1Karthaia1 was one of the four ancient city-states on the Cycladic island of Keos. It is situated at the southeast coast of the island on an imposing promontory overlooking the sea to the south, between two streams that frame it on the east and the west. The site has been extensively, yet intermittently, excavated since 1902 (fig. 1).2 The terracotta figurine under study came to light in 2005 in the course of restoration works on the pronaos paving of the Doric peripteros temple of “Athena” (fig. 2) that lies on the northern plateau.3

Fig. 1. – Plan of Karthaia and the temple of “Athena”.

Fig. 2. – Figurine in situ at the time of finding.

1. Type

2The figurine, in the type of a draped kouros, was shaped from a bipartite mould and fabricated of well-fired, fine, light-brown clay. It was apparently imported, as the local fabric is generally semi-fine, reddish, and micaceous.4 It is 17 cm tall, 6 cm wide at the shoulders and 5 cm wide on the side (fig. 3). The legs from the shank down and the rectangular base are missing; if complete it would have measured approximately 20.5 cm in height. It is quite corpulent, with sturdy shoulders, a wide chest, fleshy arms, a slim waist, and staunch legs and buttocks. The face is round with wide forehead and heavy eyebrows; it has almond-shaped eyes and splaying nose, a small mouth, and thin, closed lips. The plump cheeks and the thick neck form a hefty jowl. The coiffure is quite affected, resembling a wig. Long locks of hair grow from the middle of the forehead, rising to either side of the scalp, falling behind the ears to cover the shoulders and the back in long strands that are fashioned into thirteen rows of horizontally set astragals. The figurine wears a chiton and himation stretching diagonally over the left shoulder to beneath the right arm. The himation falls in the front in dense, zigzag folds, and was coloured a vivid red, traces of which still survive.

Fig. 3. – The terracotta kouros from Karthaia.

3The draped kouros figurine from Keos belongs to a well-established series of a type widely acclaimed in the mid-6^th century BC and encountered in over twenty four Archaic Greek sites, in sanctuaries and cemeteries alike.5 The majority of the known material comes from the island of Thasos in the northern Aegean, where S. Huysecom distinguished seven successive generations and three variations on the type (version A: hair drawn behind the ears to the back; version B: two locks of hair falling on each side on the chest; version C: left hand holds a lyre). With reference to her catalogue we may ascribe the figurine from Keos to generation Kv IIIA.6 Thus, it falls in the same group with the terracotta kouros 5280 from central Macedonia (fig. 4),7 and the fragmentary figurines from Neapolis,8 and Delos,9 that also belong to the variation A, which lacks the characteristic tresses on either side of the chest.

Fig. 4. – Terracotta kouros Archaeological Museum of Thessaloniki inv. no. 5280, from Central Macedonia.

2. Date

4The date of similar figurines is estimated either by their archaeological context (which is safe, but not always available),10 or on comparison with sufficiently well dated similar material in clay,11 or comparable large-scale sculpture in stone.12 The structure of the material from Thasos in seven successive generations is internally coherent but offers little in terms of absolute chronology.13 In brief, the type covers the second half of the 6^th century BC, evolving piecemeal over time.14 Fortunately, the figurine from Keos was found in a well-dated context. It was retrieved from a sealed stratum, at the edges between the pronaos floor and the stone foundation of the stylobate of the temple of “Athena”. This layer was used as a packed fill of earth, chips of gneiss and poros from the dressing of the temple’s building blocks, used for the laying of the lime mortar pavement of the pronaos. As it is widely agreed that the temple dates from the late 6^th century BC15 a terminus ante quem for the fill and the statuette, coinciding with the earliest phases of construction in the mid-fourth quarter of the century is acceptable.

5Moreover, the statuette was found in context with a number of Corinthian miniature kotyle (fig. 5). Miniature kotylai are notoriously difficult to date,16 but in general they follow the evolution of the full-sized Corinthian kotyle from short, wide, and convex to higher, narrower and straighter shapes.17 The profile and decoration of the kotylai we found is very close to kotyle no. 11518 from a burial in Corinth, which dates at, or little after, the middle of the 6^th century. We may therefore establish a date span for our figurine between 550 and 515 BC.

Fig. 5. – Miniature kotylai in context with the figurine.

6However, a chronology soon after the middle of the century seems more plausible and is corroborated among other things, by certain observations. For one, many later figurines are shaped with a lowered, rather than an upright, head (chiefly generations Kv IVB and later, dating from the last quarter of the 6^th century BC).19 This is often attested for generations Kv VB and later from Thasos,20 for figurine 13343 from the necropolis at Makri Langoni on Rodos21, for figurine T32 from Paros22 and AK4412 from Despotiko (Antiparos) (fig. 6)23. Expectedly, a parallel development is recorded with the figurine-shaped perfume-bottles, as these and the plain figurines were commonly fabricated in the same workshops: two draped kouros-vessels from the concluding decades of the 6^th century BC, one in the Metropolitan Museum of New York24 and another from Samos,25 have lowered heads, as opposed to some earlier bottle-figurines from Gela,26 Selinus,27 or the draped kouros bottle-figurine on exhibition at the Izmir Archaeological Museum. By analogy, the straight head of the Keos figurine is an indicator of earliness.

Fig. 6. – Terracotta kouros from Despotiko.

7Another late trait, typically encountered on statuettes, from generation Kv V onwards, is the “lean unsmiling face,28 so different from the usual late 6^th century Rhodian face”.29 The terracotta figurines from Sindos, Camiros, and Samos30 all belonging to generation Kv VA, are so alike in appearance that may well come from a single mould,31 which introduced a slimmer and somewhat austere statuette type in the later years. The origin of this slim type is possibly Rhodian or Samian but this is still unconfirmed.32 Its circulation indicates that towards the end of the 6^th century BC the Ionic ideal of a “full body and face”,33 characteristic of the figurine from Keos as well as of numerous early figurines having full faces, almond-shaped slanting eyes and splaying noses (particularly the samian ones), from the Heraion of Samos34 and Rhodes,35 was gradually loosing steam.

3. Origin and Distribution

8S. Huysecom36 recognized four centres with considerable distribution of draped-kouros type figurines: the Ionian coast and the islands, the island of Rhodes, Magna Grecia, and Northern Greece. The Cyclades have yielded less than a dozen out of the roughly hundred known figurines. Recently a terracotta fragment from the sanctuary of Gyroulas at Sangri on Naxos was identified as the shoulder and part of the hair of a draped-kouros figurine.37 Delos, which is short of useable clay deposits, seems to have imported quite a few. Surprisingly, the type is very thinly represented among the roughly two-hundred terracottas that were imported or locally produced at the Delion at Paros.38 The source of the three almost identical kouroi that were found at Paros and Despotiko (Antiparos)39 is most likely Rhodes, from where the two neighbouring sanctuaries presumably imported their figurines.40 The earliest confirmed Cycladic specimen (generation IIB) comes from Thera41 and is also probably Rhodian.42 The scarcity and heterogeneity of the excavated material from the Cyclades weakens the possibility of a local centre of production. The type of the draped-kouros terracotta figurines seems to have penetrated the Cyclades in small numbers through some nearby eastern Greek centre, which must have also been the source of the figurine we found at Karthaia.

9Samos and Miletus are both likely candidates for the formation of the type.43 Plentiful unpublished material from these sites provides a solid framework for future reference and comparison44. The figurine from Keos has the round, fleshy face, the absentminded expression, and the heavy posture of the body that is typical of the Milesian coroplathy. A kouros figurine now in the Ny Carlsberg Glyptotek (inv. no. 948)45 that is almost identical to the Kean terracotta, exhibits “Milesian” characteristics too (fig. 7). While at this stage the exact provenance of either statuette must remain inconclusive, we note that recent research has attributed a number of terracottas and vases, formerly considered to be ‘Rhodian’ or vaguely ‘Eastern-Greek’46, to Miletus.47

Fig. 7. – Terracotta kouros Ny Carlsberg Glyptotek, inv. no. 948.

10However, if we can afford a preliminary assessment, this would place our figurine closer to the Samian series of terracottas, which appear notably similar in fabric and style. To the samian figurines already noted for their resemblance to the one from Karthaia,48 we may now add significant material exhibited in the Pythagoreion Museum on Samos. Particularly the fragments nos. 7569, T460 and T653 share a close likeness to the figurine from Keos and could indicate its samian roots.

11The discovery of imported votives on the site of Karthaia, such as the kotylai49 and the terracotta kouros among others,50 is of significant value to research, especially as these were retrieved from a stratum earlier than the temple of “Athena.” Only a full-scale excavation in the future may shed some light on the earlier phases of the temple plateau, which appears to have housed a cult of some importance.