Version classiqueVersion mobile

Excavations at Sissi II

Jan Driessen

9. Bio-archaeological Assemblages and Ground Stone Artefacts

Methodological framework and preliminary results

Valasia Isaakidou, Alexandra Livarda, Christina Tsoraki et Rena Veropoulidou

Texte intégral

1. Introduction

1Integral to the excavation on the Kefali hill at Sissi has been the systematic collection of ecofacts, including animal and fish bones, plant remains and marine and land molluscs, aiming at a holistic understanding of Bronze Age society and economy. The on-going dialogue and excellent co-operation between the excavators and the bio-archaeology team (V. Isaakidou co-ordinator and zooarchaeologist since 2007, A. Livarda archaeobotanist since 2009 and R. Veropoulidou marine and land mollusc specialist since 2009) has led to the appreciation that the animal, plant and mollusc evidence are not merely additional data; in these early stages of analysis the contextualisation of the results has already provided valuable contributions to the understanding of site formation processes, use of space and organisation of life at prehistoric Sissi.

2The information assembled from these lines of evidence endeavours to shed light on various issues: what did people choose to cultivate, breed, farm and collect and why? What were the constraints and strategies employed to accommodate their subsistence needs? What agricultural and farming methods did they use and why? How different or similar were the habits of the Sissi inhabitants to their contemporaries living in other parts of the island, and hence, what were their connections and influences? The study of food preparation and consumption practices can go even further to allow insights into how people perceived and organised their space and which activities they chose to perform in internal or external/public spaces. Several other activities/practices are also investigated as ecofacts may be deposited in ritual contexts as offerings and used as raw materials for the manufacture of artefacts with practical, decorative or symbolic functions. In particular, the study of shell and animal bone objects and their manufacturing by-products allows insights into the chaîne opératoire of their production and use. Through such analyses aspects of craft specialisation, spatial organisation of production, consumption and deposition and associated processes of creation of social identities and reproduction in Bronze Age Crete can be explored.

3It is in this context that particular attention is being paid to ground stone artefacts (i.e. grinding and pounding tools, stone axes, ornaments and vases), which, with a few exceptions (e.g. Blitzer 1995; Bevan 2007), have received little analytical, methodological and theoretical attention in Cretan archaeology so far and even less integration with such related datasets. Their study (by C. Tsoraki ground stone specialist since 2010) apart from providing insights into the nature, technology and organisation of daily and craft activities, adds a valuable dimension to the contextualisation and study of bio-archaeological material, as these tools were actively employed to transform such natural materials into useful substances and objects.

2. Sampling strategy and on-site methods

4To be able to provide answers to these and many more research questions that arise from the study of all datasets, the bio-archaeology team had to ensure that a bank of all potential information would be available. To do so, a carefully planned strategy was devised to achieve a good representation of all types of data. Rather than collecting samples from a selection of potentially rich contexts, such as ash deposits or visible concentrations of various ecofacts, relying on the excavator’s personal judgement, a systematic sampling strategy was suggested. It is only this method that allows controlled detection of both presence and absence patterns across space and time, and thus, reliable identification of potential activity areas with a high level of confidence.

5A systematic soil sampling strategy meant that in addition to the recovery of large, visible to the naked-eye, bone, shell and occasionally plant fragments, during excavation, soil from the centre of every excavation unit (‘zembil’) should be collected and processed by flotation. Collection of soil from the corners of the units was not favoured in order to avoid contamination, especially when the unit boundaries were not clear. The protocol adopted for soil sampling largely follows the one devised by G. Jones and P. Halstead (University of Sheffield) for the excavation at Paliambela Kolindrou, Northern Greece, one of the few projects in Greece applying systematic and intensive soil sampling. The volume of soil samples is dictated by the dimensions of the excavated units, but 45 to 60 lt prior to dry-sieving is the standard volume from large units. The soil from small pits is collected in its entirety whereas in the case of large pits the aim is to take a sample from each stratigraphic layer. Maintaining a strategy for processing large deposits and detailed sub-sampling of these features was considered necessary in order to retrieve a wide diversity of material from as many contexts as possible. All information regarding associated finds is carefully recorded and the excavators of each trench have been marking the exact location of the collected samples on the site plan. This enables us to explore spatial distributions of bio-archaeological remains within buildings and across the entire site.

6In the case of destruction layers with visible concentrations of bio-archaeological material, such as storage deposits, an even more detailed sampling strategy is employed with smaller but more frequent soil samples from smaller areas, taking also into account the potential presence of storage vessels and their spatial relationship with sampled deposits. The rationale is to allow exploration of different accumulations of bio-archaeological remains, their association with other artefactual datasets, such as pottery and grinding tools, and identification of different crop processing activities within buildings.

7The flotation procedure has been carried out successfully with the assistance of several project participants under the supervision and co-ordination of Quentin Letesson (2007 and 2008) and Alexandra Livarda (since 2009). Processing is being conducted with a modified version of a York style flotation machine (French 1971) (fig. 9.1). The floated material is collected in a stack of two brass sieves with apertures of 1 mm (coarse flot) and 0.25 mm (fine flot) respectively, while the heavier fraction (heavy residue) is retained in a mesh with 1 mm aperture. All samples are left to dry under shade in order to ensure that the preservation condition of the charred remains is not affected by the direct sun light of the Cretan summer. Washing and organisation of shells and animal bones has been carried out with the aid of Laetitia Polain, who also helped in the preliminary scanning of animal bones, and Irene Kritikopoulos.


3. Material processing and analysis

8Processing of the different types of bio-archaeological samples is a time consuming process and it would not have been made possible without the valuable contribution of several students participating in the excavation under the supervision of the bio-archaeology team. In the early stages of analysis sorting of the heavy residues resulted in the recovery of bigger and heavier plant remains, small mammal bones, small bones of larger animals, neonate bones, fish bones, snails and other molluscs, but also beads, obsidian and other small artefacts. This material, together with the flots and any hand-picked ecofacts, is then submitted for the next stage of the analysis to the specialists.

3.1. Shell analysis

9The study of the molluscs required building up of a reference collection of contemporary marine fauna from the area, as well as the use of identification keys to assist taxon/species identification. Quantification of the specimens involves calculation of the Number of Identified Specimens (NISP), the Minimum Number of Individuals (MNI), the Number of Intact specimens (NI) and the Number of Fragments (NF). Each item is observed both macroscopically and microscopically for their condition (i. e. weathering, state of preservation) and presence of evidence for human processing (butchery or other marks of modification), recording relevant variables that will allow detailed taphonomic and processing analysis of the material. Presence, absence, fluctuations and relative abundance of different species are compared between deposits to explore spatio-temporal differences. Finally, density of shells is calculated according to NISP and MNI per 1 m3 in order to compare densities of shell assemblages between Sissi and other sites and to enable statistically valid comparisons.

10Shell artefacts also are observed macroscopically and microscopically in order to identify methods of manufacture and wear related to use and/or post-depositional processes. In particular, artefacts are recorded according to basic form and structures, while their plan and longitudinal sections are measured (length, width, thickness, as well as measurements of additional features, such as perforations and projections). Other variables recorded to describe artefacts include the presence of projections, incisions, surface treatment (polish or grinding) and indications of manufacture (facets, score-marks, cuts, shape of perforation). Recording of variables and subsequent analysis of collected data are undertaken using a database (Ostreadb), which was especially designed and constructed by Rena Veropoulidou and used for similar analyses and study of other assemblages. Documentation of the material also includes digital photography of specimens.

3.2. Animal bone analysis

11Standard zooarchaeological methods are employed (Isaakidou 2004) based, in the first instance, on the macroscopic and low magnification examination of the animal bones. For species identification atlases and modern reference skeletal material are used. A range of variables are recorded for selected specimens, including taxon/species, body part, taphonomic markers, such as state of preservation and scavenger (dog/rodent) attrition, location and morphology of butchery marks and other evidence of processing for food (e.g. fragmentation patterns and signs of burning) or processing for artefact manufacture (e.g. whittling, grinding), age data, measurements, pathologies and sex. Quantification follows the method devised by Halstead (in press) to provide counts of Number of Identified Specimens (NISP) and counts of Minimum and Maximum Anatomical Units (MinAU and MaxAU respectively). Statistical analysis of the resulting datasets is conducted using SPSS.

3.3. Archaeobotanical analysis

12Analysis of the archaeobotanical material involves sorting of all the coarse and fine flots under a powerful stereoscope in order to retrieve all plant parts, to assess their state of preservation and to record charcoal volumes. Morphological criteria are used for the identification of the plant species, based on modern reference material and several identification manuals and floras of the area. Quantification of all material is conducted with a variety of methods. In each sample the minimum number of individuals (MNI) is counted on the basis of the minimum number of characteristic plant parts. However, calculation is conducted twice according to both the apical and bottom/embryo ends of grains, seeds and fruit/nut stones for each species. A standardisation of the calculation method for each species will be assessed according to the quantification results at the end of all excavation seasons, when all the material will be collected. At present it is expected that in the case of cereals embryo ends will provide the most accurate MNI. All plant fragments are also counted and examined to investigate patterns of fragmentation and possible pre-depositional processing activities. Calculation of the density of plant remains is vital in order to investigate depositional rates and processes. Statistical analysis using various programmes (e.g. CANOCO and CANODRAW cf. Ter Braak & Smilauer 2000) and the spatial and temporal representations of the plant remains with ArcGIS are used for the final study and the contextualisation of the archaeobotanical data.

3.4. Ground stone analysis

13The methodology, database and statistical analytical methods have been designed for and applied to other prehistoric Aegean ground stone assemblages (Tsoraki 2008). All ground stone artefacts, unworked nodules and waste by-products are recorded in detail according to a series of quantitative and qualitative attributes as follows:

  • metrical attributes, such as length, width, thickness, weight, inner and outer diameter and dimensions of usefaces, for both complete and incomplete objects;

  • petrology, including information on raw material characterisation, grain size and textural characteristics (conducted at a macroscopic level);

  • typology, drawing upon the work of specialists in mainland Greece and Crete (Warren 1969; Runnels 1981; Dierckx 1992; Evely 1993; Blitzer 1995), Anatolia and Near East (Wright 1992) and other areas (Adams 2002);

  • technology (reduction sequences, manufacturing techniques and wear, presence of cortical/weathered surfaces);

  • attributes that relate to the use history of these artefacts, such as the number of use-faces, degree and type of wear combined with macroscopic characterisation of wear patterns, modification of working surface and fragmentation patterns;

  • qualitative attributes (colour, degree of finish);

  • contextual attributes (horizontal and stratigraphic position).

14The analysis draws upon the concepts of chaîne opératoire and object biographies to study the life-history of ground stone objects from raw material procurement to manufacture, use, recycling and discard. Emphasis is placed on raw material procurement strategies, the selection and use of rocks for different object categories and on rock properties. The study of raw material variability addresses questions relating to their selection for the production of specific object types. Statistical analysis of collected data combined with detailed temporal and spatial analysis of the material will shed light on the organisation of a wide range of daily and craft activities at Sissi.

4. Some preliminary results

15The analysis of bio-archaeological remains and artefacts and ground stone objects are still at a very early stage of analysis. The preliminary assessment of all the evidence, however, has started delineating the emergence of a very interesting picture for the Bronze Age habitation at Sissi.

16The study of shells is advanced to include about a quarter of the total bulk material excavated between 2007 and 2010. In total, 3295 shells from 229 units were sorted and recorded in detail from all excavated areas, revealing a moderate state of preservation with most specimens being moderately to highly worn probably as a result of postdepositional processes. The majority of specimens belong to 48 different marine species (mostly gastropods, few bivalves, one sea-urchin, and one crab) and only a few to ten terrestrial snail species. While the taxonomic richness of the assemblage is high, the number of individuals is distributed unevenly across all identified species: three congeneric species of limpets that inhabit the same ecological niche and were probably gathered indiscriminately make up 48.3% of the total MNI, while top-shells make up 18.5% of the total MNI. This low diversity shows that shell gathering was specialised and focused mainly on some particular species for food consumption (mainly limpets and top-shells, sporadically urchins and crabs). These are the preferred molluscs for food in Crete during the Bronze Age and Sissi inhabitants seem to follow the same dietary ‘rules’. Metrical data show a great dispersion of sizes (from 5 mm to 43 mm, average 25 mm), but 66.7% of limpets and top-shells are medium and large, an indication that people chose specimens with enough meat for consumption. Murex was also found at Sissi; their preservation state is consistent with purple-dye manufacture but their low number is puzzling.

  • 1 The mollusc is venomous; there are recorded cases of human deaths, resulting from the bite of mollu (...)

17Of particular interest was the concentration of 16 specimens of Conus mediterraneus shells (fig. 9.2) in a corner of Room 5.8 in Zone 5 (Building E), which is one of two molluscs that occur in the Aegean Sea and are considered inedible1. They are usually perforated, ground down and occasionally filled with lead; sometimes they are beachworn, presumably collected dead from beaches. Such finds are usually interpreted as pendants or gaming pieces (Reese 1983). The Conus found in Room 5.8 were neither worked nor beach-worn. Careful analysis of their state of preservation shows that they were gathered alive. Gathering Conus requires diving and good knowledge of their habitats. Therefore, people invested time and effort to acquire them alive. Taking into account the good preservation and the amount of shells, they arguably represent consumption debris. This is also supported by the presence in the same context of considerable amounts of other shells (limpets and top-shells) that are certainly rubbish from several food consumption events and artefacts such as a mortar, a kernos and conical cups, all of which can be related to food preparation and/or consumption. Conus flesh, however, is toxic and it seems highly improbable that it would have been consumed as an ordinary food. Therefore, it is possible that people either extracted and used Conus venom or consumed the toxic flesh as a special food or ingredient. Full study of the assemblage is required in order to draw more reliable conclusions.


18In 2009, scanning by V. Isaakidou and L. Polain of ca. 400 bone bags from contexts of all Zones excavated during the 2008 and 2009 campaigns, provided preliminary taphonomic and species-presence information. Remains of all common farmyard animals were identified. Sheep and goat are the most commonly encountered species throughout the site, followed by pigs. Cattle remains are very rare (identified in only 35 of all scanned bags), as are dog, hare and badger remains. At least one species of deer is represented: antler tines – probably manufacturing debris to judge from the presence of cut marks around the base of one of the specimens, found in 2009 in Room 3.8 (Hilltop Building CD) – were identified in three Palatial deposits. Other body parts have not been identified as yet with certainty, so the status of the species at the site is as yet unknown.

19As well as preliminary scanning some assemblages were analysed in greater detail. First, all animal bones from the Funerary Area were scanned. Animal remains are very rare from tomb deposits. The rock-shelters examined to-date yielded a number of complete hare bones. A number of features (colour, absence of butchery marks, completeness of individual elements and possible presence of near-complete skeletons) suggest that hare bones are recent intrusions (fig. 9.3). Thus, in common with other funerary contexts recently and currently explored, the Sissi cemetery does not provide any evidence either for funerary offerings of meat joints or carcasses, nor of the deposition of remains of funerary meals involving consumption of meat. This observation poses problems on previously postulated practices and resulting inferences for Minoan funerary rituals (cf. Hamilakis 1998).


20Additionally, faunal remains from three deposits from the habitation area excavated during the 2008 field season were studied in detail: an LM I (B?) deposit from Room 2.1 (‘Artisanal Area’), an LM II-LM IIIA1 waste/refuse deposit from Room 5.10 (units #903 and #907) and a few specimens from an LM IIIB late destruction deposit from the Hilltop Building CD (table 9.1). Observations are only preliminary, as individual contexts yielded too few specimens for a statistically reliable sample (cf. Van der Veen & Fieller 1982). A total of 213 specimens (NISP) were recorded representing 238 MaxAU and 216 MinAU. The presence of butchery marks on 12% of the specimens and fragmentation patterns imply that we are dealing essentially with food consumption debris. Their state of preservation was moderate, some 14% of all specimens displaying eroded surfaces. No burning was observed on recorded specimens. Carnivore attrition does not appear to have had a serious effect on the assemblages recorded: it is quite low (6.5% of the total MaxAU). The low incidence of gnawing is interesting in the case of the LM IB deposit in which articulating elements were also identified, both implying rapid deposition and lack of subsequent disturbance. Species representation is comparable with that observed in other Bronze Age east Cretan sites, with emphasis on ovicaprids. The rarity of cattle and their small size are of great interest, as the same characteristics have been observed at FN-EM I Kephala Petra (Isaakidou in prep. A) and Priniatikos Pyrgos, but not in contemporary Knossos (Isaakidou 2004; in prep. B), where average body size is consistently large throughout prehistory. It is not possible at this stage to provide a definite explanation for this pattern, but it may be due to selection of larger individuals at the palatial centre due to better fodder provision, function (i. e., their use as draught animals) or ideological reasons. Further analytical exploration is envisaged to address these questions on a regional scale.









































21The study of archaeobotanical material requires specialised laboratory facilities and equipment, and therefore, only an initial scanning has been carried out to-date on site, since the permit for their transfer is still pending. Yet, the array of plant remains is intriguing, including various cereals, legumes, fruits, nuts and wild species. The 502 soil samples that have been collected to-date across the site indicate spatial variability, with substantial quantities of archaeobotanical material being concentrated in certain rooms and buildings. A prominent example is in Zone 4, where barley, wheat, lentil, vetches and other legumes, as well as various fruits and nuts, contributed to the variety of available foods in the so-called ‘kitchen’ room. Discard of weeds from cereal processing was observed in the same area, the analysis of which will provide significant information on agricultural regimes (e.g. Jones et al. 2010, Bogaard et al. 1999). Another area with a significant presence of plant remains appears to be Zone 2, although it appears that plant foods are mainly remnants of snack consumption, such as olive stones and grape pips. Further analysis of the material is scheduled to provide the detailed level of analysis that is necessary to address issues on the social and economic organisation of the site through time and space. Similarly, only a preliminary assessment of the ground stone assemblage has been conducted so far, which, nevertheless, indicated its substantial size and its variability in terms of both the range of raw materials exploited and object types represented (tools, ornaments, stone vases), highlighting the importance of ground stone artefacts in interpretative reconstructions of past life at Sissi.

22With the completion of the individual studies of each line of evidence, these will be combined together along with the final stratigraphic and contextual information, advancing a more integrated approach to archaeological practice that will serve to highlight the vast potential of moving away from the compartmentalisation of the discipline.

5. Acknowledgements

23The authors would like to thank Quentin Letesson for his help in the environmental team and Stavroula Micha for organising flotation in 2007; Charlotte Langohr for her help in the organisation of the bio-archaeological material in the apothiki and her patience with our requests and queries; Pierre Baulain for all his systematic work on the flotation machine; Laetitia Polain for supervising and processing of all bio-archaeological material; Irene Kritikopoulos for assistance with shell washing. Emily Stevens, Maria Anastasiadou, Marco Pietrovito and all the students of the 2010 campaign worked diligently on heavy residue sorting, flotation and organisation of the archaeobotanical, shell and animal bone materials. Trench supervisors Frank Carpentier, Isabelle Crevecoeur, Maud Devolder, Florence Gaignerot-Driessen, Simon Jusseret and Quentin Letesson kindly provided crucial information and guidance for understanding the archaeological contexts of bio-archaeological datasets. Last but not least, we would like to thank Jan Driessen for entrusting us with the study of the bio-archaeological and ground stone material and for his excellent collaboration.


6. References

▪ Adams 2002 = J. L. Adams, Ground Stone Analysis: a Technological Approach, Utah, 2002.

▪ Bevan 2007 = A. Bevan, Stone Vessels and Values in the Bronze Age Mediterranean, Cambridge, 2007.

▪ Blitzer 1995 = H. Blitzer, Minoan implements and industries, in J. W. Shaw & M. C. Shaw (eds.), Kommos: An Excavation on the South Coast of Crete. Vol. I: The Kommos Region and Houses of the Minoan Town. Part I: The Kommos Region, Ecology, and Minoan Industries, Princeton 1995, 403-535.

▪ Bogaard et al. 1999 = A. Bogaard, C. Palmer, G. Jones, M. Charles & J.G. Hodgson, A FIBS approach to the use of weed ecology for the archaeobotanical recognition of crop rotation regimes, Journal of Archaeological Science 26 (9), 1999, 1211-1224.

▪ Dierckx 1992 = H.M.C. Dierckx, Aspects of Minoan Technology, Culture, and Economy: the Bronze Age Stone Industry of Crete, Ph.D. University of Pennsylvania, 1992.

▪ Evely 1993 = R.D.G. Evely, Minoan Crafts: Tools and Techniques. Volume 1. (SIMA 92), Göteborg, 1993.

▪ French 1971 = D.H. French, An Experiment in Water Sieving, Anatolian Studies 21, 1971 59-64.

▪ Halstead in press = P. Halstead, Faunal remains from FN-EH Nemea Tsoungiza: husbandry, butchery, consumption and discard of animals, in D. Pullen, P. Halstead, A. Karabatsoli, J. Hansen, S. Allen, K. Krattenmaker, Z. Stos-Gale, M. Kayafa & N. Gale (eds), Nemea Valley Archaeological Project I: The Early Bronze Age Village on Tsoungiza Hill, Princeton, in press.

▪ Hamilakis 1998 = Y. Hamilakis, Eating the dead: mortuary feasting and the politics of memory in Aegean Bronze age societies, in K. Branigan (ed.), Cemetery and Society in the Aegean Bronze Age, Sheffield 1998, 115-132.

▪ Isaakidou 2004 = V. Isaakidou, Bones from the Labyrinth: Faunal Evidence for the Management and Consumption of Animals at Neolithic and Bronze Age Knossos, Crete, Ph. D. University of London, 2004.

▪ Isaakidou in prep. A = V. Isaakidou, The Faunal Remains: in Y. Papadatos (ed.), The FN-EMI Settlement at Kephala Petra Siteias, Philadelpheia, in preparation.

▪ Isaakidou in prep. B = V. Isaakidou, Bones from the Labyrinth: Animals and Society at Prehistoric Knossos, London, in preparation.

▪ Jones et al. 2010 = G. Jones, M. Charles, A. Bogaard and J. Hodgson, Crops and weeds: the role of weed functional ecology in the identification of crop husbandry methods, Journal of Archaeological Science 37 (1), 2010, 70-77.

▪ Nikol 1964 = J. Nikol, Special Effectors: Luminous Organs, Chromatophores, Pigments, and Poison Glands, in K. M. Wilbur and C. Yonge (eds), Physiology of Mollusca. Vol. I, New York, 1964, 353-381.

▪ Reese 1983 = D. S. Reese, The Use of Cone Shells in Neolithic and Bronze Age Greece, BSA 78, 1983, 353-357.

▪ Runnels 1981 = C.N. Runnels, A Diachronic Study and Economic Analysis of Millstones from the Argolid, Greece, Ph. D. University Microfilms Michigan, Ann Arbor, 1981.

▪ Ter Braak & Smilauer 2000 = C.J.F. Ter Braak & P. Smilauer, CANOCO Reference Manual and CanoDraw for Windows User’s Guide: Software for Canonical Community Ordination (version 4.5). Microcomputer Power. New York, 2000.

▪ Tsoraki 2008 = C. Tsoraki, Neolithic Society in Northern Greece: The Evidence of Ground Stone Artefacts, Ph. D. University of Sheffield, 2008.

▪ Van der Veen & Fieller 1969 = M. Van der Veen & M. Fieller, Sampling Seeds, Journal of Archaeological Science 9, 1982, 287–98.

▪ Warren 1969 = P.M. Warren, Minoan Stone Vases, Cambridge, 1969.

▪ Wright 1992 = K.I. Wright, A classification system for ground stone tools from the prehistoric Levant, Paléorient 18 (2), 1992, 53-81.


1 The mollusc is venomous; there are recorded cases of human deaths, resulting from the bite of molluscs, but none known from the Aegean. There are many medical uses of this venom nowadays (Nikol 1964).

Table des illustrations

Fichier image/jpeg, 132k
Fichier image/jpeg, 140k
Fichier image/jpeg, 126k

© Presses universitaires de Louvain, 2011

Conditions d’utilisation :

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search