Akrotiri on Thera, the Santorini Volcano and the Middle and Late Cycladic Periods in the Central Aegean Islands
- The Late Bronze Age Eruption of the Santorini Volcano
- The Sequence of Events as Established from the Stratigraphy at Akrotiri
- The Impact of the Theran Volcanic Eruption Beyond Santorini
- The Frescoes from Akrotiri
- Comments on Theran Mural Painting in Comparison to Contemporary Minoan Frescoes
Early in the Late Bronze Age, the volcano at the center of the island of Santorini (or Thera) erupted on a scale which may have had no parallel among eruptions over the past four or five millennia by volcanoes located in or near densely populated areas of the globe. The caldera (or crater) created by this eruption of the Theran volcano is said to have measured as much as 83 square kilometers in area. It presently extends down as much as 480 meters below sea level inside of the wall of cliffs which ring it and which themselves rise up as much as 300 meters above sea level.
There has been an impressive amount of debate during the last twenty-five years in particular over the nature and sequence of the cataclysmic phenomena which led up to and resulted from this enormously impressive volcanic eruption, debate in which both vulcanologists and archaeologists have played leading roles. The impact of the eruption on the cultural history both of the smaller Aegean and of the larger eastern Mediterranean worlds has also been extensively discussed, principally by scholars from the same two disciplines. The reconstruction which follows is that most widely shared as of 1986, a little less than fifty years since Spyridon Marinatos published his landmark article postulating a connection between the Theran eruption and the collapse of Minoan palatial civilization. This theory ultimately led Marinatos in the late 1960’s and early 1970’s to begin excavation at the site of Akrotiri on the southern tip of Thera, a site which has turned out to be a prehistoric Aegean version of the better known sites of Pompeii and Herculaneum buried in 79 A.D. by the eruption of Vesuvius in the Bay of Naples. Since Marinatos’ death in the mid-1970’s, the director of the program of excavation, restoration, and publication at Akrotiri has been Christos Doumas.
A major earthquake caused extensive damage to the town of Akrotiri well before the volcanic eruption buried it. Copious evidence for the purposeful demolition by wrecking crews of buildings partially destroyed in the earthquake has been found in the form of piles of rubble and earth debris heaped up in the principal roads leading through the settlement and retained behind roughly built dry-stone walls of rubble. Marinatos’ exposure of some houses which had been repaired in a rather makeshift fashion and subsequently re-occupied led him to conclude initially that the site had been permanently abandoned by most of its inhabitants as a result of the earthquake but that parts of it were inhabited before the volcanic eruption by “squatters”. According to Marinatos, these “squatters” proceeded to loot the houses of the wealthy and to stockpile whatever wealth they were able to find, although such wealth consisted of little more than agricultural produce since the former inhabitants had evidently had enough time to strip their homes of all but a few of their readily portable valuables. Doumas has correctly denied the existence of such “squatters” in view of the now plentiful evidence at Akrotiri for a systematic program of demolition, levelling of debris, and rebuilding represented throughout the site. This indicates that the settlement’s entire population undertook an extensive program of restoration following the earthquake, one which was still in progress when the volcano erupted and buried the town. Particularly striking evidence of the intentional demolition of structures weakened by the earthquake are “demolition balls”, very large ellipsoidal ground-stone implements with two grooves around their waists which were no doubt used in very much the same way as are contemporary “wrecking balls” and which have been found in some quantity on top of the levelled debris resulting from the cleaning up of the earthquake damage.
The length of time which elapsed between the earthquake which badly damaged Akrotiri and the volcanic eruption which buried it was initially considered to be a very short one by Marinatos, and Doumas as recently as 1983 suggested that it might have been a matter of only a few months. However, more detailed comparisons by Marthari of the ceramic material from the heaped up earthquake debris with that from the volcanic destruction horizon reveal significant differences which suggest that years rather than months separated the two events, perhaps as much as two or three decades. The process of rebuilding and restoration begun shortly after the earthquake was nevertheless still in progress when the volcano erupted, as the partially plastered and painted condition of the second-storey bedroom in the West House indicates. Two vessels full of dried plaster and a third containing dried paint show that this room was actually in the process of being decorated when the site was hastily abandoned, this time for good. As both Marthari and Palyvou have shown, the repairs made necessary by the earlier earthquake were extensive in scale and would have taken an organized and numerous population a good deal of time to effect. Thus the scope of the architectural restorations is in harmony with the evidence of the pottery in requiring a period of years, probably even decades, between earthquake and eruption. Of equal significance is the fact that, while some houses were totally demolished, most were salvaged to some degree, so that the basic settlement plan of Akrotiri as preserved under the pumice of the volcanic explosion is that of the pre-earthquake phase at the site (very early Late Cycladic [=LC] I) rather than a novel creation of mature LC I. In Minoan terms, the final abandonment of Akrotiri took place late in, but not quite at the end of, LM IA; in Mainland Greek terms, the abandonment dates to LH I, some time bfore the final use of Grave Circle A at Mycenae in LH IIA.
The absence of any bodies and the dearth of metal artifacts or other portable objects of obvious material value in the ruins of Akrotiri clearly indicate that the inhabitants had ample warning of the imminence of the volcanic eruption which buried the island so deeply in ash and other volcanic debris that it became uninhabitable for as much as a century or two. At Akrotiri, the lowest stratum of this volcanic debris consists of a thin layer of pellety pumice some 3 cms. thick, the top of which was crusted as though water had fallen on it after its deposition. Slight oxidation of this layer suggests that it was exposed to the atmosphere for anywhere between two and twenty-four months before itself being sealed by a subsequent pumice fall. The first layer of pumice, preserved as a significantly deeper stratum in locations on Thera closer to the volcano than Akrotiri and less exposed to erosion, may in fact have been the warning which induced the Therans to flee, since it probably lacked the volume to have caused extensive damage or loss of life. A second stratum of rather larger pumice varying between 0.50 and 1.00 m. thick at Akrotiri but again deeper elsewhere on the island then fell. The final deposition of tephra (volcanic ash) attributable to this eruptional sequence is over five meters thick at Akrotiri but up to fifty meters thick elsewhere on Thera and includes large boulders of basalt in addition to the lighter and smaller bits of pumice which themselves now measure as much as fifteen centimeters across. There is no archaeological evidence for how long the full series of eruptions lasted, but vulcanologists have reached a consensus that the process was a fairly rapid, hence short-lived one. The absence of any clear signs of erosion at the preserved tops of the ruins of Akrotiri supports the notion that complete burial of these ruins followed close upon the heels of the events which produced the ruins in the first place, that is, the initial stages of the eruption.
The distribution of pumice derived from the eruption is quite well known thanks to a series of deep-sea cores recovered from the southern Aegean and some careful sampling of strata exposed by both archaeological excavation and road-cuts on the island of Crete. Not surprisingly, in view of the prevailing wind patterns in the Aegean, most of the pumice from the eruption is found to the southeast of Santorini. The Greek Mainland and western Crete would have been altogether unaffected by the ash fall, but eastern Crete would have been covered by a maximum of ten, and more probably by between one and five, centimeters of fine pumice. Archaeologists eager to establish a correlation between the Theran eruption and the collapse of Neopalatial Crete feel that such a quantity of ash would have had a disastrous effect on agriculture in eastern Crete. However, others point out that such a relatively thin layer of pumice would have been eroded away by wind and rain within a year or two and would in fact enhance rather than detract from the fertility of the soil. A layer of Theran ash was identified in the late 1980’s in some lake sediments in western Anatolia, indicating that the windborne dispersal of this ash had a much more northern and eastern distribution than previously suspected.
Often associated with the eruptions of insular volcanoes are tsunamis or tidal waves. In the case of Thera, a tidal wave would have been created by the collapse of the magma chamber within the volcano and the creation of a large, deep crater or caldera into which the sea would have rushed. For many of those seeking to connect the Theran eruption with the sudden decline of Minoan Crete in the fifteenth century B.C., the major destructive aspect of the eruption has been not the ash fall but the associated tidal wave. In the middle of the debate in the mid-1970’s over the nature of the Theran eruption and its effects, Doumas in fact claimed that the collapse of the magma chamber and hence the appearance of the tidal wave was an event which postdated the volcanic eruption itself by a decade or more, thus explaining how events on Santorini directly caused the collapse of Minoan civilization even though Akrotiri was buried in late LM IA while the wave of destructions of sites throughout Crete which defines the end of the Neopalatial period cannot be dated earlier than LM IB. More recently, the vulcanologists have claimed that the Santorini caldera formed quite gradually and that a tidal wave, if indeed there was one at all, would not have been on anything like the scale envisaged by Marinatos and other proponents of the link between the Theran volcano and the sudden decline of Neopalatial Crete.
Despite numerous and varied arguments by a host of reputable scholars [e.g. Marinatos (1939), Page (1970), Doumas (1974), Luce (1976)] that one or more of the events associated with the period of extreme activity of the Santorini volcano surveyed above [i.e. earthquake(s), ash fall(s), tidal wave(s)] had a direct and disastrous effect on Neopalatial Minoan civilization, the simple facts are that the great earthquake which badly damaged Akrotiri is to be dated quite early in LM IA (either ca. 1650 or ca. 1560 B.C.?), that the entire town was buried in meters of volcanic ash still within the LM IA period (ca. 1625 or ca. 1550/1540 B.C.?), and that the wave of destructions (most of them including fires) which defines the end of the Neopalatial period on Crete and to which the palaces at Mallia, Phaistos, and Zakro all fell victim cannot be dated earlier than LM IB (ca. 1480/1470 B.C.?). Hood [TAW I (1978) 681-690] claims that clear evidence of the earthquake which so severely damaged Akrotiri before the town was buried is to be found at several sites on Crete where it is clearly dated to LM IA. More importantly, tephra from the later eruption of the Theran volcano has been found within the past decade in LM IA contexts on Rhodes (at Trianda) and Melos (at Phylakopi) as well as on Crete itself, ample confirmation that the eruption preceded the LM IB destruction horizon on Crete by a significant amount of time. Thus no direct correlation can be established between the Santorini volcano and the collapse of Neopalatial Minoan civilization.
The exceptional state of preservation of all artifactual categories buried deeply under meters of volcanic debris from the eruption of ca. 1500 B.C. has almost automatically made Akrotiri one of the two or three most important prehistoric sites in the Aegean. The study of perhaps no other category of artifact has been so deeply affected by the discoveries at Akrotiri as that of mural painting or frescoes. Not only are the Theran frescoes preserved in larger fragments than the more or less contemporary murals from other sites in the Cyclades (Ayia Irini, Phylakopi) and on Crete (Knossos, Tylissos, Ayia Triadha) but most of them can be assigned to specific positions on particular walls which are themselves extremely well preserved, so that the decoration of entire rooms can be reconstructed with a considerable degree of confidence. Most of the frescoes thus far published from Akrotiri are figured, but rarely occur in more than one or two rooms within what is identifiable structurally and functionally as an individual building unit. A whole series of fascinating iconographic problems has surfaced as the result of the recovery of these magnificent paintings. For example, what is the function of mural painting within a building (i.e. in what rooms/spaces does it appear, with what artifactual associations, and depicting what sorts of figures and/or scenes)? How are the different scenes within a given room, sometimes at different scales and at different levels on the walls, to be “read” as a unit? Indeed, are they necessarily to be viewed as closely connected thematically simply because they appear in the same room?
The major groups of frescoes from Akrotiri published thus far are:
1. West House:
(a) Room 5: Two life-sized nude fishermen in narrow panels below aof variable width on the upper wall showing a fleet of ships moving between two towns (south wall; 40 cms. high), a riverine landscape (east wall; 20 cms. high), and a religious ceremony on a hill and warriors disembarking from their ships in two seemingly distinct scenes (north wall; 40 cms. high). The ran along the west wall as well, but little of this portion of it has survived.
(b) Room 4: A series of life-sized stern cabins (ikria) on the north wall, a thin partition shared with Room 5 which bore the fleet scene on its other side; a life-sized “priestess” from the eastof the doorway connecting Rooms 4 and 5; flower pots with lilies on the two jambs of a window in the west wall, the sill and the lower jambs of the same window being painted to resemble veined stone such as marble or . The rest of the room was in the process of being plastered and painted when work was suddenly stopped and the site was abandoned.
2. Sector Beta, Room 6: Blue Monkey, closely comparable to a more fragmentary and stylistically somewhat earlier composition found in the palace at Knossos.
3. Sector Beta, Room 1: Pairs of gazelles on two or three walls juxtaposed with a single pair of boxing boys in one relatively narrow panel.
4. Sector Delta, Room 2: Springtimeor of the Lilies.
5. House of the Ladies, Room 1: The west half of the room features life-sized clumps of papyrus, while in the east half two life-sized women dressed in a Cretan fashion wait on other figures who are largely missing.
6. Xeste 3, Room 3: The frescoes from this area of the building decorated the walls of both the ground floor and an upper storey. On the ground floor, the northeastern part of the room was occupied by a sunken “” of Minoan type. At the level of the upper storey, young women on the north and east walls gather crocuses in a rocky landscape and bring them from both sides to a central “goddess” seated on a platform supported by altars on the north wall. Immediately flanking the “goddess” to left and right and in postures of worship/adoration are a monkey and a respectively. On the north wall at the ground floor level, three more girls appear as follows: at the left, a girl walking right and holding out a necklace in one hand; in the center, a seated girl facing right and clutching her forehead in pain because she has hurt her foot, which is bleeding; and at the right, a girl walking left but facing right toward the door or altar on the east wall. The east wall is entirely occupied by what appears to be an ashlar wall with an elaborately decorated, closed door at its center, directly above which is a pair of “ ” dripping with a red substance which is likely to represent blood; the “wall”, “door”, and “ ” may all together constitute a large altar toward which the attention of all the girls on the north wall is directed. Other fragmentary figures, including more girls as well as at least one male figure, are considered to belong to the decoration of the west and south walls at both levels.
Male figures appear to assume greater importance in at least some Theran paintings (e.g. the fishermen, warriors, and captains from Room 5 of the West House; the boxing boys from Room B1) than in most Minoan painting, but the female plays an important and often dominant role in several Theran compositions (Xeste 3; House of the Ladies; “priestess” from the West House) and often appears dressed in a thoroughly Minoan fashion (but note the unusual garb of the “priestess”).
Relief frescoes are thus far unknown from Akrotiri, as are large scalecompositions and bull-jumping scenes, all of which are particularly characteristic of Knossian palatial murals. The compositions of Xeste 3 on both levels may have constituted forms of processional scenes, on the ground floor toward the altar on the east wall and on the upper storey from both directions toward the seated “goddess” on the north wall, compositional schemes paralleled at Knossos both in the Corridor of the Procession and on the Grand Staircase.
In details of dress (including flounced skirts, tight-fitting short-sleeved jackets leaving the breasts exposed, textile patterns, and some forms of jewelry) and in depictions of scenes of nature (e.g. the Blue Monkey and Springtime frescoes), the Theran murals often closely resemble Minoan wall paintings. At least two features common in the frescoes from Akrotiri, however, are not well paralleled on Crete and are unlikely to be patterned after Mainland/Mycenaean models either. The first is the peculiar hairstyle affected by many figures of both sexes in which much of the head is painted blue (probably indicative of a shaved head rather than of a specially deesigned skullcap of some sort) and only a few long locks whose positions vary considerably from individual to individual are indicated in black. It is possible that differing hairstyles are indicators of age, as Koehl has argued is the case for similar though seemingly not identical variations in male (but not female!) hairstyles in Neopalatial Crete. The second feature possibly peculiar to Thera is the wearing by females of exceptionally large earrings. Identical earrings from the Shaft Graves of Circle A at Mycenae are better viewed as evidence for the adoption of a Cycladic fashion by wealthy and progressive Mainlanders than as indicating the presence of Mycenaeans on Thera at this time.
The miniaturefrom Room 5 of the West House, arguably the most complex single work of art of the Aegean Bronze Age thus far recovered in anything like its entirety, has already provoked an enormous literature in which a broad spectrum of opinion is represented on such topics as the identity of the ships in the fleet scene, the identity of the warriors carrying the tower shields and wearing boars’-tusk helmets on the north wall, the locale of the two towns on the south wall, etc., etc. It is questionable whether such problems can be resolved with the evidence presently available. However, it is worth observing that scenes such as the towns on the south wall and the disembarkation of warriors from their ships on the north wall have contemporary (e.g. the Silver Siege from Grave Circle A at Mycenae) as well as earlier (the Town Mosaic from Knossos) parallels in other media. Note also that the ships of the fleet have several features in common with the vessels incised on EC frying pans a millennium or so earlier in the Cyclades.