussa, Cyprus (Pallas, D., Les Monuments paléochrétiens de Grèce découverts de 1959 à 1973 (Rome 1973), fig. 201).

90 343 KMF site plan 4.

91 Ibid., site plan 7.

92 It is impossible to tell whether this was an original doorway giving access from the outside or a doorway leading into the north annexe.

93 Megaw, fig. 1; Hayes, fig. 1.

94 Sweetman, R. and Katsara, E., ‘The Acropolis Basilica Project, Sparta: a preliminary report for the 2000 season’, BSA 97 (2002), fig. 1.

95 Megaw, 323 and fig. 1.

96 343 KMF 6, 54.

97 Ibid., 18, 12.

98 Ibid., 6, 79.

99 Ibid., 18, 6.

100 Ibid., 18, 11.

101 Megaw, 323.

102 343 KMF 6; 16, 21, 27.

103 Ibid., 6, 33.

104 The mosaic fragments are stored in the Stratigraphic Museum, Knossos, studied while the author was Acting Curator.

105 Given the colour, these could also be reconstructed as leaves.

106 343 KMF 18, 6.

107 Krautheimer pl. 146.

108 This material is the focus of a forthcoming publication by the author.

109 Hayes, 431–54.

110 343 KMF 6, 52 context 153.

111 Ibid.

112 Megaw, 323.

113 343 KMF 6, 53–5.

114 Ibid., 17,48–51.

115 i.e. the floor of the courtyard that surrounds the Basilica and is bordered by the peribolos wall.

116 343 KMF 28, 6 and 23, 14.

117 Ibid., 7, 29.

118 Ibid., 7,20 and 21.

119 Ibid., 7, 37.

120 Ibid., 21, 15

121 Ibid., 13, 11.

122 Ibid., 21, 14

123 Preusse, G. and Chaniotis, A., ‘Neue Fragmente des Preisedikts von Diokletian und weitere lateinische Inschriften aus Krete’, ZPE 80 (1990), 189202, 201. I would like to thank S. Paton for referring me to this.

124 Pers. comm. S. Paton.

125 343 KMF 20, 12.

126 Megaw, 325.

127 Full publication by the author is forthcoming.

128 Paton (n. 71).

129 Megaw, 323.

130 Tomb 158 for example had a 5th-c. coin in the fill.

131 Megaw, 327.

132 Ibid., 326.

133 Hayes, 433.

134 I should like to thank K. Hartnup for pointing this out.

135 I should like to thank S. Paton for making this valuable comment and pointing out the connection between Hayes's terminus date for the cemetery and the plague outbreak in Egypt.

136 Megaw, 326.

137 Hayes, 443–53.

138 Ibid., fig. 11.

139 There are numerous examples from across the Empire. For Rome see Grabar, A., Martyrium: Recherches sur le culte des reliques et l'art chrétien antique (Geneva, 1972), figs. 49–50; for Mitropolis in Crete see Sanders, 112.

140 For example from Perustica in Bulgaria (Grabar (n. 139), 75).

141 Krautheimer, 117.

142 No evidence for mosaic was found in the western section of the basilica.

143 Mango (n. 52), 73.

144 Tsougarakis, 113.

145 Krautheimer, fig. 159.

146 Varalis and Megaw.

147 Krautheimer, 196.

148 Grabar (n. 139), pl. 27.

149 Piccirillo, M., ‘Mount Nebo’, Studium Biblicum Franciscanum, Guide Books 2 (Jerusalem, n.d), 31–3.

150 Mango (n. 52), 87.

151 Varalis also presents a number of parallels for this particular form; see 203–13 for the catalogue.

152 Ibid., 68.

153 Duval, N. et al. , Basiliques chrétiennes d'Afrique du Nord I. Inventaire de l'Algérie (Collection des études augustiniennes. Série Antiquité 129; Paris, 1992), fig. 77. 7.

154 Shelley, J. M., ‘The Christian Basilica near the Cenchrean gate at Corinth’, Hesp. 12 (1943), 166–89.

155 Pallas, D., Les Monuments paléochrétiens de Grèce décovverts de 1959 à 1973 (Sussidi allo studio delle antichità cristiane; Rome, 1977), 88.

156 Megaw, 328.

157 Ibid.

158 Mathews (n. 51), 172.

159 Varalis, 218.

160 Krautheimer, 117.

161 Mathews (n. 51), 176.

162 Varalis, 213, 218 has made a similar suggestion, that the KMF Basilica represents a western architectural type adapted to serve local liturgical practices.

163 Megaw, 327.

164 Varalis, 199.

165 Frend, 188.

166 Ibid.

167 Frend (211) mentions the possibility that the graves might bejewish, but there is no evidence for this.

168 Frend, 203 and fig. 3.

169 Ibid., 205.

170 Ibid., 211.

171 See also Ibid., fig. 3 for plan of the burials.

172 As the Basilica is not orientated along the E-W line, strictly speaking it should be the NE wall and the SW wall. However as the aisles are usually referred to as the north and south aisle, these terms will be adhered to.

173 Eleutherna, which has a similar plan, is c. 45 × 30 m (Themelis, fig. 7).

174 Frend, 187.

175 Ibid., 198, 192.

176 Ibid., 190, 192.

177 Ibid., 196.

178 Frend (190) notes that in general the basilica was roughly constructed using a combination of large squared blocks and smaller worked blocks bonded together with a white mortar.

179 On the evidence of some fragments of green and blue glass tesserae, Frend is probably right to suggest they are from wall mosaics. He also suggests that there may have been wall paintings (Frend, 90).

180 Bowden (n. 31), 154–5 summarizes the different funding sources for church building. See also Jones, A. H. M., ‘Church finance in the fifth and sixth centuries’, Journal of Theological Studies, n.s. 20 (1969), 8494.

181 Frend, 212.

182 Sanders, 107.

183 Frend, 222.

184 Ibid., 218.

185 Themelis, fig. 7.

186 Themelis, 272–83. Frend (192) mentions two Cretan basilicas which also have high stylobates: Vizari and Sybrita. These have been dated to a later period.

187 The Eleutherna mosaic inscription discussed above mentions the name of the bishop who attended the 451 Council of Chalcedon.

188 There is no agreement regarding the date of the basilica or of the mosaics (Byz. Corp. I, 109).

189 Frend, 212.

190 Ibid.

191 Sanders, 107.

192 Frend, 198.

193 It has not been possible to locate photographs of the mosaics from the excavation archive.

194 Campbell, S. D., ‘Roman Workshops in Turkey’, AJA 83 (1979), 287–92 at 290 and Sodini, J. P., ‘Mosaïques paléochrétiennes de Grèce’, BCH 94 (1970), 699753 at 751.

195 Byz. Corp. II, pls. 126–7.

196 Duyuran, R., ‘Istanbul’, Annual of the Archaeological Museums of Istanbul, 9 (1960), 1821, pls. 1–8.

197 Hanfmann, G. M. A., Sardis, from Prehistoric to Roman Times: Results of the Archaeological Exploration of Sardis 1958–1975 (London, 1983), fig. 254.

198 Byz, Corp. I, pl. 104.

199 Assimakopoulou-Atzaka, P., “Τα παλαιοχριστιανιϰά ψηφιδωτά δάπεδα του Ανατολιϰου”, Actes du Xe Congrès international d'archéologie chrétienne (Thessalonique 28 septembre-4 octobre 1980) (Thessaloniki, 1984), 361444.

200 Byz, Corp. I, nos. 105 and 88 respectively.

201 Ibid., nos. 101 and. 87 respectively.

202 Byz. Corp. II, p1. 134.

203 For each of the designs see figs. 33–7.

204 In spite of the proposed architectural connection, the Eleutherna mosaics do not belong to any of the workshops as suggested here.

205 This pattern could also be described as tangent four pointed stars, in which the arms are filled with a chevron designs (zigzags). In the spaces left over there are lozenges, and these contain the central circle, flanked by the semicircle with the hooks either side (knot design).

206 Harrison, R. M. and Lawson, G. R. J., ‘The mosaics in front of the Vilayet Building in Istanbul’, Annual of the Archaeological Museums of Istanbul, 13–14 (1965), 216–18.

207 Byz. Corp. II, 90.

208 Ibid., 100.

209 Byz. Corp. I, 119 and 111. It is notoriously difficult to date Early Christian basilicas in Crete. More often than not a date of late 5th or early 6th c. is offered and rarely called into question. Consequently such a date is not reliable.

210 Byz. Corp. I, 130.

211 Hanfmann (n. 197), 254.

212 Byz. Corp. I, 112 and 122 respectively.

213 Frend, 212.

214 Hood and Smyth (n. 6), 26; Sanders, 152; Hayes, 433.

215 Sweetman (n. 8) and Forster, G., ‘Shadow of a city? A review of Late Roman Knossos’, in Cadogan, G., Hatzaki, E. and Vasilakis, A. (eds), Knossos:Palace, City, State. Proceedings of the Conference in Herakleion organised by the British School at Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities of Herakleion, in November 2000, for the Centenary of Sir Arthur Evans's Excavations at Knossos (BSA Studies, 12; London, 2004), 489–92.

216 Frend, 186.

217 Sweetman (n. 8).

218 It is difficult to make proposals concerning the layout of the town because the Ayia Sofia Basilica has not been investigated in any detail or dated.

219 Indeed this has been suggested by Hood at Hood and Smyth (n. 6), 26, but for an earlier period.

220 As is the case for other Cretan towns, such as Suia and Chersonisos.

221 As is likely to have been the case for the prosperity of other inland towns such as Gortyn and Eleutherna.

222 It is clear that Knossos was the seat of a bishop.

1 I should like to thank the following archaeologists who have generously made their published and unpublished material available to me: J. Carington Smith, H. Catling, W. Frend, S. Hood, and S. Paton. Dr Catling has been particularly supportive, providing me with both advice and photographs. S. Paton has provided helpful suggestions on the text and kindly allowed me to reproduce her architectural drawing in FIG. 6. Many of the thoughts expressed here are the results of fruitful conversations with M. Boyd, G. Sanders, D. Turner and Y. Varalis. Thanks are also due to the anonymous referees who made valuable suggestions on the text and content. I should particularly like to thank K. Hartnup for the time she has spent reading and commenting on the text.

Special abbreviations:

Byz. Corp. I = S. Pelekanides,Σύνταγμα των παλαιοχριστιανιϰών ψηφιδωτών δαπέδων της Ελλάδος Ι: Νησιωτιϰή Ελλάς (Κέντρον Βυζαντινών Ερευνών Θεσσαλονίϰης)(Thessaloniki, 1988).

Byz. Corp. II = S. Pelekanides and P. Atzaka,Σύνταγμα των παλαιοχριστιανιϰών ψηφιδωτών δαπέδων της Ελλάδος ΙΙ: Πελοπόννησος —Στερεά Ελλάδα (Κέντρον Βυςαντινών Ερευνων Θεσσαλονίϰης)(Thessaloniki, 1987).

Di Vita, ‘Scavi’ = A. Di Vita, ‘I recenti scavi della S.A.I.A. a Gortina: un contribute alia conoscenza di Creta tardoantico e protobizantina’, XXXVIII Corso di cultura sull'arte ravennate e bizantina: seminario internaionale sul tema ‘La Grecia insulare tra tardoantico e medioevo’, Ravenna 15–20 marzo 1991 (Ravenna, 1991), 169–83.

Frend=W. H. C. Frend and D. E.Johnston, ‘The Byzantine basilica church at Knossos’, BSA 57 (1962), 168–238.

Hayes = J. W. Hayes, ‘Early Christian pottery from Knossos: the 1978–81 finds from the Knossos Medical Faculty site’, BSA 96 (2001), 431–54.

KN Logbook = Logbook of chance archaeological finds and rescue excavations from the Knossos valley since the Second World War. Now housed in the Archives of the British School at Athens.

Krautheimer = R. Krautheimer, Early Christian and Byzantine Architecture: revised edition by R. Krautheimer and S. Curcic (Yale, 1986).

KS = Site number as referenced in S. Hood and D. Smyth, Archaeological Survey of the Knossos Area (BSA Supp. 14; London, 1981).

Megaw = A. H. S. Megaw, ‘A cemetery church with trefoil sanctuary in Crete’, Actes du Xe Congrés international d'archéologie chrétienne (Thessalonique 28 septembre-4 octobre 1980) (Thessaloniki, 1984), 321–9.

R (number) = Reference to geometric decoration in C. Balmelle et al., Le Décor géometrique de la mosaïque romaine. Répertoire graphique et descriptif ties compositions linéaires et isotropes (Paris, 1985).

Sanders = I. F. Sanders, Roman Crete (Warminster, 1982).

Sweetman = R. J. Sweetman, ‘The Roman and Early Christian Mosaics of Crete’ (unpublished Ph.D. thesis; Nottingham, 1999).

Themelis = P. Themelis, “Βασιλιϰή ϰαι προσϰτίσματα, Ελεύθερνα, Νομός PΕθύμνης”,Κρητιϰά Εστιά, 5 (1994–6), 272–83.

Treadgold = W. Treadgold, A Concise History of Byzantium (Basingstoke, 2001).

Tsougarakis = D. Tsougarakis, Byzantine Crete: From the 5th Century to the Venetian Conquest (Historical Monographs, 4; Athens, 1988).

Varalis = Y. Varalis, ‘Deux églises à chœur triconque de l'Illyricum oriental: observations sur leur type architectural’, BCH 123 (1999), 195–225.

343 KMF = Archive reference to the Knossos Medical Faculty excavation notebooks.

Related content

Powered by UNSILO

Late Antique Knossos. Understanding the city: evidence of mosaics and religious architecture1

  • Rebecca J. Sweetman (a1)


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed.