Version classiqueVersion mobile

Multidisciplinary approaches to food and foodways in the medieval Eastern Mediterranean

Sylvie Yona Waksman

Byzantium and beyond

Food production and consumption in the Byzantine Empire in light of the archaeobotanical finds

Anna Elena Reuter


Much concerning food production and consumption is known from historical sources. But research on these subjects requires the archaeobotanical investigation of plant macro-remains, as plants were a major part of human everyday life. The study of plant remains can provide direct evidence for the species used in food production and consumption. The results from more than 50 investigated sites covering the entire Byzantine period (395-1453 AD) are included in an overview of the archaeobotanical evidence for important crops in nutrition and agriculture. The investigation focuses on the plant macro-remains: seeds and fruits of cultivated cereals, pulses, oil and fibre plants as well as cultivated and imported fruits, nuts, vegetables and spices. The aim is to learn which crops were cultivated and which were economically significant in the different regions of the Empire from the Early to the Late Byzantine periods. Cereals in particular provide by far the largest database and the best results. The non-cereal species are in general less well preserved but also provide interesting evidence on production, consumption and trade.

Texte intégral


1Research on food and foodways in relation to consumption, production and trade is an important part of archaeological research. Agriculture and the production of food loomed large in the daily life of past human societies, and large parts of populations were engaged in agriculture and the production of food. The study of food can thus be very useful in the reconstruction of the daily lives of those who lived in the Byzantine Empire. There are several ways to approach the study of Byzantine food and foodways. From written sources such as the Geoponika we know a great deal about crop cultivation and plant use (Geoponika). Analyses of pottery inform us as to how food was transported, stored and consumed, and residue analysis can provide information on the original and secondary contents of amphorae and other vessels (Pecci et al., in this volume). The archaeozoological evidence provides information as to which animals were eaten, which parts were favoured and which animals were used for agricultural labour (Onar, in this volume). From archaeobotanical analysis we obtain direct evidence of the plants that were cultivated and used for human consumption. Archaeobotanical macro-remains consist primarily of the remains of crop plants, in particular cereals, pulses and plants for producing oil and fibre, but also cultivated, imported and gathered fruits, nuts, vegetables and spices. Analysis of these plant remains can provide direct evidence concerning the plants that were consumed by humans and animals, as well as agricultural practices, processing, storing, and the trade of crops and their products.

2This contribution discusses archaeobotanical approaches to the food and foodways of Byzantium. Presented are analysed macro-remains from the most important crops in human consumption that have been recovered from sites dating to the period between 395 and 1453 AD which could be associated with Byzantine territory.

Methodology in archaeobotany: preservation and retrieval of plant remains

3Archaeobotanical macro-remains are recovered during archaeological excavations at best by systematic sampling of different contexts but also through collection by hand of visible accumulations of plant remains, e.g. from storage vessels or amphorae. After sampling, the plant remains may be separated from the soil using different practices depending on the preservation, that is, by flotation, dry sieving or wet sieving. Afterwards the remains are examined in a laboratory using a binocular microscope and are identified based on their morphology, using reference collections and identification manuals. The conditions for preservation of plant remains are various, but in general carbonisation is the most common reason they are preserved on Byzantine sites. Carbonisation of plants occurs upon exposure to high temperatures with a limited supply of oxygen, for example during handling near an oven or fire, while the plant material is drying, during fire events in the storage area or in dumped material. Carbonised remains are inert and thus resist bacterial activity, fungal attack or other decomposition processes, and thus survive in most environments. Plant remains can also be preserved by desiccation in permanently dry environments in which decomposing processes are blocked by the dryness. Desiccation mainly occurs in arid regions such as Egypt. Preservation by waterlogging occurs e.g. in lakes, rivers, wells, latrines or on sites which are permanently covered by seawater such as shipwrecks.

4Which species and which plant remains are preserved depend on various factors, such as the significance of the plant as food, whether it was part of everyday food or used for special purposes. It may depend on whether the plant was processed before consumption or consumed fresh. The conditions of preservation are important too. On sites where the plant remains are preserved by desiccation or waterlogging, a wider range of plants are represented, especially those that were not processed by using heat. But desiccation and waterlogging are rare compared to the most common cause for preservation of plant remains, carbonisation.

5In general cereals provide the largest number of samples in archaeobotanical analyses. This is due to the fact that cereals provided the main part of the daily caloric intake, but also to the way they were processed, often by drying using heat, which could lead to accidental charring. Seeds and fruits which have a high oil or water content may burn to ashes or are highly deformed and thus not identifiable. Fragile plant parts like leaves or flowers often burn to ashes. This is the reason why seeds and fruits of such plants as pulses, oil-producing and fibre-producing plants and fruits and nuts are generally under-represented compared to the cereals.

State of research

6The data used for this overview of the most common crops during the Byzantine period are taken mainly from publications. To identify differences in the plant spectra for different areas of the Byzantine Empire, the results of the analyses are summarized for eight major regions: Italy, the Balkans, Greece, the Crimea, Asia Minor, Syria and Palestine, Egypt and North Africa. In spite of the empire’s long existence, more than 1,000 years, and its vast expanse, especially during the Early Byzantine period, very few sites have been archaeobotanically examined. The majority of studies concentrate on sites dated to the Early Byzantine period (395-642 AD). Few studies concern the Middle Byzantine period (643-1204 AD) and only two analysed sites date to the transition from the Middle to the Late Byzantine period (1205-1453 AD) (fig. 1). A review of the data shows that earlier investigations do not reflect the full inventory of all raw data, and for the most part only vague statements on the frequency of the taxa or presence/absence data have been published. Also the number of analyses in the different regions varies greatly. Unfortunately, the core regions of the Empire, Greece and parts of Asia Minor, are mostly inadequately examined. Also the origins of the plant material vary greatly depending on excavation practice. On some sites only small areas within specific contexts, such as material from one storage facility, have been investigated. On other sites, large-scale excavations were accompanied by systematic sampling strategies, which resulted in a much larger database.

Fig. 1 – Distribution of archaeobotanically investigated sites in the Early, Middle and Late Byzantine periods.

Fig. 1 – Distribution of archaeobotanically investigated sites in the Early, Middle and Late Byzantine periods.

7Only a few investigations from Italy are available for Early and Middle Byzantine sites. Sites of the Early Byzantine period investigated include S. Antonino (Arobba & Murialdo 2001) and Ventimiglia (Arobba 2000) in Liguria. In regard to the Middle Byzantine period, investigations are available for Classe, the port of Ravenna (Augenti et al. 2006), and for Supersano (Grasso 2011) and Apigliano (Grasso & Fiorentino 2009) in southern Italy. The results summarized for the Balkans come from sites in today’s Serbia, Bulgaria and Romania. Comprehensive analyses include those from Serbian Gamzigrad (Medović 2008), and the Bulgarian localities Nicopolis (Buysse 2007) and Iatrus (Neef 2007). For Byzantine Greece, only a few sites have been investigated. As the studies concern sites dating to the Early and Middle as well as the Late Byzantine period, it is not possible to make general statements for the entire region. For the Early Byzantine period, only the investigation at Pyrghouti (Sarpaki 2005) provided extensive material. The analysis of Agios Mamas on the Chalkidiki peninsula dates to the transition from the Middle to the Late Byzantine period (Kroll 1999). The investigation of Chersonesos in the Crimea is the other site which dates to the transition from the Middle to the Late Byzantine period (Pashkevich 2004). Chersonesos had important functions, especially in the Middle and Late Byzantine periods and was a trading crossroads between the Aegean Sea, northern Europe and the steppes of the East (Rabinowitz & Sedikova 2004). Parts of Asia Minor were under Byzantine rule during the entire Byzantine period. However, studies are only available for Early and Middle Byzantine sites. Of special interest are the investigations of the towns Sagalassos (Baeten et al. 2012; Fuller et al. 2012) in south-western Turkey and Amorium in central Anatolia (Giorgi 2012), where Early and Middle Byzantine complexes were analysed. The results from three investigated shipwrecks, the 7th century wreck Yassi Ada (Bryant & Murry 1982) and the Middle Byzantine wrecks Bozburun (Gorham 2000) and Serçe Limanı (Ward 2004), also provide interesting information, especially in regard to trade activities.

8In the south-eastern semi-arid and arid regions of the Byzantine Empire, agriculture was practised in environments different from those around the Mediterranean, especially in areas beyond the rainfall limit where no rain-fed agriculture was possible. In these areas different strategies were necessary for food security, such as irrigation or the combined cultivation of various crops. The wadis, temporary rivers which are fed from rainfall at higher elevations, provided the main source of water supply. The regions of Syria and Palestine consist of a narrow coastal strip as well as highlands and desert, but also fertile areas with climatic conditions favourable to agriculture. Analyses have been carried out on plant remains from the southern thermal baths and the cathedral at Bosra (Willcox 2003) in southern Syria, and from the village of Sergilla (Fornite et al., unpubl. report) situated in northern Syria. In the regions of Israel and Jordan (Palestine) many more studies on different types of sites are available. Of special interest are the investigations of the harbour area of Caesarea Maritima (Ramsay 2010), and of rural settlements such as Humayma (Ramsay 2014) and Khirbet Faris on the Kerak plateau (Hoppe 1999). The climate of Egypt is characterized by severe drought. With the exception of oases, the Nile Valley and the narrow coastal strips along the Mediterranean and the Red Sea, a desert climate prevails. The Eastern Desert is characterized by rocky mountains which are crossed by wadis. In the regions dominated by desert climate, agriculture was not possible. Most of the sites in Egypt are located within fertile areas of the Nile Valley. With the annual floods, fertile Nile silt was deposited in the flooded areas and provided a basis for successful agriculture. Analysis of plant remains from the Kom el‑Nana monastery, located on the eastern bank of the Nile, has provided comprehensive evidence on crop plants (Smith 2003). The port city of Berenike and the nearby rural settlement of Shenshef (Cappers 2006) as well as the fortification at Abu Sha’ar (El Hadidi & Amer 1996), located on the Red Sea coast, have provided extensive evidence on crop plants and trade activities, especially Berenike. Analyses of plant remains from North Africa, from Libya, Algeria and Tunisia, are summarized together, as the natural conditions in these countries are similar; the narrow coastal strip is influenced by Mediterranean climate and the adjoining desert regions are characterized by aridity. The eastern region of Libya, the Cyrenaica, was part of the Byzantine Empire as early as the division of the Roman Empire in 395 AD. The western regions of Libya, Tunisia and Algeria were reconquered under Emperor Justinian I after the submission of the Vandals in 533-534 AD. In Libya, Ghirza and three other localities were examined as part of the Libyan Valleys Survey (van der Veen et al. 1996), and extensive studies have been conducted at Carthage (van Zeist et al. 2001) and Leptiminus (Smith 2001), both located in Tunisia.


9As already mentioned, cereals provided by far the largest database, for different reasons. All in all, 13 cereal species have been determined for the entire Byzantine period – bread wheat (Triticum aestivum), durum wheat (Triticum durum), hulled barley (Hordeum vulgare var. vulgare), naked barley (Hordeum vulgare var. nudum), two-rowed barley (Hordeum distichum), spelt (Triticum spelta), rye (Secale cereale), broomcorn millet (Panicum miliaceum), foxtail millet (Setaria italica), sorghum millet (Sorghum bicolor), rice (Oryza sativa), einkorn (Triticum monococcum), emmer (Triticum dicoccum). Most sites show the use of various species, but in general one or two species could be identified as the main crop. These other species did not play a major role as food plants, some could have been grown for special purposes and others would have been weeds growing among the crops.

10Throughout the Byzantine period, the free-threshing wheats, bread wheat and durum wheat, and barley appear to have been the most important species. The free-threshing wheat species were especially suited for baking the most popular white bread. Therefore it is not surprising that they were grown in all regions of the empire. In the regions of Italy and Greece in particular, free-threshing wheat species dominate the cereal spectra. In Egypt also a high percentage of free-threshing wheat was grown. However, bread wheat (fig. 2) is a particularly demanding crop and its cultivation requires certain climatic conditions and a good water supply. Temperature fluctuations and poor soil conditions can lead to crop failure. Regions in which a mild, humid Mediterranean climate prevails, like most of the coastal areas, especially in the north-western regions, were the most suitable for the cultivation of bread wheat. In Egypt, characterized by an arid desert climate, the cultivation of durum wheat is most in evidence, as it is more resistant to drought and high temperatures than bread wheat. Like durum wheat, barley is also more resistant to drought and high temperatures and can be cultivated on poorer soils. Thus it is not surprising that it has been identified as the main agricultural product, especially in the eastern and southern semi-arid and arid regions, particularly North Africa, Syria and Palestine as well as in some sites in Asia Minor. But also in Egypt, Greece, the Balkans and Italy it played a major role in agriculture and nutrition. In particular six-rowed hulled barley was cultivated, two-rowed barley and naked barley being rare. Normally barley was not used for baking bread, but the grains could have been ground to flour and baked to produce flatbread or rusk, and it could have been used as an ingredient in soups as pearl barley. Aside from human consumption it was also used as animal feed.

Fig. 2 – Bread wheat (Triticum aestivum).

Fig. 2 – Bread wheat (Triticum aestivum).

11Besides the free-threshing wheats and barley, in some regions rye and millet were also cultivated. Especially in the Balkans, high proportions of rye (fig. 3) and broomcorn millet were found. The prevailing continental climate is favourable for the cultivation of rye, which can be grown in areas which are cooler and have poorer soil conditions. Millet can also be cultivated on poorer, dryer and very sandy soils and is favoured because of its very short vegetation period of only 60‑90 days from sowing to harvest. The cereal spectrum in the Balkans in the Early Byzantine period is exceptional in Byzantine lands because in most of the analyses more than two cereal species could be identified as economically significant. This situation would have been due to the insecure conditions of war and conflict prevailing in that region at the time, leading to coping strategies in agriculture. The cereal spectrum of Asia Minor also shows a greater variety of cultivated species. Unlike the Balkans, this greater variety is mainly due to the local domination of different species on different sites. Besides barley and free-threshing wheat, broomcorn millet and spelt also appear to have been important. Broomcorn millet occurs in smaller amounts, e.g. in Amorium (Giorgi 2012) and Kilise Tepe (Bending & Colledge 2007); only in Sagalassos were significant percentages identified (Fuller et al. 2012). The high percentages of hulled wheat are due to the rich finds of spelt in Kilise Tepe (Bending & Colledge 2007, p. 593). This strong dominance is limited to that site, so it must be a local phenomenon. The other hulled wheat species, emmer and einkorn, generally occurred in lower numbers in all regions. During the Roman period they had already lost their importance in agriculture, and therefore can be interpreted as weeds in the cereal crops. Of special interest are the finds of sorghum and rice in Egypt. The advent of sorghum domestication in the 1st or 2nd millennium BC continues to be discussed (Cappers 2006, p. 126). Its cultivation and spread into the Mediterranean and the Near East are often thought to have taken place starting in Islamic times (Zohary & Hopf 1994, p. 84). The finds from Berenike and neighbouring Shenshef on the Red Sea coast as well as those from the Kom el‑Nana monastery (Smith 2003) on the eastern bank of the Nile are proof that it was used during the Byzantine period. There is also strong evidence for the cultivation of domestic sorghum at Quasr-Ibrim during the transition from the Roman to the Meroitic period onward (Rowley-Conwy 1991, p. 205). While sorghum, which is native to Africa, could have been cultivated in the Nile Valley or in the vicinity of the sites, rice is native to India and must have been imported from there. Evidence for it is limited to Berenike and Shenshef and is connected to an Indian trading community.

Fig. 3 – Rye (Secale cereale).

Fig. 3 – Rye (Secale cereale).

12During the Middle Byzantine period the free-threshing wheat species and barley remained the most important cereals. Compared to the Early Byzantine period, only a few studies from Italy and Asia Minor are available and therefore it is not possible to make general statements. However, some developments are apparent. It appears that the main crop changed between the Early and the Middle Byzantine periods in Italy. During the Early Byzantine period it appears that free-threshing wheat was the most common cereal species, while during the Middle Byzantine period barley dominated the spectrum and free-threshing wheat was less frequent. However, this apparent development may reflect the low data available rather than the actual situation. All other cereal species also remained subdominant during the Middle Byzantine period. In Asia Minor the analyses of plant remains from shipwrecks and sediments in the Theodosian harbour of Constantinople (Oybak-Dönmez 2010) as well as from the shipwrecks Bozburun (Gorham 2000) and Serçe Limanı (Ward 2004) provide little evidence for cereals, as most specimens could not be identified to species level. Most of the cereal finds identified from Middle Byzantine contexts were at Amorium (Giorgi 2012) and Sagalassos (Fuller et al. 2012), from a storage vessel in Beycesultan (Helbaek 1961) and from the Byzantine garrison at Gritille (Miller 1998). Compared to the results for the Early Byzantine period the spectrum is less diverse. Hulled barley was dominant in the analyses of the material from Amorium (Giorgi 2012) and important in Sagalassos (Fuller et al. 2012) too. Free-threshing wheat was found in smaller quantities, but appears to have been important, particularly for human consumption, on both sites (Giorgi 2012, p. 396f; Fuller et al. 2012, p. 162). The increasing percentages of broomcorn millet in Sagalassos are interesting, as are the contents of a storage vessel found in Beycesultan. The latter provides a completely different combination of cereals, being almost exclusively filled with grains of rye and a few grains of hulled barley and einkorn (Helbaek 1961). This find is important to show that rye was also cultivated in the Eastern Mediterranean regions, whereas other evidence indicates that earlier only in the Balkans was rye economically significant. The finds from Agios Mamas in Greece, which date to the transition from the Middle to the Late Byzantine period, suggest the increasing cultivation of rye in the Eastern Mediterranean. Bronze Age Olynthus provided some samples of crop material which proved to be Byzantine. During the Byzantine period a little chapel with a burial ground was located on the small mound. Since the Byzantine strata were not the focus of the archaeological investigation, only a relatively broad dating to the 12th-14th centuries is available (Kroll 1999, p. 474). In spite of their small size, the samples are significant and provide insight into the development of the crop spectrum. Also in the investigation of Middle to Late Byzantine contexts in Chersonesos a larger quantity of rye was found, besides bread wheat, barley and broomcorn millet. Rye is generally considered to be a typical grain of medieval central and northern Europe and is not associated with the Eastern Mediterranean. Therefore the finds of rye may be considered as important indicators for the spread of rye cultivation into the Eastern Mediterranean.


13Besides the cereals, pulses were a main source of vegetable protein in human consumption, and could also have been used as green manure in crop rotation systems because of their ability to fix nitrogen in the soil. Cultivation of pulses could have been carried out in gardens or in fields. Especially during the Early Byzantine period a wide range of species is represented in the analyses. The most common cultivated species were faba bean (Vicia faba), garden pea (Pisum sativum), lentil (Lens culinaris), bitter vetch (Vicia ervilia), common vetch (Vicia sativa), grass pea (Lathyrus sativus), chickpea (Cicer arientum) and white lupin (Lupinus albus). In general, pulses are not as well preserved as cereals and are less present in the plant spectra. In the Early Byzantine period particularly, lentil, faba bean and garden pea appear to have been important. Lentil appears to have been predominantly cultivated in Italy, Asia Minor, Syria and Palestine, Egypt and North Africa. It is one of the oldest cultivated legume species in Mediterranean agriculture and was typically grown next to wheat and barley. It provides smaller yields, but was especially cultivated because of its high nutritional value and its good taste. The spectrum for Italy is dominated by the analysis of material from Ventimiglia, where besides lentil there were also numerous finds of common vetch (Arobba 2000, p. 203). Common vetch, like bitter vetch, is generally associated with use as animal feed or as a food for the poorer classes. Due to the toxic bitter compounds which had to be removed before consumption by cooking or soaking in water, these plants appear to have been unattractive for human consumption. Nevertheless, both types are characteristic of Mediterranean agriculture and appear in all regions. The steady and numerous occurrences of the two species indicate that they were used, especially if only a few or no other legumes were detected. The same applies to white lupin, which was cultivated mainly in Egypt. Although lentil could be identified as the most common species in Asia Minor, evidence for the cultivation of pulses is in general very low and the spectra appear to have been influenced by the local environment of the sites. Similar observations can be made for the regions of Syria-Palestine and North Africa. In Syria-Palestine there is much evidence for the cultivation of pulses, especially on the sites of Sergilla (Fornite et al., unpubl. report), Bosra (Willcox 2003), Caesarea Maritima (Ramsay 2008) and Khirbet Faris (Hoppe 1999). In Bosra (Willcox 2003, p. 183) and Khirbet Faris (Hoppe 1999, p. 128), high concentrations of garden pea were discovered. The analyses from Caesarea Maritima produced much evidence for lentil, garden pea, bitter vetch and faba bean (Ramsay 2008, p. 203). In North Africa only the analyses from Carthage show a broader spectrum of cultivated pulses; besides lentil, garden pea is frequently found, and bitter vetch and chickpea were also grown (van Zeist et al. 2001, p. 28). The investigations for the Balkans provide a wide spectrum of cultivated pulses. Some species, such as faba bean, lentil and bitter vetch, occurred on all sites. However, the number of recorded species varies greatly within the individual analyses. Thus, for example, faba bean was frequently identified at Nicopolis (Buysse 2007, p. 278), in Abritus lentil dominates the spectrum (Popova & Marinova 2000, p. 51) and in Iatrus bitter vetch was the most common species (Neef 2007, p. 419). In addition, garden pea and grass pea appeared in almost all analyses. Faba bean and garden pea could have been cultivated in warm Mediterranean climates but also in cooler, temperate conditions, since both species prefer moist habitats. For the Early Byzantine period in Greece, evidence for the cultivation of pulses is only available from the analysis of material from Phyrgouthi. Unusually, there is stronger evidence here for the cultivation of pulses compared to cereals, especially grass pea/red vetchling (Lathyrus sativus/cicera), Spanish vetchling (Lathyrus clymenum), bitter vetch, lupin (Lupinus sp.) and lentil, which are suitable for cultivation in dryer environments. The large number and the diversity of species are indicators of the importance of pulses in agriculture and nutrition. During the Middle Byzantine period, lentil remained the most common species. It appears that there was an increasing cultivation of chickpea from the Early to the Middle and Late Byzantine periods. Besides the abundant material from Classe (Augenti et al. 2006, p. 157), a significant amount was found at Agios Mamas (Kroll 1999, p. 477). In Asia Minor for the Early Byzantine period the evidence for pulse cultivation is low. From Sagalassos (Fuller et al. 2012, p. 162) and Gritille (Miller 1998, p. 215) a few finds indicate the cultivation of lentil, and bitter vetch was found at Amorium (Giorgi 2012).

Plants exploited for oil and fiber

14In almost all regions the plants exploited for their oil or fiber are very difficult to identify. This is because the seeds were often highly deformed by heat due to their high oil content or burnt to ashes. Typical plants are olive (Olea europaea), linseed/flax (Linum usitatissimum), hemp (Cannabis sativa), poppy (Papaver somniferum) and camelina (Camelina sativa). Except for olive, there is little evidence for the cultivation of other species in Early Byzantine Italy, the Balkans, Egypt and North Africa. In the Middle and Late Byzantine periods, only evidence for the use of olive has been found. The multifunctional linseed/flax appears to have been the most common species after olive. The seeds could have been used for oil production and the plant fibers derived from the vegetative parts used for the production of textiles or ropes for example. In the Balkans and North Africa there is little evidence of its use. Only in Kom el‑Nana (Egypt) was a large concentration of seeds found (Smith 2003). Poppy is also very rare, there being a few finds from Kom el‑Nana (Smith 2003) and a larger accumulation of seeds at Carthage (van Zeist et al. 2001, p. 30). Remains of camelina were identified only at Gamzigrad (Medović 2008) in Serbia. Hemp is represented by a single find from Ventimiglia (Arobba 2000, p. 208) in Italy.

15Besides the typical oil-producing plants, oil could also be obtained from other plants such as sesame (Sesamum indicum) and safflower (Carthamus tinctorius). Rare evidence for sesame has been found, particularly at Berenike (Cappers 2006, p. 125) and Nicopolis (Buysse 2007, p. 292). Local cultivation of sesame is excluded for both sites. In the case of the sesame from Berenike, an import from the Nile Valley is assumed (Cappers 2006, p. 125). Sesame could have been used not only for the production of oil, but also for seasoning. Safflower seeds may be used to produce oil, and pigments for dying were extracted from the flower heads.

16Of the oil-producing plants, olive (fig. 4) appears to have been the most important. The production of olive oil already had a long tradition during Byzantine times in the Mediterranean area. Because of the habitat requirements of the plant, cultivation was limited to areas with a Mediterranean climate, particularly the coastal regions. Areas where olives could not be grown due to climatic conditions, such as the Balkans, were supplied with olive oil through trade. Known areas for olive cultivation and production of olive oil were in the vicinity of Carthage and Athens. In Greece large accumulations of olive stones were found in Pyrgouthi (Sarpaki 2005, p. 319) and in the Agora of Athens (Margaritis 2006, p. 26). Due to the degree of fragmentation, they were identified as by-products of olive oil production. Such by-products were often used secondarily, e.g., as animal feed, alternative fuels, fertilizers or as temper in mortar.

Fig. 4 – Olive (Olea europaea).

Fig. 4 – Olive (Olea europaea).

Garden, gathered and imported plants

17Besides cereals, pulses, and plants for oil and fiber production, there is evidence for a wide range of spices, vegetables, fruits and nuts. Most of the species were cultivated in gardens, collected in the vicinity of the sites or imported from farther away. Many of the plants had more than one use, e.g. some were used for seasoning, others eaten fresh. Fruits, nuts, vegetables and spices are often less in evidence than cereals or other crops, due to the low preservation of the fragile parts used, such as leaves, flowers or roots. These parts are usually not preserved, as they burn to ashes rather than becoming carbonized. Fresh plant parts consumed without preparation by heat would also not lead to preservation. However, there is evidence for a wide range of more than 100 species. Since it is not possible to include all species at this point, only the most significant species will be discussed.

18Of particular importance were grape (Vitis vinifera), fig (Ficus carica), date (Phoenix dactylifera) and olive. During the Byzantine period their cultivation in Mediterranean regions had already enjoyed a long tradition, and they continue to be important agricultural products. The cultivation of other species such as plum (Prunus domestica), sweet cherry (Prunus avium), apple (Malus domestica) and pear (Pyrus domestica) had been introduced by the Greeks and the Romans. There is wide evidence for fig, peach (Prunus persica) and grape in particular. Grapes were consumed fresh, and dried as raisins they could be transported over long distances. The production of wine was particularly important, illustrated by the Geoponika, which in numerous sections deals with wine, its production, refining, seasoning and preservation (Geoponika). Unlike the olive tree, the grape vine (fig. 5) is tolerant of cooler temperatures and could therefore have been grown in almost every part of the Byzantine Empire. Therefore it is not surprising that evidence for grapes are found in almost all analyses of material from sites of the Early Byzantine period to the Late Byzantine period. Wine production is often difficult to establish, without proof of wine presses or large accumulations of pressing residues. By-products of wine production from pressing, as well as those from olive oil production, were probably often used for animal feed or fuel. Although archaeobotanical evidence for wine production is scarce, some analyses provide interesting insights. On the site of Supersano in southern Italy, in a layer within a well, a high concentration of waterlogged grape vine remains was found, dating to the 8th century. Whole grapes, grape skins and stalks were preserved, as well as grape seeds. These remains of grape vine and the large volume of the accumulation strongly indicate wine production. After the grapes were pressed, the waste consisting of skins, stalks and seeds was dumped in the well (Arthur et al. 2012, p. 450f). Direct evidence for the trade of wine and other products produced from grapes are found in analyses of finds from shipwrecks. The Bozburun wreck (SW Turkey), dating to the 9th century AD, contained an impressive cargo of some 1,000 amphorae which had contained mainly wine, and also olives, and grapes or raisins (Gorham 2000, pp. 133‑139). Some of the amphorae were still sealed and contained a brownish red-to-purple liquid. Some amphorae possibly contained spiced wine, others grape seeds, spices, flowers or fish (Gorham 2000, p. 153). In addition to grapes, figs and dates were traded in a dried form.

Fig. 5 – Common grape vine (Vitis vinifera ssp. vinifera).

Fig. 5 – Common grape vine (Vitis vinifera ssp. vinifera).

19Other fruits were: apricot (Prunus armeniaca), plum, sweet cherry, pomegranate (Punica granatum), raspberry (Rubus ideaus) and blackberry (Rubus fruticosus). Nuts such as almond (Prunus amygdalus), walnut (Juglans regia), hazelnut (Corylus avellana) and pine (Pinus pinea) were also grown, collected and traded over long distances. There is evidence that walnuts and hazelnuts were imported in North Africa and Egypt. In Italy, the Balkans, Asia Minor, Syria and Palestine, walnuts and hazelnuts were locally grown or collected near the sites. Almonds were typical exports from the Mediterranean area and have been found at Early and Middle Byzantine sites in all regions. Pine has been found particularly in the commercial and port cities of Carthage (North Africa), Berenike (Egypt) and Caesarea Maritima (Syria/Palestine). At Caesarea Maritima pine could have been cultivated in the vicinity of the site, but not at Berenike and Carthage, to which pine nuts would have been imported. Therefore, it is believed that the finds from the harbour area of Caesarea Maritima were originally intended for export (Ramsay 2010, p. 379).

20The widest range of spices has been found on sites in North Africa, Egypt, Syria and Palestine, principally at the port cities of Carthage (van Zeist et al. 2001), Berenike (Cappers 2006) and Caesarea Maritima (Ramsay 2010). For Berenike, coriander (Coriandrum sativum), dill (Anethum graveolens) and black pepper (Piper nigrum) have been determined. Black pepper is native to India and was among the imported finds at Berenike. The large number of remains indicates the importance of the trade in pepper, in particular the discovery of a dolium with a capacity of about 31 litres which was filled 42% with pepper corns. The desiccated remains weighed 7.5 kg. The dolium is also of Indian origin, and thus is likely to be the original vessel used for transport (Cappers 2006, p. 114). Unlike the imports, coriander and dill could have been grown locally in irrigated gardens. Dill in particular grows well in a dry environment. The leaves were used dried or fresh. Coriander and dill appear to have been the most widespread species of herbs in the Early Byzantine period.


21This overview of archaeobotanical analyses of plant remains from Byzantine sites demonstrates the wide range of plant species used in food production, consumption and trade. For some regions in particular, such as the Balkans during the Early Byzantine period, there is evidence which has led to different assumptions concerning nutrition and agriculture. The results also highlight the lack of information for the Middle and Late Byzantine periods and for the core regions of Greece and Asia Minor, thus urging for future research. Because of the lack of analyses for those periods and regions, it is possible only to provide known information and point out what appear to have been tendencies, such as the apparent spread of rye cultivation into the Eastern Mediterranean area. It must be admitted that the results of this overview reflect the state of research rather than the situation during Byzantine times. It is striking that there is a lack of interdisciplinary investigations on sites of historical periods in comparison to prehistoric periods, to which interdisciplinary research is more frequently applied. It would appear that written sources are considered to be sufficient to reconstruct the past. There is no doubt that historical sources provide important information, but they do not reflect all the information that we may obtain. Interdisciplinary research is necessary in the investigations of historical sites to obtain the fullest possible range of information. The POMEDOR conference well illustrates how enriching exchange between specialists and disciplines can be, and how it can provide a more complete picture of past societies.


Arobba 2000: Arobba D., “Macroresti botanici rinvenuti nei livelli tardoantichi e medievali del Battistero della cattedrale di Ventimiglia”, Rivista di studi liguri 66, 2000, pp. 197‑212.

Arobba & Murialdo 2001: Arobba D., Murialdo G., “Le analisi palinologiche e paleocarpologiche”, in T. Mannoni, G. Murialdo (ed.), S. Antonino, un insediamento fortificato nella Liguria bizantina, Collezione di monografie preistoriche ed archeologiche 12, Bordighera, 2001, pp. 627‑638.

Arthur et al. 2012: Arthur P., Fiorentino G., Grasso A.M., “Roads to recovery: An investigation of early medieval agrarian strategies in Byzantine Italy in and around the 8th century”, Antiquity 86, 2012, pp. 444‑455.

Augenti et al. 2006: Augenti A., Bondi M., Carra M., Cirelli E., Malaguti C., Rizzi M., “Indagini archeologiche a Classe (scavi 2004): Primi risultati sulle fasi di età Altomedievale e dati archeobotanici”, in R. Francovich, M. Valenti (ed.), IV Congresso nazionale di archeologia medievale (San Galgano, Chiusdino [SI], 26‑30 settembre 2006), Florence, 2006, pp. 124‑131.

Baeten et al. 2012: Baeten J., Marinova E., de Laet V., Degryse P., de Vos D., “Faecal biomarker and archaeobotanical analyses of sediments from a public latrine shed new light in ruralisation in Sagalassos, Turkey”, Journal of archaeological science 39, 2012, pp. 1143‑1159.

Bending & Colledge 2007: Bending J., Colledge S., “The archaeobotanical assemblages”, in N. Postgate, D. Thomas (ed.), Excavations at Kilise Tepe (1994-1998): From Bronze Age to Byzantine in western Cilicia, McDonald Institute monographs 30, Cambridge-London, 2007, pp. 583‑595 & Appendix I.

Bryant & Murry 1982: Bryant V.M., Murry R.E., “Preliminary analysis of amphora contents”, in G.F. Bass, F.H. van Doorninck (ed.), Yassı Ada, vol. I, A seventh century Byzantine shipwreck, College Station, 1982, pp. 327‑331.

Buysse 2007: Buysse J.L., “The botanical remains”, in A. Poulter (ed.), Nicopolis ad Istrum III: A late Roman and early Byzantine city: The finds and the biological remains, Oxford, 2007, pp. 260‑292.

Cappers 2006: Cappers R.T.J., Roman foodprints at Berenike: Archaeobotanical evidence of subsistance and trade in the Eastern Desert of Egypt, Berenike reports 6, Los Angeles, 2006.

Fornite et al., unpubl. report: Fornite S., Pessin H., Roitel V., Willcox G., Les restes végétaux carbonisés de Sergilla. Rapport archéobotanique intermédiaire, report (unpublished).

Fuller et al. 2012: Fuller B.T., de Cupere B., Marinova E., van Neer W., Waelkens M., Richards M.P., “Isotopic reconstruction of human diet and animal husbandry practices during the Classical-Hellenistic, Imperial, and Byzantine periods at Sagalassos, Turkey”, American journal of physical anthropology 149/2, 2012, pp. 157‑171.

Geoponika: Dalby A., Geoponika: farm work. A modern translation of the Roman and Byzantine farming handbook, Totnes, 2011.

Giorgi 2012: Giorgi J., “The plant remains”, in C.S. Lightfoot, E.A. Ivison (ed.), Amorium reports, 3, The lower city enclosure: Finds reports and technical studies, Istanbul, 2012, pp. 395‑418.

Gorham 2000: Gorham L.D., The archaeobotany of the Bozburun Byzantine shipwreck, Ann Arbor, 2000.

Grasso 2011: Grasso A.M., “Analisi archeobotaniche a Supersano (LE). Una comunità autosufficiente?”, Post-classical archaeologies 1, 2011, pp. 297‑308.

Grasso & Fiorentino 2009: Grasso A.M., Fiorentino G., “L’ambiente vegetale. Risultati delle nuove analisi archeobotaniche”, in P. Arthur, B. Bruno (ed.), Apigliano. Un villaggio bizantino e medievale in terra d’Otranto. L’ambiente, il villaggio, la popolazione, Galatina, 2009, pp. 53‑56.

El Hadidi & Amer 1996: El Hadidi M.N., Amer W.M., “The palaeobotany of Abu Sha’ar site (AD 400-700) Red Sea coast, Egypt”, Taeckholmia 16, 1996, pp. 31‑44.

Helbaek 1961: Helbaek H., “Late Bronze Age and Byzantine crops at Beycesultan in Anatolia”, Anatolian studies 11, 1961, pp. 77‑97.

Hoppe 1999: Hoppe C., A thousand years of farming: Agricultural practices from the Byzantine to Early Ottoman period at Khirbet Faris, the Kerak plateau, Jordan, PhD, Sheffield University, 1999 (unpublished).

Kroll 1999: Kroll H., “Byzantinischer Roggen von Agios Mamas, Chalkidiki”, Byzantinische Zeitschrift 92, 1999, pp. 474‑478.

Margaritis 2006: Margaritis E., “Archaeobotanical remains from the Agora’s Byzantine contexts: Wiener lab report”, Akoue, Newsletter of the American School of Classical Studies at Athens 56, 2006, p. 26.

Medović 2008: Medović A., “Gamzigradski ratari – dva koraka napred, jedan korak nazad”, Rad Muzeja Vojvodine 50, 2008, pp. 151‑173.

Miller 1998: Miller N.F., “Patterns of agriculture and land use at medieval Gritille”, in S. Redford (ed.), The archaeology of the frontier in the medieval Near East: Excavations at Gritille, Turkey, Archaeological Institute of America monographs 3, Philadelphia, 1998, pp. 211‑252.

Neef 2007: Neef R., “Archäobotanische Untersuchungen im spätantiken Iatrus/Krivina (Grabungskampagnen 1992-2000)”, in G. von Bülow, N. Benecke (ed.), Iatrus-Krivina: Spätantike Befestigung und frühmittelalterliche Siedlung an der unteren Donau, vol. 6, Mainz, 2007, pp. 415‑445.

Onar, in this volume: Onar V., “Animals in food consumption during the Byzantine period in light of the Yenikapı metro and Marmaray excavations, Istanbul”, in Waksman 2020, pp. 331‑342.

Oybak-Dönmez 2010: Oybak-Dönmez E., “Archaeobotanical studies at the Marmary and metro excavations in Istanbul”, in U. Kocabaş (ed.), Istanbul Archaeological Museums: Proceedings of the 1st Symposium on Marmaray-Metro Salvage Excavations (5th-6th May 2008), Istanbul, 2010, pp. 233‑248.

Pashkevich 2004: Pashkevich G.A., “Paleoethnobotanical investigations at Chersonesos”, in The study of ancient territories: Chersonesos & Metaponto, 2004 annual report, Austin, 2004, pp. 19‑26.

Pecci et al., in this volume: Pecci A., Garnier N., Waksman S.Y., “Residue analysis of medieval amphorae from the Eastern Mediterranean”, in Waksman 2020, pp. 417‑428.

Popova & Marinova 2000: Popova T., Marinova E., “Archaeobotanical and anthracological analysis of the Roman and Early Byzantine castle abritus in north-eastern Bulgaria: Some palaeoethnobotanical and environmental aspects”, Archaeologia Bulgarica 4/2, 2000, pp. 49‑58.

Rabinowitz & Sedikova 2004: Rabinowitz A., Sedikova L., “Excavations in the south region of Chersonesos”, in The study of ancient territories: Chersonesos & Metaponto, 2004 annual report, Austin, 2004, pp. 5‑12.

Ramsay 2008: Ramsay J.H., “Archaeobotanical remains from Caesarea: The 1997 and 1998 seasons”, in K.G. Holum (ed.), Caesarea reports and studies: Excavations 1995-2007 within the old city and the ancient harbor, BAR International series 1748, Oxford, 2008, pp. 201‑208.

Ramsay 2010: Ramsay J.H., “Trade or trash: An examination of the archaeobotanical remains from the Byzantine harbour at Caesarea Maritima, Israel”, International journal of nautical archaeology 39/2, 2010, pp. 376‑382.

Ramsay 2014: Ramsay J., “Plant remains”, in J.P. Oleson, R. Schick (ed.), Humayma excavation project, 2, Nabatean campground and necropolis, Byzantine churches, and Early Islamic domestic structures, Boston, 2014, pp. 353‑360.

Rowly-Conwy 1991: Rowly-Conwy P., “Sorghum from Quasr-Ibrim, Egyptian Nubia, c. 800 BC-AD 1811: A preliminary study”, in J.M. Renfrew (ed.), New light on early farming: Recent developements in palaeoethnobotany, Edinburgh, 1991, pp. 191‑212.

Sarpaki 2005: Sarpaki A., “The archaeobotanical material from the site of Pyrgouthi in the Berbati Valley: The seeds”, in J. Hjohlman (ed.), Pyrgouthi: A rural site in the Berbati Valley from the early Iron Age to Late Antiquity: Excavations by the Swedish Institute at Athens, 1995 and 1997, Stockholm, 2005, pp. 313‑341.

Smith 2001: Smith W., “Environmental sampling (1990-1994)”, in L.M. Stirling, D.J Mattingly, N.B Lazreg (ed.), Leptiminus (Lamta): Report no. 2: The east baths, cemeteries, kilns, Venus mosaic, site museum, and other studies, Journal of Roman archaeology supplementary series 41, Portsmouth R.I., 2001, pp. 420‑441.

Smith 2003: Smith W., Archaeobotanical investigations of agriculture at Late Antique Kom el‑Nana (Tell el‑Amara), London, 2003.

Van der Veen et al. 1996: van der Veen M., Grant A., Barker G., “Romano-Lybian agriculture: Crops and animals”, in G. Barker (ed.), Farming the desert, vol. 1, Synthesis, London, 1996, pp. 227‑263.

Van Zeist et al. 2001: van Zeist W., Bottema S., van der Veen M., Diet and vegetation at ancient Carthage: The archaeobotanical evidence, Groningen, 2001.

Ward 2004: Ward Ch., “Plant remains”, in G.F. Bass, Sh.D. Matthews, J.R. Steffy, F.H. van Doorninck Jr. (ed.), Serçe Limanı: An 11th-century shipwreck, vol. 1, The ship and its anchorage, crew, and passengers, College Station, 2004, pp. 493‑511.

Willcox 2003: Willcox G.H., “L’économie végétale à Bosra et à Si. Résultats d’analyse de restes végétaux carbonisés des périodes romaine, byzantine et islamique”, in J. Dentzer-Feydy, J.‑M. Dentzer, P.‑M. Blanc (ed.), Hauran II. Les installations de Sīʿ8: du sanctuaire à l’établissement viticole, vol. I, Bibliothèque archéologique et historique 164, Beirut, 2003, pp. 177‑195.

Zohary & Hopf 1994: Zohary D., Hopf M., Domestication of plants in the Old World: The origin and spread of cultivated plants in West Asia, Europe and the Nile Valley, Oxford, 1994.

Table des illustrations

Titre Fig. 1 – Distribution of archaeobotanically investigated sites in the Early, Middle and Late Byzantine periods.
Fichier image/jpeg, 1,0M
Titre Fig. 2 – Bread wheat (Triticum aestivum).
Fichier image/jpeg, 252k
Titre Fig. 3 – Rye (Secale cereale).
Fichier image/jpeg, 299k
Titre Fig. 4 – Olive (Olea europaea).
Fichier image/jpeg, 267k
Titre Fig. 5 – Common grape vine (Vitis vinifera ssp. vinifera).
Fichier image/jpeg, 265k


Römisch-Germanisches Zentralmuseum Mainz, Leibniz-Forschungsinstitut für Archäologie & University of Kiel, Institute of Prehistoric and Protohistoric Archaeology, Johanna-Mestorf-Straße 2‑6, 24118 Kiel, Germany,

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search