The Stratigraphy of Akrotiri

The final destruction apparently began with earth tremors, followed by earthquakes and then a period of calm, in which clearing of the ruins began. This was terminated by a series of pumice falls which buried everything in the great cataclysm.

One of the major points frequently raised in connection with the settlement at Akrotiri is its chronology, both its beginning and its end. Despite the sporadic references which have occasionally been made to earlier phases (Marinatos 1968, 11; 1969, 19, 27 - 28) scholars still continue having doubts about the founding of Akrotiri before the LM Ia period (Barber 1974, 48, 52; Immerwahr 1977, 184 n. 51). On the other hand no general agreement has been reached about the final collapse of the island (Pomerance 1970). It is true that the material which is dated earlier than the LM Ia period in Akrotiri is still very limited, principally because of the nature of the excavation. Even less of this material has been published, justifying the above lack of information. Thus the aim of this paper is to remedy this situation by outlining the stratigraphic sequence, as it has been revealed during the last decade's excavations. Despite its sporadic character, the evidence of the pre-Minoan occupation of the site is so strong that it cannot be disputed.

EARLY BRONZE AGE    

Marinatos has occasionally attributed to the Neolithic period some of the finds (1968, 11). Of these finds, however, the stone implements were not associated with any stratigraphic horizon, and, furthermore, there is no evidence that they were not in use by the Bronze Age people of Akrotiri. On the other hand, the sherds which he considered to be early cannot be dated before the Keros-Syros culture of the Cycladic Early Bronze Age.

The earliest certain evidence of habitation at Akrotiri goes back to the third millennium B.C. Although the excavation so far has been restricted to the more recent phase of the site, that is the Late Minoan Ia, which was buried under the volcanic material, because of the need to support the extensive roof by strong pillars, deep pits were often dug so that these pillars were founded on the bedrock (Marinatos 1969, 24). Quite often from such pits pottery has appeared which belongs to the Keros-Syros and Phylakopi I cultures of the Early Cycladic period (Marinatos 1969, 28). Although the pottery consists of fragments one can easily recognize shapes of the Syros, Kastri, Amorgos and Phylakopi I groups (Doumas 1977, 20 - 25). Besides pit 6 ⁽¹⁾ Early Cycladic sherds have also been found in the area round pillars 4 and 5, in pits 11, 27 and 72, on the floor of the basement of rooms Δ 3 and B 2, in the areas of Γ8 and the" Kitchen"  (Marinatos 1971, 15) as well as in the layer below the floor level of Δ 16 (for the distribution see fig. 1) ⁽²⁾.

Quite often the Early Cycladic pottery is associated with structures cut in the rock (e.g. pit 6, room Δ 3), which recall the rock-cut chamber tombs of the Phylakopi I culture (Doumas 1977, 49). Unfortunately none of these structures has been thoroughly investigated as yet, thus making their interpretation difficult. Despite its place of discovery on the north wall of room B 5, the marble figurine of the Folded-arm type constitutes further evidence of the Early Cycladic in Akrotiri (Marinatos 1971, 23 pl. 38b).

Although fragmentary, the above evidence is enough to prove that in the area of the excavation a quite sizable Early Cycladic settlement had existed. This means that the coastal site of Akrotiri has been inhabited since at least the middle of the third millennium B.C.

MIDDLE BRONZE AGE

Equally widespread is the evidence of occupation during the Middle Bronze Age. For the reasons explained above this evidence is also sporadic and has been confirmed only in places where the requirements of the work led us deep below the Late Bronze Age horizon. The Middle Bronze Age pottery in Akrotiri is rather well preserved and entire vases have already been found. Various types of jars, bowls, beakers, small pithoid vases, a hydria are among the vases of that period. When decorated, they bear matt-painted geometric patterns, either curvilinear or rectilinear. Often this pottery is associated with architectural remains (Marinatos 1971, 22).

The distribution of the Middle Bronze Age finds (s. Fig. 1) includes places like the "Sacrificial Fire" (Marinatos 1969, 19ff.), rooms Δ 2, Δ 16 and Δ 17, the pillar pits 20, 21, 24 and 35 as well as the areas around rooms Δ 10 -13 and Arvaniti 3 (Marinatos 1968, 11 plan III). Very few fragments of imported fine Kamares ware are certainly associated with the Middle Cycladic horizon (Marinatos 1972, 22 - 23 pl. 67b-d). Some other pottery may have a Middle Helladic provenance (Marinatos 1971, 31 pl. 62). All this Middle Bronze Age evidence and its distribution throughout the site suggests that the life of the settlement did not end with the Early Bronze Age (Marinatos 1971, 44 - 45). On the contrary it continued to thrive with more contacts with the outside world.

LATE BRONZE AGE

Immediately below the floor level of the Late Minoan Ia city which was buried beneath the volcanic ash lies a quite thick deposit of rubble. Among the stones and earth from demolished houses large quantities of domestic Late Minoan Ia pottery are found. It seems that this layer was formed by depositing and levelling down the debris of the city after a general destruction which must have happened not long before its final smothering (Marinatos 1970, 8 - 9; 1971, 9, 27, 44 - 45). East of Building Δ, just in front of rooms Δ 17, Δ 18 and Δ 19 this debris had been arranged in platforms which formed the street level for the East entrance to Building Δ through room Δ 19 (Marinatos 1972, pl. 14b and 15 a - b). To that disaster one could possibly ascribe the destruction of the wing west of rooms Δ 11 and Δ 13, which was later partly replaced by room Δ 16.

The walls of room Γ9 were razed to the ground before the city was buried under the pumice. However its debris did not lie next to it; the area was covered with pumice (Marinatos 1969, 51 pl. 44a). This means that the debris had been removed before the eruption. In fact, this debris was arranged in two piles, one of stones and one of earth, in Square 1 between the rooms B1 and Δ15. A third pile of small stones and earth extends all along the part of Telchines' Road, west of rooms Δ15 and Δ16 (Marinatos 1969, 28 - 31, pl 24 - 26). A dry-stone retaining wall was built to maintain the debris in front of room Δ 16 (Marinatos 1972 pl. 24a). By piling the rubble at the sides of Telchines' Road a free passage was arranged through the settlement from North to South, before the eruption (Marinatos 1969, 30). Another activity which took place prior to the eruption was apparently the demolition of dangerous walls as well as the repair of others (Doumas 1974, 111; 1974a, 368). All this work would have been unnecessary if the buildings had not suffered considerable damage. And this damage occurred shortly before the eruption, and was probably interrupted by the eruption (Doumas 1974, 111 - 12).

The earliest stratigraphic evidence of the eruption is a very thin layer of fine pumice about 2 - 3 cm thick. This layer has been observed everywhere in the site without any violation, which suggests that after its formation no human activity took place there. Marinatos (1970, 9) describes this layer as "hardened like some sort of plaster". In fact it has an encrustation on top, and due to oxidation it has a yellowish appearance. The oxidation suggests that this layer of pumice was exposed to the air for some time before it was covered by the next pumice fall (Money 1973, 52). Possibly the kind of encrustation of its surface may be due to rainfall and water accumulation. In this case the fine pumice, floating on top of this water, was gently deposited on the ground as the water was absorbed.

Immediately upon the layer of pellety pumice another stratum, up to 50 cm thick, was deposited, its individual pumice stones being coarser (about 3 - 5 cm in diameter). Much bigger pieces formed the next layer of pumice, which filled up the open areas between the buildings so that its surface was levelled; its thickness, therefore, sometimes reaches two metres (Marinatos 1970, 9). There is no clear distinction between the two strata of coarse pumice except for the size of the individual pieces. It is interesting, however, that the upper layer often contains stones fallen from nearby walls or even entire parts of collapsed walls (Marinatos 1969 pl. 13a; 1970 pl. 9b; 1971 pl. 21a).

Neither signs of erosion nor any soil formation distinguish the upper layer of pumice from that of the dust which is the volcanic material deposited last. This may indicate that no considerable time span intervened between the deposition of the two layers. But it must be of major importance that enormous boulders often found in the site as well as in the fields are associated with this ash layer. The progress of the excavation has proved a theory proposed by Marinatos (personal communication) that these boulders were ejected from the volcano and literally bombarded the site. In fact three such boulders have hit the imposing building of Xeste 4 : the North wall of the building has been damaged at its west extremity in the third storey; much more serious is the damage to the north-east and south-east corners. In all three cases the boulders responsible for the damage have been found in situ indicating also the northerly direction from which they came. This discovery was made by Marinatos during his 1974 digging season. His subsequent sudden death did not let him enjoy the announcement of the confirmation of his theory. It is hoped that specialists will be able to estimate the force with which the boulders were ejected and flung so far from the volcano.

From this outline of the Akrotiri stratigraphy the following points can be sumarized:

i.     An Early Cycladic settlement existed there since at least the middle of the third millennium B.C. (Keros - Syros and Phylakopi I cultures).

ii.    No break is evidenced between the Early and Middle Cycladic period. The Cycladic character of the settlement continued to prevail though a few imports from Middle Minoan Crete and from Middle Helladic Greek mainland are present.

iii.   Without break the settlement entered the Late Bronze Age. A rich Late Minoan Ia culture is attested mainly by the ceramic evidence and the wall paintings. During this period the site suffered two serious blows:

iv.   The first destruction of the LM Ia city probably due to earthquakes was followed by a general levelling of the ruins which could not be repaired.

v.    In the layer of this debris the foundations of the new city were laid, though in a few cases earlier walls have also been used.

vi.   The second and fatal blow against the city seems to have occurred in the following stages:

a.       Earthquakes which damaged the buildings. Possibly the earthquakes started with tremors since the inhabitants managed to evacuate before the destruction. No victims have been found so far from the excavation.

b.       Clearing of the ruins. This implies that a period of calm elapsed after the destruction, so that people felt secure to come back and work in their ruined city.

c.       First fall of pumice probably followed or accompanied by rain.

d.       Second fall of pumice in two successive layers, and volcanic ash accompanied by the ejection of the boulders.

e.       There is neither stratigraphic nor archaeological evidence so far about the time of the final collapse of the island from Akrotiri.

- (1).   Although the description given by Marinatos for pillar pit 7 (1969, 27) is that of pit no 6, in plan IV pillar 6 is marked in the bed of the ravine where the stratigraphy consisted of plain sand only. As he explained to me after the publication of the report, the changes in the enumeration of the pillars were made in order to avoid number thirteen. These changes were: pillar 5 became 4a, 6 became 5 and pillar 13 disappeared becoming 6. Regardless of the entries in the daybooks and the labels of the finds, similar changes have not been uncommon both in numbering the pillars and labelling buildings or rooms. Finally, as the general plan in Thera VII, 1973 shows, number 13 has been restored; unfortunately not to its original place but in the place of pillar 5 which was previously marked as 4a.  

- (2).   A more detailed presentation of the pre-Minoan material from Akrotiri is under preparation and it is hoped that it will appear soon.

---------------------------------------

-----------------------------------