Santorini and its eruption in the Late Bronze Age
Chapter 2. Earth: The Minoan eruption in its archaeological landscape
(For the pdf version of this document with correct page references download the chapter 2 pdf.)
• A consideration of three aspects of Minoan culture at risk
The Santorini eruption is, in principle, of sufficient magnitude to have caused serious damage to Bronze Age sites in the Aegean . The most densely populated area, Crete , lay approximately 120 km to the south, consisting of a variety of sites of different sizes and components from villages and ports to major towns. The two broad elements that might affect a particular site are tephra fallout and tsunami inundation. In order to determine how much damage the eruption may have caused I have created a schematic for a 'typical' site and divided this into three zones (figure 2.8). These zones reflect an object orientated approach rather than a more intangible socio-political or religious one, which will be dealt with in the summing up of the effect on the culture as a whole. The three zones are Minoan buildings, the coastal littoral, and crops and subsistence. Not only will each of these zones be affected in different ways, their preservation in the archaeological record is very different as is the amount of research undertaken on them.
Figure 2. 8 Theoretical model of three zones at risk, buildings, crops and the coastal littoral. White blocks represent buildings. Specific items of interest are designated by a letter.
In figure 2.8 the three zones are shown by the red dotted lines, the letters indicates specific features. Area A shows small boats used by local fishermen drawn up onto the beach area, whilst B indicates the main harbour used by some of the larger vessels. The structure of Bronze Age harbours may have been more complicated than was originally thought (Shaw 1990, Raban 1991). Two vessels are drawn up to quays for unloading, one quay is made of wood and the other of stone. A ship is also anchored further from shore in the centre of the bay. Area C represents warehouses and shipsheds in the coastal zone. Building D indicates a shrine outside the main concentration of the site. Crops at E represent delicate crops that need a specific and rare environment. Area F is the main subsistence area close to the site which is used for a variety of fields and crops. Area G is the main site and shows a number of different building types and sizes. The building at H represents a temporary shepherd's hut where the majority of sheep/goats are kept at a particular time of year. The field at I is situated by the river, but this means that it is more vulnerable to a tsunami propagating up the lower lying river area. Finally J shows a number fields and crops in the coastal zone and temporary buildings used by the fishermen. This theoretical model tries to illustrate some of the complexity of use in the Minoan landscape and how much information is missed if excavation is concentrated on the main settlement area.
The majority of archaeological work has focussed on Minoan buildings, and in particular on the best preserved buildings at a site since they can by inference provide the greatest amount of data. The result of this is that buildings from outside the core area of a site are often ignored or are simply not found. Survey work may provide some details on them but they would not necessarily be excavated. Temporary structures such as shepherds' huts may not survive at all.
Data on the coastal littoral during the Late Bronze Age on Crete is particularly poor. The change in sea level and the fragmentary nature of remains are just two reasons why it is has been poorly studied in relation to a culture that obviously had a great deal of interaction with maritime resources.
The subsistence resources of the Minoan population is of critical importance in the debate on the impact of the Santorini eruption but evidence is ephemeral and often extrapolated from a single site across the whole of the Minoan culture. The Late Bronze Age period is not covered by a single pollen core from Crete (Rackham and Moody 1996, p125) .
• Minoan buildings
The most direct method for collateral damage to the Minoan culture is through the fallout of significant amounts of tephra on Minoan sites leading to a structural collapse of buildings and therefore the possibility of injury. Any structural damage would probably occur through the overloading of tephra material on roof structures leading to the collapse of upper stories (Blong 1980; Blong 1984) . The site of Akrotiri and other sites on Santorini were directly affected by a wide range of phenomena such as bombs, direct blast and pyroclastic flows. Sites on Crete however would not have been subjected to bombs or large projectiles which would have fallen out before reaching Crete and if there was any fallout it would be of tephra particles settling in a similar way to snow fallout. The weight of the material could then form a critical load and the supporting structures give way. Blong has suggested that a figure of around 10 cm of tephra with a density of around 1000 kg/m 3 may have been enough to cause structural damage (Blong 1980, p221) . This figure will decrease if the tephra became wet, and so rain following on from a tephra fall could have had a greater effect.
The evidence for Minoan roof structures is very fragmentary, mainly because in most sites nothing above the footings of walls has survived and only minimal amounts evidence can be extrapolated about the roof construction. Therefore any comments about roofs will be a synthesis of material drawn from archaeological evidence, iconography of the period and ethnographic resources.
• Archaeological evidence
The range of building types across Late Minoan culture is extensive, from the multi-storied palace of Knossos visualised, for example, in one of Piet de Jongs' pictures (e.g. Higgins 1973, facing p33) to the stable/store from Ftellos on Santorini (Doumas 1973, p162) . Many of the main synthesis works on Minoan architecture, such as The Palaces of Crete (Cadogan 1980) or the Palaces of Minoan Crete (Graham 1987) , have concentrated on the central palace buildings rather than the ordinary buildings that would have made up the majority of structures. There are however some basic principles of construction that probably held for both domestic and palatial buildings. The walls would usually be built of unfired mudbrick or adobe built upon low stone footings in a tradition going back to the Neolithic (Graham 1987, p149; Dickinson 1994, p144) . The height of these footings would depend upon the building but the upper story would consist of mudbrick, (except for Phaistos see Graham, 1987, p151). Foundations for walls appear to have been limited in depth or generally do not exist (Dickinson 1994, p147; Graham 1987, p149) . Mud mortar would be used to face the majority of buildings, hiding the often haphazard construction of the wall itself. Specific features may have been constructed of different stone types, such as gypsum in the grander buildings. Wood would often be used to brace or add support to the building, either internally within the walls or sometimes externally creating a half-timbered effect. There is no evidence for the use of tile in this period in Crete .
The majority of buildings must have been made up of domestic dwellings and it is convenient to use the subdivision of types as produced by McEnroe (McEnroe 1982) . He divided Neo Palatial housing into three groups on the basis of size, construction and types of rooms. As McEnroe points out each building is unique and therefore any distinction between houses may be completely arbitrary. Therefore he uses combinations of attributes or characteristics to form the boundaries between his groups. Writing in 1982 he was working from a corpus of approximately 180 houses.
Type 1
Type 1 houses contain the most distinctive set of rooms and include such areas as lustral basins, pillar crypts and light wells, and correspond to the sets of rooms that can be found in the palatial complexes,
" …the relationship between the Type 1 houses and the palaces cannot be overstated; it is the most important characteristic of this group of houses."
(McEnroe 1982, p5)
Type 2
The main characteristic of the Type 2 structures is that the residential and storage/residential areas of the house are combined within the ground floor space.
McEnroe divided his Type 2 into two subgroups, a and b, the difference being that Type 2a adopt some elements of the Type 1 house.
Type 3
These houses are usually less than half the size of Type 2 houses, they do not contain Type 1 details and they only rarely have details finished off with metal saws or chisels. They make up the majority with about 70% of McEnroe's corpus.
Using the figures presented in McEnroe's table 1 the average areas for the three groups are:
• Type 1 352 m 2
• Type 2 316 m 2
• Type 3 128 m 2
(McEnroe 1982, table 1, p18) .
McEnroe also makes some mention of local traditions. Architecture at a particular site may well have been influenced by the availability of local resources. This is evidenced, for example, at the island site of Pseira where although some walls were constructed of limestone boulders many are dry walled of schist and limestone slabs. This factor of local tradition has also been used to explain the difference between buildings on Kea which are also of schist slabs and the use of ashlar blocks evidenced at Akrotiri (Barber 1987, p12) . Thus the apparent Minoan influence at Akrotiri may actually be due to availability of resources biasing the archaeological record.
Another interesting conclusion of McEnroe's work is that of the 46 excavated buildings at Gournia, all but the 'palace' are of Type 3. This may be a construct of tradition, and when a new house is built it is modelled on those around it and has a similar layout, thus restricting the innovative element and proving that the extrapolation of data over limited sites is viable.
Hallager (1990) specifically looked at the question of roofs and upper stories and proposed a reconstruction for two buildings from Chania. He agreed with McEnroe's division of Minoan domestic housing but made a significant point about the area of the ground floors, with his opinion being that this area is too large for a normal dwelling space. This is coupled with the lack of open areas in LM I sites. He used five criteria to reconstruct the second story.
• Staircases
• Different construction and thicknesses of ground floor walls
• Preserved fragments of roof/ceiling construction
• Architectural remains , such as thresholds fallen from above
• Drainage systems connected to the upper story
From this his main conjecture was that much of the second story space was not devoted to sealed rooms but was rather an open space which would be used for a variety of activities including storage, preparation and cooking (Hallager 1990, p289) .
Plate 2- 1 Roof construction from a building in Tunisia abandoned in the 1960's.
The actual data on roof construction is limited but is probably relatively similar to the method used in the area up to the introduction of concrete. The greatest amount of evidence comes from Santorini where timbers used in the construction process are preserved as voids. Floors and ceilings were supported by beams across which were laid slabs, or probably more commonly reeds or branches. These were then overlaid by vegetation such as hay or spiny burnet on top of which there was a thickness of earth. If this was the roof then the earth would be waterproofed in some fashion (Rackham 1978) . Plate 2-1 illustrates this type of roof construction from a house in Tunisia which was abandoned around 1960. Some of the roof has collapsed revealing in section how the roof is constructed. Timbers support a layers of reeds upon which is a layer of leaves; this is then sealed by a number of earth layers providing the waterproofing.
Graham assessed the palace roof structures and believed it unlikely that there would be three stories to the main Knossos complex (Graham 1979) . The stresses combined with the tectonic nature of the area would not have made this viable. Rather he suggested that the riser that Evans identified as leading up to a third floor forms a stairwell leading to the roof of the second story. He conjectured that from here views could again be gained of the Central Court activities. Small structures over such a stairwell to protect them from the elements would explain the attic-like structures visible on the Town Mosaic (Morgan 1988, fig 53, p68) .
A problem with Graham's interpretation is an over emphasis of the religious and ceremonial nature of the palace structures and the fact that reconstruction's are based on an ideal rather than the practical. Klynne (1998) discussed the appearance of the palaces in graphical reconstruction's. He suggested that the cleanliness of the sites and the aesthetic are over emphasised also that the architecture is often a concrete blankness with sharp lines. They are a construction as much of a modern viewpoint as of reality and based fairly rigidly on Evans work, which was of its time, but also was necessary to be able to excavate the areas underneath (Klynne 1998, p207) . The inability to recognise palatial structures in Minoan iconography may be linked to a fixed view of their appearance, central courts with tripartite shrines topped by horns of consecration. The proposal put forward by Klynne for a reconstruction of Knossos is for a much less sterilised site (Klynne 1998, fig 9, p223) . Individual elements are emphasised so that sections of the building are painted and decorated differently and it is obvious that some were built at different times. He emphasises the importance of the roof space as an activity centre with projections, parapets and temporary coverings.
Seen in this context the roof area becomes more vulnerable to tephra fall. Parapets and obstructions may have led to a drifting of material which might have increased the chance of roof collapse. Tephra could also have blocked drainage channels from the roof, which depending upon the time of year may have increased risk of collapse. Also food stuffs and preparation areas on the roof may have been contaminated by the tephra.
• Iconography
The iconographic evidence for buildings is limited to a number of sources, namely frescoes, seals and sealings, house models and furniture inlays. The most important of these are fresco depictions and in particular the Miniature Fresco from Akrotiri which shows two towns (Morgan 1988) . Seals and sealings provide limited evidence mainly because the small scale of the media predicated against many buildings being shown. The tripartite shrine appears to be the most popular form shown in this medium. The House model from Archanes is a clay model of a Minoan house that confirms in the round many of the details outlined by the archaeological evidence, in particular the flat roof area and parapet. The Town Mosaic is a series of faience plaques discovered by Evans at Knossos which may have been from an inlaid piece of furniture (Evans 1921, fig 226). They show individual houses and buildings from a front-on 2-Dimensional perspective with the viewer placed centrally and suspended in mid air.
The depiction of multiple buildings and by inference a town or village needs a particular medium, which is influenced by scale. It needs to be large enough to depict a scene, but the individual buildings also need to be reduced in size. So called miniature frescoes depicting buildings exist from Ayia Irini on Kea and there is a fragmentary example from Tylissos but the best example is the Miniature Fresco from the West House at Akrotiri which depicts two towns as well as other structures.
All these sources confirm the individual nature of many buildings. There is great detail on the Miniature Fresco but distinguishing between stories and different buildings is difficult. Colour appears to be used in registers on the buildings reflecting different plasters and white washes. Some ashlar details can be distinguished but much detail such as mud brick is hidden beneath plaster facade.
The elements that are most important for this study are that the buildings are 2 or 3 stories and have flat roofs which must have been accessible because people are watching from them. There are parapets and stairwells as well as small dome-like features.
• Summary
Using the general principles outlined by Blong for tephra fallout a number of conclusions can be drawn about the fallout of tephra on buildings (Blong 1984, p206) . It is likely that there would have been sufficient wind to create uneven distribution of fallout; many of the Minoan sites are coastal sites which have significant changes of sea breeze during the day. The parapets and structures on flat roofs combined with wind drifting would probably lead to asymmetrical loads on roofs. However, as was noted above it appears that the Minoan roof space was a fully functioning part of the building structure. If areas of the roof were used for storage then the amount of weight that these structures could bear may have been significant. The depth of 10 cm of tephra fall should thus be seen as a minimum depth of tephra that could cause some structural collapse in buildings. In areas experiencing more than 10 cm of deposit then structural collapse of some buildings is more likely. The roof space appears to be an integral part of a functioning Minoan community and was probably used for storage, food preparation and some industrial activity. Tephra fall of less than 10 cm may have contaminated some foodstuffs and caused minor damage but such an amount was probably disruptive rather than catastrophic.
• The coastal littoral and maritime zone
The coastal littoral or maritime zone components include ships, harbour installations and cultivated land. These components of Minoan sites situated on the north coast of Crete could have been at risk from tsunami inundation. A number of articles from Marinatos (1939) to Antonopoulos (1992) stress the massive destruction that could have been caused by tsunami waves, but these arguments are usually a gross simplification of not only the area at risk but also of the possible damage caused. To be able to assess if this threat from tsunami inundation was a significant one I feel it is necessary to draw up some conclusions about how the Minoans interacted with the sea, the types of ships that they used and how they secured these vessels.
The main difficulty with trying to reconstruct the potential damage to this littoral zone is that the evidence is very fragmentary and poorly researched. There are a number of reasons for this but a breakdown of some of them is given by Raban in his discussion on Minoan and Caananite harbours (Raban 1991, p136) .
• The preconception that harbours were only naturally sheltered bays where ships could be drawn up onto sandy beaches and were lacking in any man made features.
• The interaction between earth sciences and 'traditional archaeology' was limited in the region and ignored by a number of major excavations which could have carried out this type of study.
• The cost of involving expertise from other areas and expensive equipment to examine both land and near shore sites was too much for most excavations when well preserved in-situ deposits were available inland.
• The regulations and different permits required to investigate land and sea zones work against a holistic approach to this problem.
• The significant change in topography means that the search for a harbour site is difficult and will not always be successful.
These factors need to be borne in mind when trying to assess this important and poorly understood zone. The change in sea level since the Late Bronze Age is particularly relevant and this is dealt with in depth in chapter 5. The main conclusion though is that the north coast has dipped into the sea on the eastern side, but has been raised out of the sea on the western side.
The evidence for maritime subsistence products is dealt with below in section 2.5.3.1. The coastal zone would have been at risk from tsunami inundation with many sites situated on the low-lying coastal plains. Inundation of fields and wells may have occurred leading to saline poisoning but such an event cannot be quantified. Driessen and Macdonald (Driessen and Macdonald 1997, p100) state that a number of wells went out of use or were freshly dug in this period but there is no direct evidence of a link to Santorini.
• Ports and harbours
The only Minoan port site to have been examined in depth is Kommos on the south coast which serviced the Messara plain (Shaw and Shaw 1996) . The reconstruction drawings of Building J/T and Building P which are presented in Shaw's paper on Aegean harboursides reveal a sophistication and scope lacking from the excavated remains (Shaw 1990, fig 8, p424 and fig 9, p425) . A combination of change in sea-level, tectonic forces and winter storms has battered this poorly preserved resource. Whilst it may be that some sites used lighters and beached vessels to unload supplies, the Arrival town with its double harbour in the Miniature Fresco from the West House and the unloading of produce shown by the Miniature Fresco from Kea (Morgan 1990, fig 2, p255) are indicative of a more sophisticated harbour arrangement. Chapter 6 shows walls and structures of probable Minoan date at a number of coastal sites on Crete that were noted during fieldwork on the island in 1999 and 2000.
• The Minoan Thalassocracy
The question of the Minoan relationship with the sea has become intimately bound with a description by the Classical writer Thucydides. He mentions a Minoan thalassocracy, or rulership of the sea, when the legendary king of Crete, Minos, kept the sea lanes free of pirates and colonised the Aegean islands (Book 1:4 and Book 1:8). This has meant that many discussions on Minoan interaction with other areas have focussed on concepts of domination and whether trade was based on the state or the individual. Cherry (1999) in a recent introduction to the Minoan economy stated that we were a long way from,
"..naively unproblematized charaterisations of the Minoans as a nature-loving folk, under the beneficent rule of priest-kings, enjoying the fruits of a pax minoica and the Minoan thalassocracy."
(Cherry 1999, p18) .
I am not sure that this has filtered through to disciplines outside archaeology and it is obviously difficult to get away from the term when discussing the interaction of the Minoan heartland of Crete with that of other Aegean centres. A recent in-depth scientific study of Late Bronze Age pottery was termed 'The Minoan thalassocracy reconsidered: provenance studies of LH IIA/LM IB pottery from Phylakopi, Ay Irini and Athens ' (Mountjoy and Ponting 2001) .
In my opinion the term should not be used since it creates the image of a large centrally organised fleet for which we have no evidence. In terms of the present discussion this is important since the eruption cannot have destroyed a brittle, centrally controlled, politically dominating fleet if we have no evidence that such a construct existed.
• Ships and shipping
No actual shipwrecks of the period have been identified but there are some wrecks from which some conclusions can be drawn. The Dokos wreck has Early Helladic II pottery and other items but the lack of preserved timbers and the findings of material on both sides of Cape Myti Komeni have raised doubts that this is a wreck (Wachsmann 1998, p205). Cape Iria , the other shipwreck found in Greece dates to the 13 th century BC as does the wreck from Cape Gelidonya off Turkey (Wachsmann 1998, p205-208) . The shipwreck of Ulu Burun off the Turkish coast is the most impressive and the best preserved (Bass 1986; Bass 1987; Wachsmann 1998) . A number of absolute dates for some dunnage from the wreck have been published (e.g. Kuniholm et al 1996; Manning 1999, p345), the latest published date is of 1327 +4/-7 BC (Manning et al. 2001, note 38) . The extraordinary wealth and preservation raise fundamental questions about the nature of Late Bronze Age trade. The layout of the cargo could be reconstructed (Bass 1987, p694) and enough timbers were preserved to show the construction was by pegged mortice and tennon, pushing back our knowledge of this technique by a thousand years (Wachsmann 1998, p206) .
The iconographic evidence is the largest comprising many depictions on seals, models and frescoes and in particular the Miniature Fresco from Akrotiri (Wachsmann 1998, Chapter 6) . The Miniature Fresco provides many details on Minoan ships and has been studied and interpreted by a number of different researchers and there are issues of its cultic significance and the role of a water procession (Morgan 1988; Wachsmann 1998 for summary) . The main issue from the point of view of this study is that there were a wide variety of seagoing craft from small fishermen's boats to large merchant ships. These boats represented a great deal of effort and wealth in an area where timber was a limited resource. The loss and wreckage of boats and ships by tsunami inundation could have caused a long term damage to both large and small communities that relied on shipping and trade to function and survive.
• Trade and interconnections
An unusual element of Minoan shipping is that there is little evidence for anchor devices. Most cultures at the time used large drilled stones for attachment to hawsers to anchor their ships at sea. A small number have been found at Minoan sites (one at Akrotiri, two at Kommos for example), but not enough when compared to other Mediterranean areas. They may have used smaller unworked stones encased in wicker as a substitute and these leave little evidence behind. Stone anchors have been used to track the ports of call used during the period (McCaslin 1980) . However a review of McCaslins thesis on stone anchors and maritime trade routes by Knapp (Knapp 1985) criticises this study for still being reliant on an object orientated approach to archaeology. Knapp states that any study of international trade at this period should draw upon ethnographic studies and theories of economic anthropology, (whether the dominant paradigm be Substantavist, Formalist or Marxist is unimportant), to enhance the object approach.
This difference in approach is emphasised in the Aegean sphere by two books on trade both of which call upon Homeric echoes to alert the reader to their subject matter. Ulysses' Sail (Helms 1988) examines the ethnographic ideas behind trade and the power that long distance trade can generate, whilst Sailing the Wine Dark Sea (Cline 1994) essentially catalogues the artefacts that define long distance trade for the Late Bronze Age Aegean. Cline's work shows the problem of only relying on archaeological object evidence, the Ulu Burun shipwreck for example is so rich that it provides anomalous data. Table 60 (Cline 1994, p99) shows the incidences of Canaanite Jar for all sites in the Aegean totalling 93 for the Late Bronze Age; this compares with a figure of over 140 recovered from Ulu Burun alone.
In a study based on Helms' approach Broodbank has used point analysis between islands and the geological sourcing of materials to show that there was complex interactivity between Cycladic islands in the Early Bronze Age (Broodbank 1993) . He concludes that the physical result of trade, i.e. artefact exchange, might not be the only justification for distance travel and the use of boats but might equate with a social or psychological need as well. Whilst such a discussion is beyond the remit of this work it does show that by the time of the Late Bronze Age there may well have been a multi-faceted complex tradition of interaction between Cycladic islands, and by this stage the Helladic and Minoan spheres as well, and that the disruption caused by the Santorini eruption may have had not just a physical manifestation, the loss of Santorini, but also a social and psychological blocking and disruption of interaction.
• Summary
The fragmentary nature of the evidence in the coastal zone has meant that it has been little studied. Rather, many works have concentrated on issues of trade and interconnection at the object level or on the issue of the Thalassocracy of Minos from Classical literature (Hägg and Marinatos 1984; Wiener 1990; Knapp 1993) . It is my opinion that this zone which provided food, tradable items, contact and identity was at risk from the Minoan eruption. Because of the disruption caused by the possible destruction of shipping, food supplies and coastal buildings this effect may have continued for a number of years and possibly caused some communities to fail.
• Crops and subsistence
The third major zone that would have been at risk from the Santorini eruption was the food supply of the Minoan settlements. This threat could have come from a number of sources, including tephra falling on crops and destroying them, tsunami inundation of coastal and low-lying areas leading to destruction and salt pollution, sulphur content from the eruption creating noxious gasses and affecting animals, climatic change leading to a loss of crops.
All of these ultimately have the same result; possible starvation of the Minoan population. This in turn could lead to stress in the community leading to internecine conflict. It will be shown however that there was great diversity both of the agriculture exploited and of the ecological zones.
• Subsistence evidence
Our evidence for crops and animals exploited during this period is limited. There are two main direct sources of archaeological evidence; excavated sites producing animal bones and seed remains and pollen cores. Other sources of information include ethnographic information, processing tools and equipment from archaeological contexts and some elements from Linear B texts. Rackham and Moody in their work on the Cretan landscape detail the difficulties of reconstructing the wider landscape during the Late Bronze Period, with fragmentary archaeological evidence and no pollen cores from Crete for the period (Rackham and Moody 1996, p125) .
The mainstays of Bronze Age agriculture have been seen as a supposed triad of three cash crops; cereals, the vine and the olive (Renfrew 1972) . These cash crops allowed the build up of a stratified society and dedicated craftsmen through the redistribution of surplus crops. Sarpaki would add legumes to this list as a mainstay of the economy, but one masked by poor preservation (Sarpaki 1992) . Although there have been discussions about whether certain crops could be regarded as being domesticated in the Early Bronze Age (Hansen 1988; Halstead 1996) , by LM IA the use of the olive and the vine seem established.
The subsistence economy of the Minoans appears to be based on a suite of crops supplemented by animal produce. Some of the crops grown or harvested, such as carob or vetch, could have been used as a fodder crop for animals. Animals could be used as a supplementary food source in times of stress, effectively acting as a walking larder.
Agricultural resources would also produce products for more than just sustenance, they would also have provided a major input into craft specialisation and provided tradeable goods. Thus sheep would be kept for wool and leather as well as milk and meat. Linen could be grown for cloth, the deposit of over four hundred loomweights from the West House at Akrotiri illustrating the factory nature of some of these products (Schofield 1990, p210) . Many of these goods may not survive in the archaeological record and their loss on a society may not be immediately catastrophic but could place considerable strain on a redistributive society.
For the period of the Late Bronze Age the site of Kommos on the south coast provides the most comprehensive and well studied site on which to base an analysis of agricultural resources (Shaw and Shaw 1996) . This is due to a dry and wet sieving regime of excavation and sampling backed up by a range of specialist studies encompassing oft neglected groups such as small mammals and fish (Reese et al. 1996) . The range of groups and species found at Kommos is probably typical for Neo Palatial coastal sites.
In their table 4.11 Shay et al. (1996, p123) show the charred seeds recovered from Kommos and their frequency by period. This illustrates the problem of data preservation with, for example, only two cereal grains recovered from the MMI to LM III . The data that have been collected however, combined with the charcoal wood data, suggest a mixed environment of natural, semi-natural and planted communities in the vicinity of the site. This confirms the use of olive, cereal, vine and legumes as the main crops but these are supplemented by such items as bean trefoil, almonds, carob, lentisc and pear. Table 2.3 shows the modern harvest times for some of the crops that might have been exploited by the Minoans (Food and Agriculture Organization of the United Nations 1959, p12) . The crops could be for immediate consumption or stored for later human use or animal fodder.
Crop |
Harvest period |
Main harvest period |
Wheat |
May-August |
June-July |
Barley |
June-July |
June-July |
Onions |
May-August |
May-July |
Dry beans |
July-October |
July-September |
Dry peas |
June-July |
June-July |
Lentils |
June-July |
June-July |
Chick peas |
June-July |
June-July |
Vetch |
May-July |
June-July |
Grapes, table |
July-October |
August-October |
Grapes, wine |
September-October |
September-October |
Carobs |
August-September |
September |
Olives, table |
October-December |
November |
Olives, oil |
October-May |
November-January |
Almond |
July-August |
July-August |
Flax |
May-June |
May-June |
The animal bone assemblage from Kommos compares with sites across Greece and is dominated by the three main species of sheep/goat (53%), pig (31%) and cattle (16%) (Reese, et al. 1996, p165) . The sheep/goat bones indicate that many are slaughtered young, which is consistent with exploitation for meat. Pigs likewise show that they were bred for meat, whilst the cattle were probably mainly used for traction and ploughing and only partly as a food resource. These main animals were probably supplemented by other species such as deer and hare.
Kommos also provides important information on the rarely recovered marine resources. In all over a thousand fish bones and otoliths were recovered from MM IB to LM IIIB contexts (Reese et al. 1996, p204) . The majority of the fish bones come from shallow coastal waters and the assemblage is dominated by small sea bream. There are some remains from fish found in slightly deeper water, such as grouper or cardinal fish. The presence of one tunny and one jack could indicate open water fishing. Nets and hooks were used to capture the fish. There were also a number of marine invertebrates, of which 98% come from rocky, shallow water shore locations.
• Storage
The element of storage within Neo Palatial society is crucial. Christakis outlines three types of food storage during this period (Chistakis 1999, p2) . Intra-corporeal is the accumulation of fat to help survival during a lean period. This does not really work for humans but livestock can be fed with fodder crops and surplus which is then retrieved when the animal is eaten. Social storage is where food is exchanged for social obligation which can be reconverted in times of stress. Material storage is where the food is directly stored at the household level. The range of foods listed for material storage include wine, water, salted olives, meat and fish, dried fruits, butter, cheese, carobs, beans, raisins, honey and beeswax (Chistakis 1999, p4) . These were stored in a variety of containers, of which pithoi are the most prevalent. Christakis used the division of households developed by McEnroe (1982) and noted that there was a large range of storage potential, but that 76% of households had storage facilities for no more than a productive season (Christakis 1999, p14). However, insect pests have been found within stored products on Santorini and these may have reduced further the amount of edible produce (Panagiotakopulu and Buckland 1991). Christakis found no evidence to support the contention made by Driessen and Macdonald (1997, p54) that there was a decrease in the palatial storage capacity and an increase in household storage in the LM IB phase possibly as a reaction to the Santorini eruption (Christakis 1999, p15).
• Summary
Parts of the Minoan subsistence economy would have been at risk from the eruption of Santorini. Crops in the field could be have been damaged by tsunami inundation or tephra fallout, marine resources may have been decimated because of the tsunami or shock caused by the eruption. Processing or storage facilities may have been destroyed or damaged. Any short term effect such as the destruction of a crop in the field could probably be dealt with by the society as it would be used to crop failure from other sources. What might have been more serious was if there were a combination of factors such as destruction of storage buildings, damage to crops and then a climatic down-turn for the next year which meant another failure. It is probably this synchronicity of factors which is the most harmful but also the hardest to assess. These effects will not be uniform across all Minoan sites however, and widespread destruction is unlikely, rather there might be some loss of life across sites and maybe the abandonment of particular sites which fail.
| Back | Next |
This website is devoted to the eruption of Santorini volcano in the Late Bronze Age and its impact on the cultures and civilisations of the time. The website is owned by Dr. David Sewell.