KNOSSOS URBAN LANDSCAPE PROJECT - 2011

KULP Survey, Knossos. T. Whitelaw (BSA/University College London) and A. Vasilakis (for the ΚΓ’ ΕΠΚΑ) report on further study in 2011.

Units collected in 2007-2008 (outside the city, covering most of the eastern half of the survey area) were reviewed and re-dated, doubling the number of sherds assigned dates finer than the gross categories of Prehistoric, Hellenic, Roman and Post-Roman. This particularly helped to define the date and extent of prehistoric occupation on the summit of Ailias, on the east of the valley, and the outlying scatters on the slopes of southern Ailias and in the north of the valley. The latter large but low density scatters of prehistoric sherds are most likely to represent cemeteries, though confirmation awaits detailed study of the types of vessels represented.

Fig. 1A plots all diagnostic Prepalatial sherds (primarily EM III-MM IA) and Fig. 1B MM sherds (as well as less diagnostic black-painted fragments, which will be dominantly of MM date). This confirms interim assessments that no major occupation extended to the east bank of the Kairatos river, and also indicates that the suburb on the central summit of Ailias, first noted by Hood, Smyth and Carington Smith in the mid-1970s, extends back into the MM period. A light scatter of EM III-MM IA sherds (Fig. 1A) may indicate some occupation before the end of the Prepalatial period.

A light scatter of MM sherds on the west slope of southern Ailias presages the denser though still light scatter of LM material, almost certainly representing material eroding from chamber tombs on this slope, parallel to those excavated further north at the Mavrospelio and Ailias cemeteries, and lower down at the KSP tombs excavated by Hood (KS2.278). An extensive but light scatter of MM sherds on both sides of the Kairatos in the north valley, east of the Kephala ridge, may represent other burials flanking the river.

Fig. 1C plots all diagnostic LM sherds (as well as less diagnostic red-painted sherds, the latter dominantly of LM date). Most patterns identified for MM are better defined by the much larger quantity of LM diagnostics, but several new patterns are unlikely to be the result simply of the larger sample. The first is a distinct concentration on the very exposed northwest point of Ailias which has no evidence for earlier occupation (see also Figs. 1A-B). In the north valley, the scatter flanking the Kairatos continues in this period, but LM material is now also found widely on the west slope of the Kephala ridge, the east slope of the ridge underlying modern Ag. Ioannis, and flanking the North Cemetery on the east and north in the limited fields available for survey. Whether a very low density scatter across the north point of the Isopata ridge is sufficiently dense to document the extent of the Isopata cemetery awaits re-study of the collection units on the centre and west of the ridge. Also better defined are high-density patches of sherds slightly separated from the main city distribution in the northeast, immediately south of the Zapher Papoura cemetery and on the west slope of Lower Gypsadhes. Confirmation as cemeteries awaits full study of the vessel forms represented.

Since understanding of the LM II-III periods is hampered by the low quantities of highly diagnostic fine wares recognised, interpretation of Fig. 1D is problematic. There is no evidence for continued occupation on the summit of Ailias, but equally, recognised diagnostics mark the locations of the numerous excavated LM II-III cemeteries surrounding the site.

For the Early Iron Age (Fig. 2A), the substantial nucleated town is now well documented. The cemetery tested by Evans, Hogarth, Payne and Hood on the west slope of the Kephala ridge stands out from one dense deposit. Around the south and west fringes of the town, patches of sherds on the northwest and west sides of the Acropolis may represent clusters of tombs, as may the fairly extensive distributions on the northeast and west slopes of Lower Gypsadhes. Detailed re-assessment of the 2007-2008 collections provides no support for the theory of dispersed hamlets during this period.

For the Hellenic period (Fig. 2B), some correspondence with LM distributions in the outlying low-density scatters (e.g. Ailias, north valley) raises the possibility of mis-classification, particularly among the finest buff wares. A real correspondence in distributions in the different periods is, however, suggested by the distributions of more firmly attributed diagnostics, and some spatial shifts in concentrations. The same potentially applies to the fine local Roman wares, which when poorly preserved can resemble LM material (Fig. 2C). Late Antique diagnostics (Fig. 2D) reveal that the declining city was still highly nucleated. The concentration of Post-Roman material in the northwest of the study area, in the areas most easily accessible from medieval and early modern Candia, remains very clear.