Précédent Suivant

Roman Life in the Eastern Desert of Egypt: Food, Imperial Power and Geopolitics


Texte intégral

Introduction

1The Eastern Desert of Egypt, located between the Nile and the Red Sea, has a mean annual rainfall of just 5mm and is today classified as hyper-arid, and these arid conditions were already in place well before the start of the Roman period (Zahran and Willis 1992). Consequently, vegetation is sparse, except in some well-watered wâdis, and the region has seen neither agriculture nor permanent occupation during the last 10,000 years. The Eastern Desert is, however, rich in precious resources, ranging from gold and emeralds used in jewellery and other valuable objects, to high quality stone used for building, for statuary, for baths, basins and sarcophagi, employed largely in imperial prestige projects. Additionally, one of the main wâdis, the Wâdi al-Hammâmât, offers an accessible way through the mountains from the Nile to the Red Sea coast, and this route has been used at least since pharaonic times; in the Roman period it formed the main route to the port of Myos Hormos (Sidebotham et al. 2008). The scarcity of water, the extreme heat and the lack of locally available foodstuffs make for a harsh environment and travel in or through the region was and is difficult and unforgiving. Nevertheless, the region was a hive of activity during the early Roman period, with the development of two major ports for the trade with India (already established during the Ptolemaic period), several quarries and mines, as well as roads and service stations to supply these. The inhabitants of these sites came from both Egypt and other parts of the Roman Empire, and included soldiers and their superiors, specialist and unskilled workmen, crafts people, passing merchants, wives, prostitutes, and possibly some slaves (e.g. Cuvigny 2003a and b, 2005). So what was life like for the people working at the ports and quarries, and at the service stations? Were they living a life of bitter hardship, away from family and friends and without the trappings of standard aspects of Roman life? Were they far removed from Roman culture, living as they did in a remote region of the Empire? Or was their work so essential to the core of the Empire that they were well-integrated and provided for?

2During the last 30 years many archaeological projects have addressed this and other questions, and many of the results are presented in this volume. This paper focuses on one key aspect: food. What did the people working at these various sites eat and how did they obtain their food? How varied was their diet, do we see differences between the various sites, and how did their diet compare with that of people living in the Nile Valley and other parts of the Egypt? Here the botanical remains of foodstuffs, recovered in abundance from the rubbish heaps associated with the archaeological sites, are synthesized and discussed. Such remains are available from 10 sites, all dating to the 1st - early/mid-3rd century AD, though some were occupied before or after this. The supply of meat and other animal protein is discussed by Martine Leguilloux (2018).

Background

3The Romans were not the first to exploit the rich resources of the Eastern Desert. Apart from Pharaonic activity, the first systematic exploitation started in the Ptolemaic period. This intensified after the Roman occupation of Egypt, and during the late 1st century AD a well organised system was put in place, when the need for reliable water sources and security became more strongly felt (Brun 2018; Redon 2018; Sidebotham 1986; Sidebotham et al. 2008). During this time the two major quarry complexes at Mons Claudianus and Mons Porphyrites were started, and work at the two ports (Berenike and Myos Hormos) was expanded (Fig. 1). To supply and otherwise support these ports, quarries and mines a number of roads were constructed from the Nile valley, along with a series of fortified road-stations (thereafter way-stations), fortlets or praesidia (Sidebotham 1986, 2018; Sidebotham et al. 2008). Soldiers were stationed at these way-stations to assist with security and to police the roads. The way-stations had wells and accommodation, providing water, food and fodder to weary travellers and their animals, at intervals of a day’s travel, ca. 20-25 km depending on the terrain.

Fig. 1

Image 1000000000000313000002DC8C19F4D10105CE49.jpg

Map of the Eastern Desert showing the location of sites mentioned in the text. Map: S. Goddard.

© M. Van der Veen

4Food and water were needed by those travelling the roads, but also by the people manning the way-stations and those working at the ports and quarries. As virtually no food was produced locally in the Eastern Desert, most of it would have to have been brought in from the Nile valley, or, in the case of fish, from the Red Sea. These supply routes were long; for example, in antiquity the journey from Qena (Kainepolis) to the quarry of Mons Claudianus would have taken five days when carrying supplies, that from Coptos to the port of Myos Hormos six or seven days when using donkeys as the main transport animal, and from Coptos to Berenike 12 days (Bülow-Jacobsen 2003; Pliny, Natural History 6.29.103; Strabo, The Geography 17.1.4). Transport of the large, heavy columns and other stone blocks from the quarries may have taken more than a month. The quantities of food needed would have been very considerable, in view of both the large number of quarries, mines, way-stations and ports involved and considering the number of people working there at certain times of the year. While exact numbers are not known and would have fluctuated, one ostracon from the quarry of Mons Claudianus lists the presence of 917 people on a specific day (Cuvigny 2005). Another 500-900 men may have resided at the way-stations (30+ praesidia with approximately 15-30 men each; Cuvigny 2003; Maxfield 1996). Additionally, huge amounts of fodder would have been needed too, as large numbers of working animals (donkeys and camels) transported the goods from the ports and the stone from the quarries to the Nile valley, as well as bringing food and water to all the sites. These animals would have needed to be fed, as the amount of grazing available in the desert was insufficient.

5Thus, the logistics of bringing huge amounts of food and fodder to all these sites required very considerable effort and organisation. The rich archives of ostraka recovered from the rubbish heaps at these desert sites, including accounts, private letters and instructions, offer a detailed picture of these logistics (e.g. Bagnall 1986; Bagnall et al. 2000; Bülow-Jacobsen 1997, 2003; Cuvigny 1996, 2000, 2003; Rufing 1993). We learn that regular food caravans travelled the roads and delivered supplies, while private letters highlight that many further foods were requested from and sent by family and friends, either via the caravan or via people travelling to and between the various stations. Meat and vegetables are frequently mentioned, see also below.

6Additionally, we have information on the wages of the workforce at the quarries; the skilled workforce (pagani; stone-masons, smiths, quarrymen) received a monthly wheat allocation (1 artaba = ca. 39 litres), a wine-ration and a salary of, usually, 47 drachmae. Once a month these workers wrote down instructions to the quartermaster (entolai) specifying which foods they wanted and how they wanted their wages spent. From these we learn that they often arranged for their wheat allocation to be given to their female relatives in the Nile valley, to be turned into bread before being brought to the desert, and that their salary was used to buy oil, lentils, onions and dates (Cuvigny 1996). We know that the other, unskilled, workers at the quarries (the familia) were also paid a salary (amount unknown) and received 1 artaba of wheat, lentils and oil each month, and, once a year, a set of clothes (Cuvigny 1996, 2000). At the way-stations the grain for the soldiers was usually delivered in kind, to be ground, converted into bread and stored there (Brun 2011; Cuvigny 2012: 30; Maxfield 2007; Van der Veen and Tabinor 2007). While these texts inform us about the supply mechanisms and list considerable numbers of foodstuffs and food products, much additional information can be obtained from the botanical, faunal and ceramic evidence. See Leguilloux (2003, 2011, 2018) for the supply of meat and fish, Bouchaud et al. (2018) for the supply of wood, Bender (2018) and Wild and Wild (2018) for textiles, and Tomber (1996, 2006, 2008, 2018) for the ceramic evidence. Here the evidence from the surviving plant foods is discussed.

Table 1

Type of

site

Site name

(modern name)

Abbrev

No. of samples

0,5 mm

No. of liters

0,5 mm

No. of samples

2 mm

No. of liters

2 mm

Hand-picked

Total no. ids. Food plants only

Reference

Ports

Berenike

BE

228

456

-

-

45,142

Cappers 2006

Myos Hormos

(Quseir al-Qadim)

MH

43

57

40

490

9,789

Van der Veen 2011

Quarries

Mons Claudianus

MC

59

55

63

1260

8,063

Van der Veen 2001

Mons Porphyrites

MP

35

37

31

620

7,530

Van der Veen, Tabinor 2007

Tiberianè (Barud)

TI

8

10

8

160

353

Van der Veen 2001

Domitianè/Kainè

Latomia (Umm

Balad)

KL

6

11

-

-

814

Newton, in prep.

Way- stations

Badia

BA

10

12

1

20

361

Van der Veen, Tabinor 2007

Maximianon

(al-Zarqā’)

MA

26

30

2

2

1,975

Newton, unpublished

Xeron Pelagos

(Jirf)

XE

19

63

-

-

14,729

Bouchaud, unpublished

Didymoi (Kasm

al-Menih)

DI

34

54

-

-

2,266

Tengberg 2011

Archaeological sites in the Eastern Desert with botanical remains dated to the 1st-early/mid 3rd c. AD, with sample information, abbreviations of site names used in the tables and figures, and bibliographic references to full archaeobotanical reports. See also Appendix 1.

The botanical evidence

7Thanks to the arid conditions in the Eastern Desert organic remains are generally well preserved at these Roman sites, and plant food remains such as grains, seeds, fruit stones, vegetative plant tissues including chaff, as well as animal bones, textiles, leather and some papyri, have been recovered during the archaeological excavations. Botanical food remains are available from 10 excavation projects (Table 1, Fig. 1). These include the two major ports (Berenike and Myos Hormos), four quarry sites (Mons Claudianus, Mons Porphyrites, Kainè Latomia (initially called Domitianè, but renamed Kainè Latomia once Domitian fell from grace; today known as Umm Balad) and Tiberianè (also known as Barud), as well as four way-stations (Badia, on the way from Mons Porphyrites to the Nile; Maximianon (also known as al-Zarqā’), on the way from Myos Hormos to the Nile, and Didymoi and Xeron Pelagos on the way from Berenike to the Nile). As the excavations were conducted by different archaeological teams and the sampling for botanical remains was executed at different scales, we use presence/absence of food plants and relative proportions of key components to compare the plant assemblages from each site. For exact details of the sampling strategies and size of the assemblages recovered and studied, the reader is referred to the publications of each site (Table 1, Appendix 1). We concentrate on the 1st to 3rd centuries AD, the period for which most evidence exists. Data from Ptolemaic period sites are still very scarce, and while some data are available from late Roman/Late Antique phases of occupation, these are not yet sufficient for a full synthesis, and are thus only briefly summarised below, but also listed in Appendices 1 and 2.

8Almost all of the botanical remains were recovered from so-called sebakh, the large rubbish heaps of domestic waste found in and around these archaeological sites. They contain a rich archive of everything the Romans discarded, ranging from papyri and ostraka, pot sherds, fragments of glass vessels, vestiges of clothing and footwear, charcoal and ash from the fireplaces, to food and fodder remains such as animal and fish bones, seeds, fruit stones, nutshells, grain and chaff.

9It is worth stressing here that food is, of course, intended for human and animal consumption, so that in archaeology we tend to find primarily those parts of the plants that are not edible or were not digested, and were discarded in the process of food preparation and consumption. Thus we find the seeds and stones of fruits such as olives, grapes, dates, citron and sebesten, the shells of nuts such as walnut, hazelnut and pine nut, as well as the chaff and straw of the cereals (Fig. 2 and 3). Over and above these ‘waste’ products –though the chaff and straw is not really ‘waste’, see section on animal fodder below– we occasionally find table left-overs, and, in small amounts, some food remains that have been accidentally lost, for example, grains of wheat, barley and rice, the seeds of herbs and spices, such as coriander, cumin, fennel and black pepper. The actual number of items is not necessarily indicative of the importance in the diet. Not only do fruits have varying numbers of ‘seeds’ –e.g. each olive contains one stone, each fig contains several thousand seeds– their nutritional values also vary, as do their chances of survival. Rather than concentrating on actual numbers, we rely here primarily on presence/absence, frequency, and relative proportions of major components. We hope that this largely counteracts the differences in numbers of samples and volumes of deposits analysed at each individual site. Where sampling differences are a concern, this is discussed in the text.

Fig. 2

Image 10000000000002D400000400A4FF00F5CC7D03C8.jpg

1. Desiccated grains of barley (Hordeum vulgare) from Badia (after Fig. 4.19 in Van der Veen and Tabinor 2007: 106); 2. charred grains of hard/durum wheat (Triticum durum) from Badia (after Fig. 4.20 in Van der Veen and Tabinor 2007: 107); 3. desiccated rachis segments of hard/durum wheat (Triticum durum) from Mons Porphyrites (after Fig. 4.21 in Van der Veen and Tabinor 2007:108); 4. desiccated husk fragments of rice and 5. hulled grain of rice (Oryza sativa), both from Myos Hormos (after Fig. 2.5 in Van der Veen 2011: 46). Photographs: Jacob Morales.

© M. Van der Veen

Fig. 3

Image 10000000000003B1000002253C47B7B60F1FC317.jpg

1. Seeds of citron (Citrus cf. medica) (after Fig. 4.9 in Van der Veen and Tabinor 2007: 94); 2. persea (Mimusops laurifolia) (after Fig. 4.10 in Van der Veen and Tabinor 2007: 95); 3. pine nut (Pinus pinea) (after Fig. 4.11 in Van der Veen and Tabinor 2007: 95); 4. sebesten or Egyptian plum (Cordia myxa) (after Fig. 4.7 in Van der Veen and Tabinor 2007: 93). All desiccated and from Mons Porphyrites. Photographs: J. Morales.

© M. Van der Veen

Preservation

10Botanical remains, like other organic materials, would normally decay over time, but we do find them in archaeological contexts when certain preservation conditions are met; these concern charring, waterlogging, desiccation and mineral-replacement. The hyper-arid conditions in the Eastern Desert mean that desiccation is the main mode of preservation here. There is little to no bacterial decomposition and most plant remains are preserved in excellent condition, in desiccated form. Hence, the food remains discarded by the people living and working at the Eastern Desert sites are still preserved in the rubbish strewn in and around the settlements –possibly aided by the fact that there were few browsing omnivores (e.g. goats) to consume the food leftovers (as is often the case in agricultural settlements). These include vegetative remains, such as the bracts of artichoke, the skin of onions and the baseplates of garlic (Fig. 4), that is, items rarely recovered in temperate climates, as well as shells of nuts, cereal chaff, fruit stones, seeds of herbs, grains and many more. Despite the arid conditions, some sites suffer from high levels of humidity, for example at sites in parts of the Wâdi al-Hammâmât, see Bi’r Umm Fawâkhir below.

Fig. 4

Image 1000000000000313000001512D9E3BAC888DC994.jpg

1. Artichoke bracts (Cynara cardunculus, var. scolymus) from Mons Claudianus (Van de Veen 2001); 2. base plates and clove of garlic (Allium sativum) from Mons Porphyrites (after Fig. 4.15 in Van der Veen and Tabinor 2007: 97); 3. base plate and skin of onion (Allium cepa) from Mons Porphyrites (after Fig. 4.14 in Van der Veen and Tabinor 2007: 97). All desiccated. Photographs: Jacob Morales.

© M. Van der Veen

11Charred remains are also present at each of the sites (e.g. Fig. 1, item 2). These are food remains charred during daily domestic activities, such as food preparation and discard, but also during accidental fires, handicraft activities, etc., and some food remains were discarded into the fires of the smithies at the quarries. Additionally, animal dung was used as fuel, and as these animal droppings contained undigested food remains (grains and chaff, see below), these represent a significant component of the charred assemblage (see Bouchaud et al. 2018; Bouchaud and Redon 2017; Van der Veen 1999).

12At all ten sites both desiccated and charred remains have been found, but comparing the two we see that the number of food plants found in desiccated form is considerably higher than the number found charred (Fig. 5). This highlights the value of sites in Egypt (and other arid zones), as we obtain a fuller picture of the foods consumed than is possible in areas where only charred remains are preserved, thus helping us to assess the direction of loss at those latter sites (Van der Veen 2007).

Fig. 5

Image 1000000000000313000001EA8089934814C422A1.jpg

Number of food plants recovered at each site, by mode of preservation (charred and desiccated); 1st-early 3rd c. AD only. For abbreviated site names and size of each dataset, see Table 1. DI (Didymoi): information not available.

© M. Van der Veen

What did they eat?

13Remarkably, considering the remote location of the sites and the long supply routes, the range of different plant foods recovered is much broader than originally expected, with some sites having as many as 50+ different plant foods (Fig. 6). There are variations across the different types of site, see below, but 27 food taxa have been recovered in either all ten sites or in at least eight of these (Table 2; a full list, including Latin names is provided in Appendix 2). It is worth remembering here that the actual number of occurrences is a very rough measure of the importance of a food in the diet. Not only are some foods more nutritious or more essential in the diet than others, but some have a higher chance of being preserved. For example, many fruits have pits or stones that are discarded when the fruit is consumed, thus giving a higher survival rate than foods that leave little waste material, such as certain vegetables, herbs (e.g. coriander), pulses and certain oil-rich seeds (e.g. sesame). The presence of these is remarkable and highly significant.

Table 2

Image 100000000000045A000005C4EF875E5E01618A81.jpg

The most common food plants found at the ten Early Roman sites discussed in the text, that is, those that are found in 8 or more of the 10 sites; desiccated and charred preservation combined; 1st-early/mid 3rd c. AD sites only. √=present, √√=present in 50+% of samples, ?= identification not certain. See Table 1 for abbreviations and Appendix 2 for a complete list of all food plants found.

Fig. 6

Image 1000000000000313000001FA4538C32C39A6C985.jpg

Total number of plant food taxa, by food category, recovered at each site, combining desiccated and charred remains; 1st-early 3rd c. AD only. For abbreviated site names and size of each dataset, see Table 1.

© M. Van der Veen

14Two cereals, hard or durum wheat and hulled six-row barley (Fig. 2), are included in these 27 very common food taxa, with barley not just occurring in all ten sites, but also in 50% or more of the samples at all of these. Most sites also have small amounts of one or two other types of wheat, that is, emmer wheat and bread wheat (Appendix 2). They occur in such low quantities that they may simply represent contaminants from previous crops and/or poorly developed or immature specimens of durum wheat, but at a few sites (KL, MA, XE) slightly larger amounts of emmer wheat are present, suggesting emmer may have come in as a separate crop. Emmer, a hulled wheat, was the dominant wheat crop of the Pharaonic period, but was replaced by free-threshing durum or hard wheat during the Ptolemaic period, though it may have continued to be grown on a small scale during the Roman period (e.g. Clapham and Rowley-Conwy 2007; Cappers et al. 2007; Thanheiser 2002). Note that a considerable amount of emmer wheat was also found in Islamic period Qusayr-al Qadim/Kusayr, though in one sample only (Van der Veen 2011). Bread wheat, another free-threshing wheat, is common at some Roman sites, for example at Kellis in the Dahkleh Oasis and at Qasr Ibrim in southern Egypt (Clapham and Rowley-Conwy 2007; Thanheiser 2002), but its spread throughout Egypt is a modern phenomenon (Cappers 2016). The Roman Eastern Desert sites are predominantly receiving hard or durum wheat, which was grown in the Nile valley. The high frequency of barley grain, as well as wheat and barley chaff, is discussed below, in the section titled ‘Animal fodder’.

15Lentils are also very common at all sites, as are peas and vetches, while termis beans, fava beans and chickpeas occur in smaller numbers. All offer important nutrition, especially vegetable protein, in diets where meat protein was a rare luxury, while the two oil-rich seeds, safflower and linseed (flaxseeds), contribute essential fatty acids. This group of common foods also includes eight different types of fruit, i.e. dates, grapes, figs, olives, sebesten, dom, pomegranate and Christ’s thorn, which among them provide important vitamins and minerals, including essential Vitamin C. The other fruit, colocynth, may have had medicinal applications. Less common but still frequent are watermelon, melon and/or cucumber –the seeds of these are difficult to tell apart– persea and caper. Vegetables such as garlic, onion, leaf beet and cabbage also contribute vital minerals and vitamins (e.g. vitamin A, C, K, magnesium, calcium, iron), and onion and garlic also have important anti-inflammatory properties. Other vegetables, such as bottlegourd, cress and lettuce, occur less frequently (see below). Somewhat surprisingly, there are five herbs included in this list of most common food taxa, that is, coriander, fennel, cumin, black cumin and ammi. These are important in that they offer different flavours and tastes, thus facilitating the creation of variety in the basic staples (cereals, pulses). They are also commonly used in a wide variety of medicinal applications. Other herbs, found slightly less frequently, are dill, black mustard, aniseed and celery (Appendix 2).

16These results highlight that the supply of basic foods to each of the settlements was consistent and good, though, of course, shortages will have occurred. The full range of foods will not have been accessible to all (see below), but the botanical evidence highlights that the potential range of foods that the workers and soldiers received was the foundation of a nutritious and well-balanced diet, providing essential carbohydrates (energy), protein, essential fatty acids, essential vitamins and minerals, medicinal properties, as well as enough different flavours to create variety.

17The rich archaeobotanical record from the Eastern Desert stands in contrast to that of the Nile valley, the oases and the Faiyum, from where archaeobotanical evidence is still relatively scarce (though see Cappers 2005; Cappers et al. 2007; Clapham and Rowley-Conwy 2007; Murray 2000; Newton and Clapham in press; Newton et al. 2005; Smith 2003; Thanheiser 2002; Thanheiser and König 2008; Thanheiser and Walter 2015; Thanheiser et al. 2016). Enough sites have been published, though, to carry out a simple comparison, and this highlights that all the foods available at those sites have also been found at the Eastern Desert sites. This indicates that the people working and living in the Eastern Desert had access to all the foods available in the Nile Valley, subject to the usual factor of social status.

Access to luxuries

18As mentioned above, there is remarkable consistency in the number of different plant foods (the number of different taxa, not the number of seeds) across the ten sites. People at all the sites were in receipt of cereal grain, pulses, fruits, oil-rich seeds, vegetables and some herbs and spices, but there is a marked contrast between the sites in the number of different taxa (Fig. 6). While differences in the number of samples analysed may have some bearing on this, the pattern seems robust. Four sites, namely the two ports of Berenike and Myos Hormos, and two of the quarries, Mons Claudianus and Mons Porphyrites, have access to more than 50 different plant foods, while the remaining six sites (two quarries and four way-stations) have access to between 16 and 32 taxa, which, nevertheless, represents a good dietary range, as discussed above. The four sites with 50 or more food plants are characterized by greater variety, by a wider range of nuts, fruits, as well as herbs, spices and vegetables (Table 3). For example, almond, pine nut, hazelnut and walnut are fairly common at these four ‘food-rich’ sites, but absent at the way-stations and two of the quarries, with the exception of some very scarce remains of hazelnut and walnut at Maximianon. Most of the plants found at the four ‘food-rich’ sites are foods that must have been rare and expensive in the Nile valley at the time, such as the nuts, but also fruits such as apple, peach, apricot, plum and citron, certain vegetables (e.g. artichoke), as well as imports such as black pepper and rice (Fig. 2 and 7). Other rare foods are sesame, fenugreek and tamarind. These foods can be described as ‘extras’ or luxuries, that is foods not essential to human nutrition, but offering something extra (Van der Veen 2003), such as extra flavour and aroma (black pepper, fenugreek, sesame, several herbs, citron), variety (apple, peach, plum, artichoke, endive, rue, purslane), additional tastes and texture (the nuts, which also provided extra protein, minerals and fatty acids).

Table 3

Image 100000000000031E00000303DD50701E83C70677.jpg

Occurrence of so-called ‘extras’ or 'luxury foods' at the four 'food-rich' sites, i.e. at Berenike, Myos Hormos, Mons Claudianus and Porphyrites; desiccated and charred remains combined; 1st-early/mid 3rd c. AD sites only. See Table 1 for abbreviations and Appendix 2 for a complete list of all food plants found.

Fig. 7

Image 100000000000034D00000313B56F9584EA2247A6.jpg

1. Black peppercorns (Piper nigrum) (after Fig. 2.3 in Van der Veen 2011: 44); 2. belleric myrobalans (Terminalia bellirica) (after Fig. 2.8 in Van der Veen 2011: 51); 3. black cumin (Nigella sativa) (after Fig. 4.20 in Van der Veen 2011: 168), and 4. sesame (Sesamum indicum) (after Fig. 4.15 in Van der Veen 2011: 160). All desiccated and from Myos Hormos. Photographs: Jacob Morales.

© M. Van der Veen

19This pattern of food occurrences highlights that at Berenike, Myos Hormos, Mons Claudianus and Mons Porphyrites there were individuals who were in receipt of a wider range of foods than most of the workers and soldiers, suggesting that at these four sites there was a regular presence of senior personnel, including army officers, who could afford to get such extras on a regular basis. This fits very well with what we know of these sites. All four would have had significant numbers of higher ranking personnel; the two ports because of the need to supervise the spice trade, which was both expensive and lucrative, and the two quarries Mons Claudianus and Mons Porphyrites, because of the need to oversee the workings at the quarries, both large quarry complexes supplying prestigious imperial building projects in Rome. The other two quarries, Tiberianè and Kainè Latomia, are much smaller, subsidiary operations, run from Mons Claudianus and Mons Porphyrites, and thus without the need for permanent senior personnel. At the way-stations (praesidia) such individuals would not be resident, but would travel through on the way to the ports or the two large quarry complexes. Notably, their sporadic presence is reflected in the botanical assemblages, in that these ‘extras’ are only rarely found here. Similarly, these foods are rarely mentioned in the ostraka; for example, we know of only one mention each of citron, fenugreek, and four mentions of apples in the ostraka from these way-stations, and of rare references to chickpea, pine nut, sesame and artichoke in those from Mons Claudianus (Bulow-Jacobsen 2003; Cuvigny pers. comm.), corroborating the botanical evidence that these foods were available to certain individuals only.

Vegetable gardens

20A surprising find was that of seeds of certain vegetables, such as leaf beet, cabbage,1 lettuce, cress and endive (Table 4). It is normally the fresh green leaves of these plants that are consumed, not the seeds (though oil can and has been extracted from the seeds of the brassicas and lettuce). We would expect to find these seeds in locations where the seeds are sown, and where the plants are cultivated and harvested, that is, in horticultural locations such as gardens. The leaves are harvested before the plant sets seed, but occasionally the plants will bolt before the leaves have been picked. When vegetables have bolted, they are not normally sold, as the leaves will no longer be palatable. The presence of the seeds of these green vegetables thus suggests that green vegetables were cultivated in the desert, at several sites, with either waste water or water from the wells (Cappers 1998, 2006: 140-141; Van der Veen 1998; 2001: 200-201; Van der Veen and Hamilton-Dyer 1998). This makes sense, as fresh green vegetables would have formed an important contribution to the intake of vitamin C and iron, and the long supply routes would have made regular delivery of such greens problematic; most would have wilted before arrival. This botanical evidence for the local cultivation of green vegetables is corroborated by textual evidence found at these sites, by the ostraka, which include private letters requesting or confirming delivery of vegetables, e.g. beet (O.Claud. 150, 228, 232), cabbage (O.Claud. 226, 229, 255, 256; O.Krok. 85.4-6; O.Did. 391, 377, 461), endive/chicory (O.Claud. 228), lettuce (O.Claud. 226, 370) and ‘vegetables’ unspecified (O.Claud. 238, 256, 258, 270). The origin and destination of these letters likewise suggest that these vegetables were grown at several locations in the desert (Bingen 1997; Bülow-Jacobsen 1992, 1997; Cuvigny 2007, 2011). See also discussion section below.

Table 4

Ports

Quarries

Way-stations

Site name abbrev:

BE

MH

MC

MP

TI

KL

BA

MA

XE

DI

Cabbage

-

√√

√√

√√

√√

?

√√

?

-

Leaf beet

√√

√√

√√

√√

√√

-

Cress

-

√√

√√

?

-

-

-

-

Lettuce

-

-

-

-

-

Endive

-

-

-

-

-

-

-

Basil

-

-

-

-

-

-

-

Mint

-

-

-

-

-

-

-

Rue

-

-

-

-

-

-

-

Purslane

-

-

√√

√√

-

-

-

-

-

Occurrence and relative abundance of seeds of green vegetables at each site; desiccated and charred preservation combined; 1st-early/mid 3rd c. AD sites only. √= present, √√= in 20% or more of samples, ?= identification not certain. See Table 1 for abbreviations and Appendix 2 for a complete list of all food plants found.

21It is worth noting that the vegetable seeds are more common at the quarry sites, than at the ports or way-stations, and this despite the fact that at both Kainè Latomia and Tiberianè the number of samples analysed was very low (Tables 1 and 4). As the sampling strategy at the port of Myos Hormos was identical to that at the quarries of Mons Claudianus, Porphyrites and Tiberianè, where such seeds were relatively widespread, a difference in sampling cannot be the (only) explanation. It may be that the workers at the quarries stayed on location longer than those working in the ports, and thus were in need of fresh greens more regularly, and, maybe, had more spare time as well. It may also be that locations where such vegetables could be grown were located more closely to the quarries than the ports. Certainly, fresh water, rare though that is in the desert, was easier to access in the desert than on the coast, where most water would have been brackish. The ostraka highlight that there were certain way-stations specialising in vegetable growing, such as at Persou (Fawâkhir), Phoinikon (al-Laqîta) and Kompasi (Bi’r Daghbag), all three located on the roads to the ports (Berenike and Myos Hormos), as well as at Raïma, location unknown (Bingen 1997; Bülow-Jacobsen 2003; Cuvigny, 2007). However, the common occurrence of seeds of these leafy vegetables at the quarries of Mons Claudianus and Porphyrites indicates that they were grown here too. This is corroborated by an, as yet unpublished, ostracon from Mons Claudianus which instructs one of the men to go and work in the garden: “let Petraerès go work in the garden” (O.Claud. inv. 729; Cuvigny, pers. comm.). Given the remote location of the quarry and settlement of Kainè Latomia, the presence of seeds of green leafy vegetables here (lettuce, beet and possibly cress and cabbage) also points to local cultivation, probably by a nearby well, Sabelbi, mentioned in ostraka, or at Prasou (Cuvigny, pers. comm.).

22That sufficient water was always an issue is clear from two ostraka: from O.Claud. 280, which is a request for water and manure in connection with the comment that “the vegetables have not grown yet” (Bingen 1997), and from an ostracon recovered at Umm Balad (late 1st or early 2nd century AD; O.KaLa. inv. 200; Cuvigny pers. comm.). It is a letter written by the curator of a nearby way-station where there was a vegetable garden. The text explains that a shortage of water has meant that the vegetables have not grown very well. He sends the bunches of greens (roots and leaves) from 7 small planting beds in the hope that they may be of some use, possibly to be given to the workers, as they would be unworthy for the table of the addressee, Hieronymos, an architectus also known from Mons Claudianus. This letter might be taken to mean that vegetables were normally intended for people of high status, but other interpretations are possible too, see discussion below.

Animal fodder

23One of the most numerous categories of plant remains found was chaff (rachis nodes) and straw, and especially that of durum or hard wheat, while grains of barley were also numerous (Fig. 2). To give an impression of the quantities involved, the total number of wheat chaff fragments from the ten sites exceeds 20,000 and the number of barley grains exceeds 3,000 (data from samples sieved over 0.5mm only). When we plot the relative proportions of wheat and barley chaff as well as wheat and barley grain, a distinct pattern emerges (Fig. 8 and 9). Wheat chaff is consistently more numerous than barley chaff (on average wheat represents ca. 80%, barley chaff 20%), while these proportions are the reverse for the grain (barley grain ca. 75%, as against 25% of wheat grain). While people do and did eat barley, most will by preference eat wheat. We also know from the ostraka that the workers at the two main quarry complexes were paid partly in wheat and that this wheat was often made into bread by family in the Nile valley, before being sent to the quarry (see above). Thus, the lower amounts of wheat grain do not mean that wheat was not consumed, but that most of it was eaten and thus not discarded in the rubbish heaps. Moreover, a considerable amount of wheat arrived in the desert already converted into flour or bread. Besides, the wheat would have been a precious foodstuff, not to be wasted. Barley reached the site as grain and may have been used to make bread, gruel or porridge, but may additionally, or, more likely, have been brought in primarily to feed the animals.

Fig. 8

Image 10000000000003130000023197B9081F491DFBF1.jpg

Relative proportions of desiccated barley and wheat grains at each site; 1st - early 3rd c. AD only. N = cereal grain (minimum number of individuals) in samples sieved over 0.5mm mesh (i.e. excluding hand-picked material and ‘large’ samples). For abbreviated site names, see Table 1.

© M. Van der Veen

Fig. 9

Image 1000000000000313000002309071503BE1810909.jpg

Relative proportions of desiccated barley and wheat chaff (rachis nodes) at each site; 1st - early 3rd c. AD only. N = cereal rachis nodes (1 rachis node = 1) in samples sieved over 0.5 mm mesh (i.e. excluding hand-picked material and ‘large’ samples). For abbreviated site names, see Table 1.

© M. Van der Veen

24The quantities of barley grain and wheat chaff recovered are interpreted as animal fodder. The working animals at these sites were donkeys and camels and large numbers of them will have been needed for the long journeys across the desert. Today these pack animals are fed green fodder, chaff and straw of wheat as roughage, as well as barley grain as a power food. Barley grain is especially important when animals have to perform demanding tasks, such as pulling carts with heavy stone from the quarries. Corroboration that the working animals were indeed fed these foods comes from the remains of dung found at the excavations. Both undigested wheat chaff and barley grains have been found in the animal droppings of camels and donkeys (but not in those of sheep or goat) (Fig. 10). See also Bouchaud and Redon 2017; Van der Veen 2001: 207-8; Van der Veen and Tabinor 2007: 99; Van der Veen 2011: 175). Moreover, the texts indicate that some barley was intended for the pigs and piglets that were kept at some sites (Bülow-Jacobsen 2003; Leguilloux 2018). Some of the desiccated chaff and barley grain recovered from the sebakh deposits may represent dung that has disintegrated, or remains of animal and human bedding material, but much of it will be fodder that was not consumed. Clearly these animals were rather messy eaters, and the fodder that was not consumed, especially the wheat chaff and straw, would have been blown around the site by the wind and thus became incorporated into the many refuse dumps. These remains were found at every site and in virtually all sebakh, highlighting the importance of these animals to the exploitation of the Eastern Desert and the numbers in which they must have been present at the settlements (see also Leguilloux 2018).

Fig. 10

Image 1000000000000313000002FA236E57CF330AACEA.jpg

Desiccated camel dung from Mons Porphyrites, showing complete barley grains and wheat chaff incorporated in the droppings (after Fig. 4.16 in Van der Veen and Tabinor 2007: 99). Photographs: Jacob Morales.

© M. Van der Veen

Indian trade: pepper and other exotics

25As mentioned, two of the sites studied were ports for the trade with the East, especially the trade in spices and other commodities from India and further afield, and some evidence for this trade has, indeed, been found. Foodstuffs imported from the Indian subcontinent include black pepper, recovered in quantity at both ports, as well as rice, mung bean, coconut, belleric myrobalan, emblic myrobalan and Job’s tears, the latter possibly representing beads rather than food, though also edible (Table 5; Figs 2 and 7) (Cappers 2003, 2006; Van der Veen 2011; Van der Veen and Morales 2015; Wendrich et al. 2003). Most of these imports are found exclusively at the two ports, though two peppercorns were found at Mons Claudianus, one rice grain at Didymoi, and one fragment of coconut at Xeron. Remarkably, there is no botanical evidence for the other spices traded, such as cinnamon, ginger, long pepper, or cardamom. It is worth remembering, however, that these latter spices were very expensive and thus out of reach of most people living and working in the ports. Their expense also meant that every effort would have been made to avoid spillage during transhipment in the ports. The eastern foodstuffs were largely intended for the wealthy in Rome and other parts of the Empire and they were used in perfumery, ritual and medicine, rather than exclusively in cuisine. The exception is black pepper, which was not only more widely available, but also more extensively used in cuisine. Thus, the absence of cinnamon, ginger and other spices at the ports does not mean that they were not traded and imported, but, instead, that they were not consumed by the residents of the ports and that accidental spillage during transhipment was avoided.

Table 5

Ports

Quarries

Way-stations

Site name abbrev:

BE

MH

MC

MP

TI

KL

BA

MA

XE

DI

Black pepper

√√

√√

-

-

-

-

-

-

-

Rice

√√

√√

-

-

-

-

-

-

-

Coconut

√√

√√

-

-

-

-

-

-

-

Mung bean

-

-

-

-

-

-

-

-

Belleric myrobalan

-

-

-

-

-

-

-

-

-

Emblic myrobalan

-

-

-

-

-

-

-

-

-

Job’s tears

-

-

-

-

-

-

-

-

-

Occurrence and relative abundance of spices and other foodstuffs imported from the Indian subcontinent at each site; desiccated and charred preservation combined; 1st-early/mid 3rd c. AD sites only. √=present, √√=in 20% or more of the samples. See Table 1 for abbreviations and Appendix 2 for a complete list of all food plants found.

26In contrast, much of the pepper found at the ports may represent accidental loss during transhipment or storage. For example, at Berenike much of the pepper was found near buildings associated with the trade, while at Myos Hormos the pepper is mostly found in an area near the harbour, where the road sets off for the Nile valley, which is where transhipment took place, though at both sites some pepper was recovered from areas representing food preparation and consumption (Cappers 2006: 114; Van der Veen 2011: 190-194). As mentioned, pepper is the most numerous spice in the deposits at the ports; in Berenike in excess of 3,000 peppercorns were recovered. It was, in actual fact, the most common spice traded in Roman times, used across the Roman Empire in cooking and as a table spice, but also in medicine, as an antidote to poison, and, importantly, in religious ritual (Dalby 2000). It is worth noting here that some 80% of the peppercorns at Berenike were charred, and much of these were found in association with shrines and with the temple dedicated to Serapis, suggesting that the pepper was used in offerings. This matches what we know from historical sources that pepper, and spices more generally, were indeed used in religious rituals. One spectacular find is that of a storage jar, containing some 7.5 kg of desiccated peppercorns in the courtyard of the so-called Serapis temple at Berenike. The jar originates from India, suggesting the pepper was still in its original packing (Cappers 2006: 114; Tomber 2008: 76-77).

27Pepper was one of the few imported spices regularly used in cuisine, as can be seen in the cookbook of Apicius, a collection of ca. 400 recipes probably compiled in the late 4th or early 5th century AD (Warmington 1928: 182). Out of these recipes some 80% have black pepper as one of the ingredients, including meat, fish and vegetable dishes, as well as sweet desserts. Cumin occurs in only 3% of recipes and rice in 1% (Petersen 1980). Rice was imported and consumed in the Roman period, but only used infrequently and largely in medicinal contexts (André 1981: 54; Dalby 2000: 197; Dioscorides, De Materia Medica II.117; Konen 1999). The rice at Berenike and Myos Hormos includes both hulled grains (spikelets, i.e. the grains still enclosed by the hulls) and also hull fragments. The latter indicate that the rice was dehusked and consumed at the ports, because the hulls are not easily digested by humans and dehusking thus takes place prior to food preparation and cooking. It is likely that the rice was consumed by Indian sailors who temporarily resided in the ports. Their presence is hinted at by ceramics of Indian origin, epigraphic evidence (graffiti) and other finds (Tomber 2008: 73-77, 2011; Sidebotham 2011: 74-75).

28The rare occurrences of mung bean and coconut may also represent foods brought and consumed by Indian sailors on the journey from India to Egypt, rather than imports or foods for the workers and other personnel in the ports. Certainly, mung bean has, to our knowledge, never been found in Roman deposits in the west, and coconut only once, at Xeron, one of the way-stations on the route from Berenike. Coconut may, in fact, have been carried on the journey from India not as a foodstuff, but as a raw material to make rope, cordage and caulking. The remains found at the ports consist of fragments of the shell (endocarp) and sections of the fibrous husk (mesocarp; coir). Both belleric myrobalan and emblic myrobalan are fruits mainly used as health supplements and in medicines. Again, both may represent products brought for personal use by the Indian sailors (Cappers 2006: 108-109; Van der Veen 2011: 50-53).

Changing worlds

29The picture of a consistent and stable food supply described above is only valid for the later 1st to early 3rd centuries AD. The political and economic turmoil of the 3rd century that affected the Roman Empire also had repercussions for the exploitation of the Eastern Desert. It was no longer viable to maintain both ports and Myos Hormos was abandoned, while activities at Berenike were much reduced from the later 2nd century, though these pick up again during the 4th to early 6th centuries. The quarries at Mons Claudianus, Tiberianè and Kainè Latomia, are also abandoned, the latter two possibly during the later 2nd century AD. The quarrying at Mons Porphyrites also sees a hiatus, but, as at Berenike, activities resume during the 4th and 5th centuries. Finally, occupation at many way-stations ceases. It is too early to offer a detailed synthesis of how this affected the food supply during the later Roman period, but here the little evidence that currently exists is mentioned.

30The clearest picture comes from Berenike where there is evidence for a continuation of the trade with India (black pepper, rice, coconut). The food supply similarly continues, including wheat, barley, lentils, lupin, dates, grapes, olives, coriander, cumin, and even a few luxuries (peach, pine nut and walnut), though there is a reduction of those in range and number (Cappers 2006: 160-162). However, the evidence also points to change, in particular towards a greater reliance on foods available from the local, desert area, such as the fruits of Cordia nevillii/sinensis (no English name known), sugar date, Christ’s thorn, dom palm and bentree. This suggests that contact between the Roman occupants of the port and the local nomadic population increased, with some of the native population possibly even residing in the port and satellite sites (ibid.; Appendix 2). Additionally, we see the appearance of grains of sorghum, a cereal crop more common further south (in the Sudan), but increasingly becoming important in southern Egypt (Nubia) and the Western oases by then (Fuller 2014; Thanheiser 2002; Thanheiser et al. 2016). With several of the way-stations on the road from Berenike to the Nile Valley abandoned, some of the food supply may have been delivered from Clysma (ancient Suez) in the north, via the Red Sea (Brun, pers. comm.).

31At Mons Porphyrites the main fort is abandoned, but one quarry, the Lykabettus quarry, high up in the mountains, is in use during the 4th and early 5th centuries. It has a small village housing the workmen nearby (Maxfield 2001a). The quarry operations are at this time linked with the way-station at Badia, where we find large storage facilities (probably for food), stone mortars and quern stones suggestive of large-scale milling of grain (Maxfield 2001b: 223; 2007), which is matched by large deposits of burnt wheat grain. It is likely that during the Late Roman period communal bread baking was carried out at Badia, with the bread sent up to the quarry and village of Lykabettus. Here too we see a reduction in the range of plant foods, especially in the range of luxuries or extras, though the smaller scale of sampling may be a factor too (Appendix 2; Van der Veen and Tabinor 2007).

32The only other Late Roman/Late Antique botanical data currently available come from the fort of Abu Sha’ar on the Red Sea coast and from the gold mines at Bi’r Umm Fawâkhir in Wâdi al-Hammâmât. At both these sites the sampling for botanical remains was very limited and the data can thus not be directly compared with those of the other sites, but the identifications have been added to Appendix 2. Abu Sha’ar is a Late Roman fort on the Red Sea coast, built at the beginning of the 4th century AD; it was then abandoned between the middle and the end of this century, and then reoccupied by Christians during the 5th and 6th c. AD (Sidebotham 2008: 59). Many of the food plants are similar to those at the earlier Roman sites, though with a reduced range and especially fewer herbs (the lack of herbs may be related to different sampling methods used). Wheat and barley together with lentils, dates, onion and garlic appear to have been the staples, while fava beans, chickpeas, coriander, olives, grapes and sebesten also feature. Remarkably, there are good numbers of almonds, walnut and hazelnut, as well as some peach and one seed of Citrus sp., probably citron, see Appendix 2 (el-Hadidi and Amer 1996; el-Hadidi and el-Fayoumi 1996; el-Hadidi et al. 1997; Fadl 2013).

33The assemblage from Bi’r Umm Fawâkhir is very sparse, not just because sampling was more limited, but because only charred plant remains were recovered. No desiccated material has survived, probably because there was more water present in the area, the site being located in a side section of the main Wâdi al-Hammâmât, which is comparatively well-watered and sees regular flash floods, which must have affected the preservation of organics. In fact, one of the main vegetable growing sites during the early Roman period, Persou, was located very close to the gold mine village. The gold mines and associated settlement were in use during the 5th-6th centuries AD, though earlier exploitation is known (Meyer 2014). The food plants comprise wheat and barley, as well as dates, dom, olives, grape, and bottlegourd, see Appendix 2 (Smith 2014).

34Despite this narrow dataset, the evidence for a somewhat reduced range and greater reliance on local food plants seems to tally with that of the faunal, textual and archaeological evidence. The faunal remains also highlight changes in supply and in particular a greater reliance of locally available foods. In this case a switch from pork (brought in from the Nile valley) to meat of sheep and goats, most likely obtained from local nomads (Leguilloux 2018). While several texts refer to hostilities between the Romans and the local nomadic population (usually referred to as barbaroi), there are some that imply better relations from the 3rd century onwards, with food and other products obtained from and sold to them. For instance, several ostraka from Xeron do refer to the delivery of substantial quantities of wheat to the barbaroi (Cuvigny 2014).

Table 6

Crops from South and South-East Asia

African crops

Rice 1

Cardamom

Sorghum 2

Lime

Ginger

Pearl millet 2

Banana

Turmeric

Black-eyed bean

Aubergine

Betelnut

Cotton 3

Taro

Fagara

Sugarcane

Cotton 3

Food plants from South-East Asia and Africa that become common in Egypt during the Islamic period. 1) rice was imported into Egypt from the Roman period, and cultivated there from the Islamic period; 2) sorghum and pearl millet start being grown in Egypt during the later Roman period, possibly as early as the 2nd c. AD (Thanheiser et al. 2016), but their production increases in scale during the Islamic period; 3) cotton was imported from India during the Roman period, and starts being cultivated in Egypt at that time, but, as with sorghum and pearl millet, its production increases in scale during the Islamic period. African cotton (a different plant species from the Asian one) has been found in Roman Qasr Ibrim, southern Egypt (Nubia) (Clapham and Rowley-Conwy 2009; Palmer et al. 2012), meaning that Egyptian cotton finds may have African as well as Indian origins (aDNA analysis can help distinguish between the two).

Fig. 11

Image 1000000000000313000002226B86D9B1B35B561A.jpg

Total number of plant food taxa recovered at each site, combining desiccated and charred remains; comparing the Early Roman sites (1st-early 3rd c. AD) with the medieval Islamic remains from Kusayr (11th-13th c. AD). For abbreviated site names and size of each dataset, see Table 1, plus KU=Kusayr. See also Appendix 2.

© M. Van der Veen

35By the end of the Late Roman period, all these sites are abandoned, but at one port we see renewed activity in the medieval Islamic period. At Qusayr al-Qadim, the location of the early Roman port of Myos Hormos, the port is brought back into use during the 11th century AD, but it is now called Kusayr. Two phases of occupation have been recognized, during the 11th-13th centuries, and on a much reduced scale during the 14th-15th centuries (Peacock and Blue 2006, 2011). While chronologically outside the scope of this paper, it is worth briefly reviewing the evidence for food plants here, as it helps highlight the close link between food and geopolitics, and the specific ‘Roman’ character of the 1st-3rd century evidence. A listing of the food remains is given in Appendix 2 and the data are described in full in Van der Veen (2011). The most striking aspect of this botanical assemblage is the significantly greater range of food plants reaching the port during the 11th-13th centuries (80 in total), including the larger number of spices and other food plants originally from the Indian subcontinent and Africa (Fig. 11). These include ginger, cardamom and turmeric, as well as sugar cane, aubergine, banana and taro from South and South-East Asia, along with sorghum, pearl millet and black-eyed bean from Africa (Table 6). This increase in foods from those regions implies changes in the nature of the trade, as well as changes in the agricultural system in the Nile valley, as, with exception of the spices, these crops become integrated into Egyptian agriculture. Importantly, it also highlights stronger ties with the Indian Ocean world and countries bordering Egypt, and with this the geopolitical shift that Egypt has undergone after the end of the Roman and Byzantine periods (Van der Veen 2011; Van der Veen and Morales 2015, 2017). While there are as yet very few sites of Islamic date with botanical evidence, thus making it difficult to evaluate the prevalence of these new foodstuffs in Egypt at that time, the ample textual and historical evidence highlights that these foods had rapidly become part of Egyptian agriculture and cuisine (Van der Veen 2011: Chapter 3 and Table 3.6).

Discussion

36Going back to our original question of what life was like for the Romans working in the Eastern Desert, our evidence indicates that their lives, while obviously different from that of those living in the Nile valley and other parts of Egypt, were not as dissimilar as one might expect, especially when we focus on food. Plant foods would have been the main source of energy and protein in the diet for most of the inhabitants of Egypt, and the range of plant foods available in the desert during the later 1st to early 3rd centuries AD was very similar to what was available in other parts of Egypt, from where most of the food originated. This is not as surprising as it may seem at first. Great care and effort was invested in organising the trade of eastern commodities and the quarrying of prestigious building stones (granodiorite and porphyry), and this same care and effort was invested in looking after the people who made this possible. This supply seems very organised and stable with a wide range of staple foods delivered on a regular basis. This range is considerably greater than the rations mentioned in the ostraka imply, and also much greater than can be inferred from the private letters. It is illustrative in this respect that most of the extra or luxury foods at the two ports and two major quarry complexes (Table 3) are not mentioned in the ostraka, highlighting the fact that many foodstuffs were delivered to the sites without the need for written documentation. It is clear that private enterprise flourished, including the growing of green vegetables in small garden plots at sites where enough water was available (including at the major quarries), using private letters along with oral communications to request and receive additional foods and other goods from family and friends in the Nile valley and in other stations in the desert, as well as ad hoc, individual purchases from the food caravans.

37We will never know in exact detail what degree of difference existed between the various classes of people working and living in the desert in terms of access to foodstuffs. The large heaps of domestic waste (sebakh) found in and around each of our sites were often repositioned, moved from one room to another or redeposited outside of the settlement when the need arose, something observed in many areas of Egypt and other arid zone countries (Rowley-Conwy 1994; Van der Veen 2007), so that we are rarely able to define different social groups by spatially differentiated refuse. Our evidence does, however, indicate that higher ranking officers were able to access a wider range of extras, such as hazelnuts, walnuts, almonds and pine nuts, artichoke, peach, citron, mulberry, persea, black pepper, and fenugreek. Social differences in access to certain foods did, of course, exist throughout the Roman Empire, and Egypt would have been no exception. What the botanical evidence suggests is that within their social rank the people living and working in the desert had access to more or less the same plant foods as they would have had in the Nile valley.

38One unexpected aspect of our results concerns the growing of green vegetables in the desert, something the ostraka have corroborated. The ostraka give the impression that it was primarily the soldiers and their officers who asked for and received these green vegetables (Bülow-Jacobsen 2003), though care is needed with this interpretation. Firstly, soldiers and civilian workers often had the same names and the letters rarely mention the profession of the writer or recipient, though future detailed analysis of the names may provide more information (Cuvigny pers. comm.). Secondly, while the soldiers earned more than the skilled civilian workers, the latter earned more than their counterparts in the Nile valley (Cuvigny 1996), probably as compensation for living away from family for long spells of time. Consequently these workers would have had the means to purchase extra foods if they so wished. Thirdly, we should remember that the writing of letters is culturally and socially determined. The workers at the quarries may, for example, have utilized different forms of communication to stay in touch with relatives and friends in the Nile valley, such as oral messages sent via the caravan, and obtained extras that way. Finally, the ostraka mentioning vegetables and the seeds of green vegetables may represent evidence from different groups of people, that is, the ostraka may principally represent the activities of the soldiers, and the seeds those of the workmen at the quarries (see Van der Veen 1998 for a fuller discussion). Most of the soldiers were stationed at the various way-stations; at some of these there was enough water to grow vegetables, and these were sent from there to other stations. We know that they were sent from an unknown station to the quarries at Mons Claudianus too, but vegetables were also cultivated at Mons Claudianus itself, which is clear not just from the many vegetable seeds recovered here, but also from an ostracon from Mons Claudianus instructing one of the men to go and work in the garden (see above). Significantly, having access to several lines of evidence from these sites (botanical, ceramic, faunal and textual) allows us to appreciate that only by combining all this evidence are we able to get closer to a full reconstruction of the food supply and what it signifies.

39Plant foods were the most important component of the diet, but they were, of course, not the only foods consumed. The ceramic evidence highlights that vast quantities of wine and olive oil (that is, processed plant foods), as well as fish-sauce, were delivered to each of the sites, including some from as far afield as Spain, Italy and Greece (Tomber 1996, 2006, 2008, 2018). The ostraka also mention processed foods, such as wine, beer, vinegar and oil (Bülow-Jacobsen 2003). Moreover, the faunal remains indicate that the two ports, as well as the quarries Mons Claudianus and Porphyrites received large quantities of Red Sea fish, while all sites received pork, small amounts of sheep or goat, as well as chicken. The chicken and the piglets were often kept on site. At the quarries the working animals were also consumed, probably when these animals had reached the end of their useful lives (Hamilton-Dyer 2001, 2007, 2011; Leguilloux 2003, 2011, 2018; Van Neer 1997). The faunal remains show more variability across the sites in availability of animal protein, partly related to closeness to the Nile valley and/or the Red Sea, and partly to the relative abundance, and thus availability, of donkeys and camels for work purposes (see Leguilloux 2018).

40We can thus conclude that, despite their remote location, the people working and living in these Eastern Desert sites were, in terms of food, well-looked after. This is further corroborated by other evidence. For example, to improve the psychological comfort of the quarry workers the Roman authorities arranged for their families to be brought closer to the quarries, by creating a new town, Kainè (modern Qena), in the Nile valley, at the point where the roads to the quarries departed (Cuvigny 1998). These relatives converted the wheat rations of the workers into bread before it was brought into the desert, and no doubt other foods and goods will have been delivered along with the bread. At several sites, including at the way-stations on the roads to the two ports and the quarries, women and children were able to live alongside the men, and at many sites there is evidence for the regular arrival of prostitutes (Cuvigny 2003; Van Rengen 1992). All this suggests that many of the needs of the men were catered for in one form or another. This is not to say that life was always easy and comfortable; the heat and dust of the desert combined with, especially at the quarries, the hard labour would have made for a difficult life, with occasional shortages, and with at times the threat of attacks by the local nomads, who, in the ostraka, were often referred to as ‘barbarians’.

41One notable aspect of our results is the fact that the botanical evidence highlights that these Eastern Desert sites and their inhabitants were fully embedded within the Roman Empire and its political and economic fortunes. While the eastern trade was driven by the Roman elite’s desire and demand for “eastern” goods, the costs of sustaining the ports and the roads could be met by levying taxes on the spices and other eastern imports. In contrast, the extraction of the stone was a purely imperial vanity project, it made no economic sense. The stone was used in prominent imperial building projects (e.g. the Pantheon, Trajan’s Forum, baths, basins, sarcophagi, statuary), and both the complexity of its extraction and the remoteness of its origin emphasized and enhanced the greatness and prestige of the emperors; the stone was a symbol of imperial power (Peacock 1992). Both the trade and the quarrying required complex logistical linkages between the Eastern Desert and the Nile valley, the Red Sea coast and India, and between Rome, the Mediterranean and further west. When the Empire was hit by political and economic upheaval in the 3rd century AD, this directly affected the sites in the Eastern Desert, with one port and one major quarry complex, as well as a number of way-stations, abandoned, others temporarily reduced in size and function. By the 4th century operations resumed. By then, power relations in the desert had also changed, with evidence for greater cooperation with the local nomadic population and greater reliance on foods provided by them, rather than brought in from the Nile valley, though the latter still occurred, with further provisions possibly being shipped in from Clysma (ancient Suez).

42The link between food, identity and geopolitics becomes even clearer when we compare the foodstuffs available in the Eastern Desert during the Roman period with those during the medieval Islamic period when Egypt had become part of the Islamic world. Many eastern spices that were imported during the Roman period, but largely used in ritual and perfumery –the exception is black pepper–, are now more widely available and used in cuisine. Additionally, many other foodstuffs from South and South-East Asia have become incorporated into agriculture and cuisine. In many ways we can see the food choices of the people working and living in the Eastern Desert as expressions of identity. Despite working in a remote region, or more accurately, because of working in a remote desert location, the choice of foods was strongly focused on those that were current in the Nile Valley. In the Roman period these are foods widely available in Egypt and the Mediterranean at that time. In contrast, by the medieval Islamic period the political alignment of Egypt had changed, and we now see many of the spices and other foodstuffs from South and South-East Asia accessible and incorporated into both agriculture and cuisine. Foodways had changed and shifted from a focus on Rome to a focus on the Middle East (for a fuller discussion, see Van der Veen 2011; Van der Veen and Morales 2017). Importantly, we should regard these changes in diet (and agriculture and trade) not as passive outcomes or passive consequences of geopolitical transformations. Identities are created and maintained through food. What it is to be part of the Roman Empire is connected to food practices, to the daily routine of acquiring, preparing and consuming food. Thus, the changes we see in the archaeobotanical record between the Roman and medieval Islamic periods were part of the geopolitical realignment of the Eastern Desert and of Egypt at that time. This realignment was not just a political reality; it was a reality in everyday life.

Conclusion

43There can be no doubt that the goods produced by the quarries were of great significance to the expression of imperial power, and that the “eastern” goods imported via the ports were central to the lives of many, including the social elite. This explains the huge investment in infrastructure and protection in the Eastern Desert. Roads and way-stations were constructed to supply these sites and regular caravans brought food, water and many other items needed. The botanical and faunal remains, together with the ceramics and textual evidence, recovered at these Roman sites highlight that the delivery of food was well organised and consistent. The botanical assemblages at the different sites are very similar to one another, and together with the documentary evidence point to a well-planned supply system, with standard supplies augmented through a variety of private enterprise, such as growing green vegetables in the desert, requesting foods from family and friends back in the Nile Valley, as well as individual purchases from passing caravans. Notwithstanding the many logistical obstacles and inevitable shortages and delays in the supplies arriving at the sites, the range of foods available to the workers and soldiers was impressively wide, including all basic essentials (cereal grain, pulses, fruits, oil-rich seeds, and vegetables), as well as several herbs, the latter including taxa such as coriander, fennel, black cumin and ammi. Some were also able to access luxuries such as artichoke, almonds, pine nuts, walnuts, citron, persea, peach, apple, plum and sesame. These latter foods were, considering their expense, probably restricted to the higher-ranking officers, which is borne out by the fact that they were recovered primarily from sites that housed such personnel.

44The Eastern Desert was important to the Roman Empire, as a source of prestigious building materials, precious stones and metals, as well as the eastern commodities brought to the Red Sea ports. These high-status, as well as a few quotidian, goods played a central role in the way the Empire functioned and this is reflected in the effort and care that went into supplying the people who worked to bring these goods to Rome. Thus, despite being located on the geographical fringe of the Empire, the Eastern Desert’s sites primary function, namely the provision of prestigious building materials and eastern goods, lay at the very heart of the Empire, feeding both imperial vanity and elite display. While the soldiers and workmen at these sites undoubtedly experienced life in the desert as difficult and harsh, neither the physical distance from the Nile valley, nor the environmental constraints of the desert environment prevented them from maintaining contact with their families and friends or from eating the types of foods that they had become accustomed to in other parts of Egypt or the Mediterranean. Their foodways were an integral part of their identities, and the changes observed in the archaeobotanical assemblages demonstrate changes in these identities over time, with a strongly Roman emphasis during the heyday of the Empire and of the exploitation of the Eastern Desert (later 1st to early 3rd centuries), more integration with local Egyptian peoples (both Nile valley and desert nomads) during the later Roman period, after the political and economic turmoil of the 3rd century, and a more starkly observed transformation by the medieval Islamic period, when Egypt had become part of the Islamic world. This highlights that the Eastern Desert was not remote in any real sense; instead, it was closely linked to and affected by the social, economic and political fortunes of the Empire and of later political entities. Our results emphasise the importance of food in the dynamics of geopolitics.

Acknowledgements

45Marijke Van der Veen and Charlène Bouchaud would like to thank Jean-Pierre Brun and the Collège de France for inviting them to the conference in Paris where this paper was presented by the first author. Marijke Van der Veen is grateful to Charlène Bouchaud, René Cappers and Claire Newton for making available their unpublished data and for collaborating on this paper. Marijke Van der Veen would like to thank Valerie Maxfield and David Peacock for inviting her to join their projects in the Eastern Desert, while Charlène Bouchaud and Claire Newton thank Hélène Cuvigny for inviting them to participate in the French archaeological missions. Thanks are also due to Wilfried Van Rengen and Hélène Cuvigny for information regarding unpublished ostraka concerning vegetable gardens, to both Hélène Cuvigny and Valerie Maxfield for comments on an earlier draft, to Jacob Morales for the photographs, and to Seán Goddard for drawing up Figure 1.

Bibliographie

Des DOI sont automatiquement ajoutés aux références bibliographiques par Bilbo, l’outil d’annotation bibliographique d’OpenEdition. Ces références bibliographiques peuvent être téléchargées dans les formats APA, Chicago et MLA.

André J. 1981. L’alimentation et la cuisine à Rome, 2nd ed., Paris, Les Belles Lettres.

10.4000/books.lesbelleslettres.6724 :

Bagnall R.S. 1986. “Papyri and ostraca from Quseir al-Qadim”. Bulletin of the American Society of Papyrologists, 23, pp. 1-60.

Bagnall R.S., Helms C. and Verhoogt A.M.F.W. 2000. Documents from Berenike, Volume 1. Greek Ostraka from the 1996-1998 Seasons. Brussels, Fondation Égyptologique Reine Elisabeth.

Bender L. 2018. “Textiles from Mons Claudianus, ‘Abu Sha’ar and other Roman Sites in the Eastern Desert”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5234.

10.4000/1895.2 :

Bingen J.A., Bülow-Jacobsen A., Cockle W.E.H., Cuvigny H., Rubinstein L., Van Rengen W. 1992. Mons Claudianus. Ostraca Graeca et Latina I (O. Claud. 1-190), Cairo IFAO. Documents de Fouilles 29.

Bingen J.A., Bülow-Jacobsen A., Cockle W.E.H., Cuvigny H., Rubinstein L., Van Rengen W. 1997. Mons Claudianus. Ostraca Graeca et Latina II (O. Claud. 191-416), Cairo, IFAO. Documents de Fouilles 32.

Bingen J. 1997. “Lettres privée de Raima (255-278)” and “Lettres privées (279-303)”. In Mons Claudianus. Ostraca Graeca et Latina II (O. Claud. 191-416). Bingen J., Bülow-Jacobsen A., Cockle W.E.H., Cuvigny H., Rubinstein L., Van Rengen W., Cairo, IFAO. Documents de Fouilles 32, pp. 81-140.

Blue L. 2018. “Port of Myos Hormos and its Relation to Indo-Roman Trade”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5235.

10.4000/1895.2 :

Bouchaud C. unpublished. The botanical remains from Xèron Pelagos.

Bouchaud C. and Redon B. 2017. “Heating the baths during the Ptolemaic and Roman periods in Egypt: Comparing the archaeobotanical and textual data”. In Redon B., The Collective Baths in Egypt 2, Cairo, IFAO, Etudes urbaines 10, 323-349.

Bouchaud C., Newton C., Van der Veen M. and Vermeeren C. 2018. “Fuel and Wood Supplies in the Eastern Desert of Egypt”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5237.

10.4000/1895.2 :

Brun J.-P. 2011. “Les meules”. In Didymoi. Une garnison romaine dans le désert Oriental d’Égypte, Volume 1. Les fouilles et le matériel (Praesidia du désert de Bérénice IV). Cuvigny H. (ed.), Cairo, IFAO, pp. 243-249.

Brun J.-P. 2014-2015. “Le commerce entre l’Empire romain, l’Arabie et l’Inde à la lumière des fouilles archéologiques dans le désert oriental d’Égypte (2e partie)”. Annuaire des cours et travaux du Collège de France, 114, pp. 471-502.

Brun J-P. 2018. “Chronology of the Forts of the Routes to Myos Hormos and Berenike during the Graeco-Roman Period”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5239.

10.4000/1895.2 :

Bülow-Jacobsen A. 1992. “The private letters”. In Mons Claudianus. Ostraca Graeca et Latina I (O. Claud. 1-190). Bingen J., Bülow-Jacobsen A., Cockle W.E.H., Cuvigny H., Rubinstein L., Van Rengen W., Cairo, IFAO. Documents de Fouilles 29, pp. 123-159.

Bülow-Jacobsen A. 1997. “The correspondence of Dioscorus and others (224-242)”. In Mons Claudianus. Ostraca Graeca et Latina II (O. Claud. 191-416). Bingen J., Bülow-Jacobsen A., Cockle W.E.H., Cuvigny H., Rubinstein L., Van Rengen W., Cairo, IFAO. Documents de Fouilles 32, pp. 4368.

Bülow-Jacobsen A. 2003. “The traffic on the roads and the provisioning of the stations”. In La Route de Myos Hormos : l’Armée Romaine dans le désert Oriental d’Égypte. Volume 2. Cuvigny H. (ed.) Cairo, IFAO, Fouilles de l'IFAO 48, pp. 399-426.

Cappers R.T.J. 1998. “Botanical contribution to the analysis of subsistence at Berenike”. In Life on the Fringe. Living in the Southern Egyptian Deserts during the Roman and early-Byzantine Periods. Kaper O.E. (ed.), Leiden, Centre of Non-Western Studies, pp. 75-86.

Cappers R.T.J. 2003. “Exotic imports of the Roman Empire: an exploratory study of potential vegetal products from Asia”. In Food, Fuels and Fields: Progress in African Archaeobotany. Neumann K., Butler A. and Kahlheber S. (eds.), Köln, Heinrich Barth Institut (Africa Praehistorica 15), pp. 197-206.

Cappers R.T.J. 2005. “Onderzoek aan plantenresten uit Grieks-Romeins Karanis (Fayum, Egypte): een doorstart na 70 jaar”. Paleo-Aktueel 16, pp. 89-95.

Cappers R.T.J. 2006. Roman Food Prints at Berenike: Archaeobotanical Evidence of Subsistence and Trade in the Eastern Desert of Egypt. Los Angeles, Cotsen Institute of Archaeology, Monograph 55.

Cappers R.T.J. 2016. “Modelling shifts in cereal cultivation in Egypt from the start of agriculture until modern times”. In News from the Past: Progress in African Archaeobotany. Proceedings of the 7th International Workshop on African Archaeobotany in Vienna, 2-5 July 2012. Thanheiser U. (ed.), Groningen, Barkhuis, pp. 27‑36.

10.2307/j.ctt20p56fr :

Cappers R.T.J., Sikking J.C., Darnell J.C. and Darnell D. 2007. “Food supply along the Theban desert roads (Egypt): the Gebel Roma, Wâdi el-Hôl and Gebel Qarn el-Gir caravansary deposits”. In Fields of Change: Progress in African Archaeobotany. Cappers R.T.J. (ed.), Groningen, Barkhuis/Groningen University Library, pp. 127-138.

Clapham A.J. and Rowley-Conwy P. 2007. “New discoveries at Qasr Ibrim, Lower Nubia”. In Fields of Change: Progress in African Archaeobotany. Cappers R.T.J. (ed.), Groningen, Barkhuis/Groningen University Library, pp. 157-164.

Clapham A.J. and Rowley-Conwy P. 2009. “The archaeobotany of cotton (Gossypium sp. L.) in Egypt and Nubia”. In From Foragers to Farmers, Papers in Honour of Gordon C. Hillman. Farbairn A. and Weiss E. (eds.), Oxford, Oxbow Books, pp. 244-253.

Cuvigny H. 1996. “The amount of wages paid to the quarry workers at Mons Claudianus”. Journal of Roman Studies, 86, pp. 139145.

10.2307/300426 :

Cuvigny H. 1998. “Kainè, ville nouvelle: une expérience de regroupement familial au iis. è. chr.”. In Life on the Fringe. Living in the Southern Egyptian Deserts during the Roman and early-Byzantine Periods. O.E. Kaper (ed.), Leiden, CNWS, pp. 87-94.

Cuvigny H. 2000. Mons Claudianus. Ostraca Graeca et Latina III. Les reçus pour avances à la familia (O. Claud. 417-631), Cairo, IFAO, Document de Fouilles 38.

Cuvigny H. 2003a. “Le fonctionnement du réseau”. In La route de Myos Hormos. L’armée romaine dans le désert Oriental d’Égypte. Praesidia du désert de Bérénice I, 2. H. Cuvigny (ed.), Cairo, pp. 295-359.

Cuvigny H. 2003b. “La société civile des praesidia”. In La route de Myos Hormos. L’armée romaine dans le désert Oriental d’Égypte. Praesidia du désert de Bérénice I, 2. H. Cuvigny (ed.), Cairo, pp. 361-397.

Cuvigny H. (ed.) 2003c. La route de Myos Hormos. L’armée romaine dans le désert Oriental d’Égypte. Praesidia du désert de Bérénice I, 2. Cairo, Fouilles de l’IFAO, 48.

Cuvigny H. 2005. “L'organigramme du personnel d'une carrière impériale d'après un ostracon du Mons Claudianus”, Chiron, 35, pp. 309-353.

Cuvigny H. 2007. “Les noms du chou dans les ostraca grecs du désert Oriental d’Égypte κραμβη, κραμβıotν, καυλıoυ”. Bulletin de l'IFAO 107, pp. 382-383.

Cuvigny H. (ed.) 2011. Didymoi. Une garnison romaine dans le désert Oriental d’Egypte, Volume 1. Les fouilles et le matériel, (Praesidia du désert de Bérénice IV), Cairo, IFAO, 2011.

Cuvigny H. (ed.) 2012. Didymoi. Une garnison romaine dans le désert Oriental d’Égypte, Volume 2. Les textes (Praesidia du désert de Bérénice IV), Cairo, IFAO.

Cuvigny H. 2014. “Papyrological evidence on ‘barbarians’ in the Egyptian Eastern Desert”. In Inside and Out. Interactions between Rome and the Peoples on the Arabian and Egyptian Frontiers in Late Antiquity. J.H.F. Dijkstra and G. Fisher (eds.), Leuven, Peeters, pp. 165-198.

Dalby A. 2000. Empire of Pleasures: Luxury and Indulgence in the Roman World, London, Routledge.

Dioscorides. 2000. De Materia Medica II.117. Translation by Osbaldeston T.A. and Wood R.P.A., Johannesburg, Ibidis Press.

Fadl M.A. 2013. “Comparison between archaeobotany of inland and coastal sites in the Eastern Desert of Egypt in 300 B.C.-700 A.D.”. International Research Journal of Plant Science 4(5), pp. 117-132.

el-Hadidi M.N. and Amer W.A. 1996. “The palaeoethnobotany of Abu Sha’ar site (400-700 A.D.), Red Sea, Egypt. I. Food plants and related industries”. Taeckholmia 16, pp. 31-42.

el-Hadidi M.N. and el-Fayoumi H.H. 1996. “Catalogue of the archaeobotanical specimens in Cairo University Herbarium. I. Abu Sha’ar site, Red Sea coast of Egypt, season 1990”. Taeckholmia 16, pp. 11-30.

el-Hadidi M.N., Amer W.A., and Waly N.M. 1997. “Catalogue of the archaeobotanical specimens in Cairo University Herbarium. II. Abu Sha’ar site, Red Sea coast of Egypt, season 1991”. Taeckholmia 17, pp. 47-60.

Fuller D.Q. 2014. “Agriculture innovation and state collapse in Meroitic Nubia”. In Archaeology of African Plant Use, Walnut Creek. C.J. Stevens, S. Nixon, M.A. Murray & D. Fuller (eds.), Left Coast Press, pp. 165-177.

Hamilton-Dyer S. 2001. “The faunal remains”. In Survey and Excavation at Mons Claudianus, 1987-1993. Volume II: The Excavations: Part 1. V.A. Maxfield and D.P.S. Peacock (eds.), Cairo, IFAO Documents de Fouilles, 43, pp. 251-311.

Hamilton-Dyer S. 2007. “Faunal remains”. In The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998. Volume 2 : The Excavations. V.A. Maxfield and D.P.S. Peacock (eds.), London, Egyptian Exploration Society, pp. 143-175.

Hamilton-Dyer S. 2011. “Faunal remains”. In Myos Hormos-Quseir al-Qadim: Roman and Islamic ports on the Red Sea, Volume 2: Finds from the excavations 1999-2003. D.P.S. Peacock and L. Blue (eds.), Oxford, Archaeopress (BAR. International Series 2286; University of Southampton. Series in Archaeology, 6), pp. 245-288.

Ikram S. 2014. “Zooarchaeological remains”. In Bir Umm Fawakhir 3. C. Meyer, Chicago, The Oriental Institute of the University of Chicago, Oriental Institute Publications 141, pp. 91-96.

Konen H. 1999. “Reis im Imperium Romanum: Bemerkungen zu seinem Anbau und seiner Stellung als Bedarfs- und Handelsartikel in der Römischer Kaiserzeit”. Münstersche Beiträge zur Antiken Handelsgeschichte 18, pp. 23-47.

Leguilloux M. 2003. “Les animaux et l’alimentation d’après la faune : les restes de l’alimentation carnée des fortins de Krokodilô et Maximianon”. In La route de Myos Hormos. L’armée romaine dans le désert Oriental d’Égypte, vol. 2. (Praesidia du désert de Bérénice I). H. Cuvigny (ed.), Cairo. IFAO, pp. 549 588.

Leguilloux M. 2011. “Les animaux à Didymoi d’après les restes fauniques du dépotoir extérieur”. In Didymoi. Une garnison romaine dans le désert Oriental d’Egypte (Praesidia du désert de Bérénice IV). Volume 1. Les fouilles et le matériel. H. Cuvigny (ed.), Cairo, pp. 167-204.

Leguilloux M. 2018. “The Exploitation of Animals in the Roman Praesidia on the Routes to Myos Hormos and to Berenike: on Food, Transport and Craftsmanship”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5245.

10.4000/1895.2 :

Maxfield V. 1996. “The Eastern Desert forts and the army in Egypt during the principate”. In Archaeological Research in Egypt. D.M. Bailey (ed.), Portsmouth (RI), Journal of Roman Archaeology Supplement 19, pp. 9-19.

Maxfield V. 2001a. “Lykabettus Village”. In The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998. Volume 1: Topography and Quarries. V.A. Maxfield and D.P.S. Peacock (eds.), London, Egypt Exploration Society, pp. 110-129.

Maxfield V. 2001b. “Badia”. In The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998. Volume 1: Topography and Quarries. V.A. Maxfield and D.P.S. Peacock (eds.), London, Egypt Exploration Society, pp. 215-239.

Maxfield V. 2001c. “Stone quarrying in the Eastern Desert with particular reference to Mons Claudianus and Mons Porphyrites”. In Economies beyond Agriculture in the Classical World. D.J. Mattingly and J. Salmon (eds), London, Routledge, pp 143-170.

Maxfield V. 2007. “Excavations at Badia”. In The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998. Volume 2: The Excavations. V.A. Maxfield and D.P.S. Peacock (eds.), London, Egyptian Exploration Society, 2007, pp. 25-81.

Maxfield V.A. and Peacock D.P.S. (eds.) 2001a. Survey and Excavation at Mons Claudianus, 1987-1993. Volume II: The Excavations: Part 1, Cairo (IFAO Documents de Fouilles, 43).

Maxfield V.A. and Peacock D.P.S. (eds.) 2001b. The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998. Volume 1: Topography and Quarries. London, Egypt Exploration Society.

Meyer C. 2014. Bir Umm Fawakhir 3. Chicago, The Oriental Institute of the University of Chicago, Oriental Institute Publications 141.

Murray M.A. 2000a. “Fruits, vegetables, pulses and condiments”. In Ancient Egyptian Materials and Technology. Nicholson P.T. and Shaw I. (eds.), Cambridge, Cambridge University Press, pp. 609-655.

Murray M.A. 2000b. “Cereal production and processing”. In Ancient Egyptian Materials and Technology. Nicholson P.T. and Shaw I. (eds.). Cambridge, Cambridge University Press, pp. 505-536.

Newton C. and Clapham A. in press. “Egyptian oasis crop productions and potential exports from the Pharaonic Period to Late Roman times: a first archaeobotanical review”. In La céramique du désert Occidental de la fin du Néolithique à l’époque médiévale. La Marmarique et les oasis de Bahariya, Dakhla et Kharga. S. Marchand (ed.), Cahiers de la Céramique Égyptienne 11.

Newton C., Gonon T., Wuttmann M. 2005. “Un jardin d’oasis d’époque romaine à ‘Ayn-Manâwir (Kharga, Égypte)”. Bulletin d’IFAO 105, pp. 167-196.

Newton C. Unpublished. “The plant remains from Domitianè/Kainè Latomia”. In J.-P. Brun (ed.) (report 2005).

Newton C. Unpublished. Rapport de mission d’étude archéobotanique; sites du Désert oriental. (report 2004).

Palmer S.A., Clapham A.J., Rose R., Freitas F.O., Owen B.D., Beresford-Jones D., Moore J.D., Kitchen J.L. and Allaby R.G. 2012. “Archaeogenomic evidence of punctuated genome evolution in Gossypium”. Molecular Biology and Evolution 29(8), 2012, 2031-2038. doi:10.1093/molbev/mss070.

10.1093/molbev/mss070 :

Peacock D.P.S. 1992. Rome in the Desert: A Symbol of Power. Southampton, Inaugural Lecture Southampton University.

Peacock D.P.S. and Blue L. 2006. Myos Hormos-Quseir al-Qadim, Roman and Islamic Ports on the Red Sea, Volume 1: The Survey and Report on the Excavations, Oxford, Oxbow Books.

Peacock D.P.S. and Blue L. (eds.) 2011. Myos Hormos-Quseir Al-Qadim: Roman and Islamic ports on the Red Sea, Volume 2: Finds from the excavations 1999-2003, Oxford, Archaeopress (BAR. International Series 2286; University of Southampton. Series in Archaeology, 6).

Peacock D.P.S. and Maxfield V.A. (eds.) 2007. The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998. Volume 2: The Excavations, London, Egyptian Exploration Society.

Peacock D.P.S. and Maxfield V.A. (eds.) 1997. Survey and Excavations at Mons Claudianus 1987-1993. Vol. I. Topography & Quarries. Cairo, IFAO, Documents de Fouilles 37.

Petersen T. 1980. “The Arab influence on western European cooking”. Journal of Medieval History 6, pp. 317-340.

Pliny: Natural History. Trans. H. Rackham (Volume 4 1952) and W.H.S. Jones (Volume 6 1962). Loeb Classical Library. London/Cambridge MA, Heinemann/Harvard University Press.

Redon B. 2018. “The Control of the Eastern Desert by the Ptolemies: New Archaeological Data”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5249.

10.4000/1895.2 :

Rowley-Conwy P. 1994. “Dung, dirt and deposits: site formation under conditions of near-perfect preservation at Qasr Ibrim, Egyptian Nubia”. In Whither Environmental Archaeology. R. Luff and P. Rowley-Conwy (eds.), Oxford, Oxbow Books, pp. 25-32.

Ruffing K. 1993. “Das Nikanor-Archiv und der Römische Süd- und Osthandel”. Münstersche Beiträge z. Antiken Handelsgeschichte 12(2), pp. 1-26.

Sidebotham S.E. 1986. Roman Economic Policy in the Erythrea Thalassa 30 BC- AD 217, Leiden.

Sidebotham S.E. 1993. “University of Delaware Archaeological Project at Abu Sha’ar: the 1992 season”. Newsletter of the American Research Center in Egypt 161/162, pp. 1-9.

Sidebotham S.E. 1994. “Preliminary report on the 1990-1991 field seasons of fieldwork at Abu Sha’ar (Red Sea Coast)”. Journal of the American Research Centre in Egypt 31, pp. 133-158.

Sidebotham S.E. 2011 Berenike and the Ancient Maritime Spice Route, Berkeley, University of California Press.

10.1525/9780520948389 :

Sidebotham S.E. 2018. “Overview of Fieldwork at Berenike 1994-2015”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5250.

10.4000/1895.2 :

Sidebotham S.E. Hense M., Nouwens H.M. 2008. The Red Land. An Illustrated Archaeology of Egypt’s Eastern Desert, Cairo, New York, The American University in Cairo Press.

Smith W. 2003. Archaeobotanical Investigations of Agriculture at Late Antique Kom El-Nana (Tell El-Amarna). London, Egypt Exploration Society.

Smith W. 2014. “The floral remains (Chapter 6)”. In Bir Umm Fawakhir 3, Excavations 1999–2001. C. Meyer, Chicago, The Oriental Institute of the University of Chicago, pp. 97-110.

10.7560/746473 :

Strabo, The Geography, Trans. H.L. Jones 1969. The Loeb Classical Library. Cambridge MA/London, Harvard University Press/Heinemann.

Thanheiser U. 2002. “Roman agriculture and gardening in Egypt as seen from Kellis”. In Dakhleh Oasis Project: preliminary reports on the 1994-1995 to 1998-1999 field seasons. C.A. Hope and G.E. Bowen (eds.), Oxford, Oxbow Books, pp. 299-310.

Thanheiser U. and König C. 2008. “Plant remains from habitation areas at Kellis: some considerations concerning their accumulation”. In The Oasis Papers 2: Proceedings of the Second International Conference of the Dakhleh Oasis Project, Vol. 2. M.F. Wiseman (ed.), Oxford, Oxbow Books, pp. 141-150.

Thanheiser U. and Walter J. 2015. Forthcoming but available online. “Plant use in a Romano-Egyptian household in the third century CE”. In Amheida II: A late Romano-Egyptian house in the Dakhleh oasis. Amheida House B2. A.L. Boozer (ed.), New York: The Institute for the Study of the Ancient World, pp. 375‑392, URL: http://dlib.nyu.edu/awdl/isaw/amheida-ii-house-b2/

Thanheiser U., Kahlheber S., Dupras T., 2016. “Pearl millet, Pennisetum glaucum (L.) R.Br. ssp. glaucum. In the Dakhleh Oasis, Egypt”. In News from the Past: Progress in African Archaeobotany. Proceedings of the 7th International Workshop on African Archaeobotany in Vienna, 2-5 July 2012. U. Thanheiser (ed.), Groningen, Barkhuis, pp. 115‑125.

Tengberg M. 2011. “L’acquisition et l’utilisation des produits végétaux à Didymoi”. In Didymoi. Une garnison romaine dans le désert Oriental d’Egypte (Praesidia du désert de Bérénice IV). Volume 1. Les fouilles et le matériel. H. Cuvigny (ed.), Cairo, pp. 205-214.

Tomber R. 1996. “Provisioning the desert: pottery supply to Mons Claudianus”. In Archaeological Research in Roman Egypt. D.M. Bailey (ed.) Journal of Roman Archaeology Supplement S19, pp. 39-49.

Tomber R. 2006. “The pottery”. In Survey and Excavations at Mons Claudianus 1987-1993. Volume 3: Ceramic Vessels and Related Objects. V. Maxfield and D. Peacock (eds.), Cairo, IFAO, Fouilles de l’IFAO 54, pp. 3-236.

Tomber R. 2008. Indo-Roman Trade. From Pots to Pepper. London, Duckworth.

Tomber R. 2011. “Pots with writing”. In Myos Hormos-Quseir Al-Qadim: Roman and Islamic ports on the Red Sea, Volume 2: Finds from the excavations 1999-2003. D.P.S. Peacock and L. Blue (eds.), Oxford, Archaeopress, 2011 (BAR. International Series 2286; University of Southampton. Series in Archaeology, 6), pp. 5-10.

Tomber R. 2018. “Quarries, Ports and Praesidia: Supply and Exchange in the Eastern Desert”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5251.

10.4000/1895.2 :

Van der Veen M. 1998. “Gardens in the Desert”. In Life on the Fringe: Living in the Southern Egyptian Deserts during the Roman and early-Byzantine Periods. O.E. Kaper (ed.), Leiden, CNWS, pp. 221-242.

Van der Veen M. 1999. “The economic value of chaff and straw in arid and temperate zones”. Vegetation History and Archaeobotany 8, pp. 211-224.

10.1007/BF02342721 :

Van der Veen M. 2001. “The botanical evidence”. In Survey and Excavation at Mons Claudianus, 1987-1993. Volume 2, Part 1: The Excavation. V. Maxfield and D. Peacock (eds.), Cairo, IFAO de Caire; Documents des Fouilles 43, pp. 174-247.

Van der Veen M. 2003. “When is food a luxury?”. World Archaeology 34(3), pp. 405-427. [doi: 10.1080/0043824021000026422]

10.1080/0043824021000026422 :

Van der Veen M. 2007. “Formation processes of desiccated and carbonised plant remains - the identification of routine practice”, Journal of Archaeological Science 34, pp. 968-990. [DOI 10.1016/j.jas.2006.09.007]

10.1016/j.jas.2006.09.007 :

Van der Veen M. 2011. Consumption, Trade and Innovation: Exploring the Botanical Remains from the Roman and Islamic Ports at Quseir al-Qadim, Egypt, Frankfurt, Journal of African Archaeology, Monograph 6.

Van der Veen M. and Hamilton-Dyer S. 1998. “A life of luxury in the desert? The food and fodder supply to Mons Claudianus”. Journal of Roman Archaeology 11, pp. 101-116.

Van der Veen M., Morales J. 2015. “The Roman and Islamic spice trade: new archaeological evidence”. Journal of Ethnopharmacology 67, pp. 54-63. [DOI 10.1016/j.jep.2014.09.036]

10.1016/j.jep.2014.09.036 :

Van der Veen M., Morales J. 2017. “Food globalization and the Red Sea: new evidence from the ancient ports at Quseir al-Qadim, Egypt”. In Human Interaction with the Environment in the Red Sea. D.A. Agius, A. Williams, E. Khalil and E. Scerri (eds), Leiden, Brill, pp. 256-291. [DOI 10.1163/9789004330825_013]

10.1163/9789004330825_013 :

Van der Veen M. and Tabinor H. 2007. “Food, fodder and fuel at Mons Porphyrites: the botanical evidence”. In The Roman Imperial Quarries. Survey and Excavation at Mons Porphyrites 1994-1998. Volume 2: The Excavations. V. Maxfield and D. Peacock (eds.), London, Egypt Exploration Society, pp. 83-142.

Van Neer W. 1997. “Archaeozoological data on the food provisioning of Roman settlements in the Eastern Desert of Egypt”. Archaeozoologica 9, pp. 137-154.

Van Neer W. and Ervynck A.M.H. 1999. “Faunal remains from Shenshef and Kalalat”. In Berenike 1997. Report of the 1997. Excavations at Berenike and the Survey of the Egyptian Eastern Desert, including Excavations at Shenshef. S.E. Sidebotham and W.Z. Wendrich (eds.), Leiden, pp. 431-444.

Van Rengen W. 1992. “Les laissez-passez (48-820)”. In Mons Claudianus. Ostraca Graeca et Latina I (O. Claud. 1-190). J.A. Bingen, A. Bülow-Jacobsen, W.E.H. Cockle, H. Cuvigny, L. Rubinstein, W. Van Rengen, Cairo, IFAO. Documents de Fouilles 29, pp. 57-74.

Warmington E.H. 1928. The Commerce between the Roman Empire and India. Cambridge, Cambridge University Press. Reprint 1995 (New Delhi: Munshiram Manoharlal Publishers).

Wendrich W.Z., Tomber R.S., Sidebotham S.E., Harrell J.A., Cappers R.T.J. and Bagnall R.S. 2003. “Berenike crossroads: the integration of information”. Journal of the Economic and Social History of the Orient 46, 1, pp. 46-87.

Whitcomb D.S. and Johnson J.H. 1979. Quseir Al-Qadim 1978: Preliminary Report. Cairo, The American Research Center in Egypt.

Whitcomb, D.S. and Johnson, J.H. 1982. Quseir Al-Qadim 1980: Preliminary Report. Malibu, Undeena Publications

Wild J.-P. and Wild F. 2018. “Textile Contrasts at Berenike”. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5254.

10.4000/1895.2 :

Zahran M.A. and Willis A.J. 1992. The Vegetation of Egypt, London, Chapman and Hall.

10.1007/978-94-015-8066-3 :

Annexe

Appendix 1: Summary information of Eastern Desert sites with archaeobotanical remains

Abu Sha’ar (AS)
- Late Roman fort on the Red Sea coast, 20km north of modern Hurghada. Occupation: late 4th-early 5th c. AD. After a period of disuse (length not known) re-occupied by a small Christian community, length of occupation not known, but possibly going into the 6th c. AD. Excavations 1987-1991. The archaeobotanical data derive from Late Roman rubbish dumps and concern mostly hand-picked material, but sampling details are not published.
- Archaeology: Sidebotham 1993, 1995, 2011. Archaeobotany: el-Hadidi and Amer 1996, el-Hadidi and el-Fayoumi 1996, el-Hadidi et al. 1997; Fadl 2013.

Badia (BA)
- A way-station (praesidium) on the route from Qena to the quarries at Mons Porphyrites, consisting of a fortified settlement with large animal lines, and a well close by. Chronology: late 1st-5th c. AD. While functioning as a way-station during the early Roman period, it also formed logistical support for the Mons Porphyrites quarries during the Late Roman (4th-5th c. AD) period. Excavations: 1994-1998.
- Archaeology: Maxfield and Peacock 2001b, Peacock and Maxfield 2007. Archaeobotany: Van der Veen and Tabinor 2007. Archaeozoology: Hamilton-Dyer 2007.

Berenike (BE)
- A port on the Red Sea coast, on the far southern border of Egypt. Large settlement, temple and shrines, as well as storage buildings. Occupation: founded in the Ptolemaic period (3rd c. BC), heyday during the 1st-2nd centuries AD, a hiatus or at least much reduced activity during the 3rd century, activities resumed mid-4th-mid-5th c. AD, and abandoned before the mid-6th c. AD. Excavations 1994-2001, and again more recently.
- Archaeology: Sidebotham 2008, 2011, 2018; Wendrich et al. 2003. Archaeobotany: Cappers 2006, see also Cappers 2003 and Cappers in Wendrich et al. 2003. Archaeozoology: Van Neer 1997.

Bir Umm Fawakhir (BF)
- Gold mines and village, just north of the Coptos to Myos Hormos road. Occupation 5th-6th c. AD, though earlier exploitation is known. There was also a way-station (Persou) during the early Roman period. Excavations: 1992-1999.
- Archaeology: Meyer 2014. Archaeobotany: Smith 2014. Archaeozoology: Ikram 2014.

Didymoi – Kasm al-Menih (DI)
- Way-station (praesidium) on the Coptos to Berenike road. Occupied early 2nd-mid/late-3rd c. AD (abandonment ca. AD 269/271, Brun 2018. Excavations 1998-2000.
- Archaeology: Cuvigny 2011. Archaeobotany: Tengberg 2011. Archaeozoology: Leguilloux 2011.

Kainè Latomia/Domitianè – Umm Balad (KL)
- A small quarry and settlement close to Mons Porphyrites and administered from there. Initially called Domitianè, but renamed Kainè Latomia after Domitian’s fall from grace. Occupation late 1st-early 2nd c. AD, and again briefly during the mid-2nd c. AD. Excavations 2002-2003.
- Archaeology: Brun in preparation. Archaeobotany: Newton in preparation, unpublished report 2005. Archaeozoology: Leguilloux in preparation.

Kusayr - Qusayr al-Qadim (KU)
- Reoccupation of the Roman port at Qusayr al-Qadim (see under Myos Hormos), 8km north of modern Qusayr. During this time period the port is called Kusayr. Chronology: main phase of occupation 11th-13th c. AD, but continuing on a more reduced scale during the 14th-15th c. AD. Excavations: 1999-2003 (and previously 1978-1982, Whitcomb and Johnson 1979, 1982).
- Archaeology: Peacock and Blue 2006, 2011; see also Blue 2018. For radio-carbon dates, see Van der Veen 2011. Archaeobotany: Van der Veen 2011; see also Van der Veen and Morales 2015, 2017. Archaeozoology: Hamilton-Dyer 2011.

Maximianon – al-Zarqā’ (MA)
- Way-station (praesidium) on the Coptos to Myos Hormos route. Fortified settlement with well in the centre. Occupation: mid 1st and 2nd c. AD. Excavations 1993-1994.
- Archaeology: Cuvigny 2003 (2 volumes). Archaeobotany: Newton unpublished report 2004. Archaeozoology: Leguilloux 2003.

Mons Claudianus (MC)
- A large quarry complex for the extraction of grey granodiorite, used in major imperial building projects in Rome, e.g. columns at the Pantheon and Trajan’s Forum, as well as baths and basins. A fortified settlement, containing accommodation, communal kitchen, bath house and temple, as well as animal lines, granary and well, and a quarry field of some 750ha and 130 quarries. Occupation during mid/late 1st-early-3rd c. AD. Excavations 1987-1993.
- Archaeology: Peacock and Maxfield 1997, Maxfield and Peacock 2001a; Maxfield 2001. Archaeobotany: Van der Veen 2001, see also Van der Veen 1998 and Van der Veen and Hamilton-Dyer 1998. Archaeozoology: Hamilton-Dyer 2001.

Mons Porphyrites (MP)
- A major quarry complex for the extraction of purple and (to a much lesser extent) black porphyry, used in imperial building projects (largely in Rome and Constantinople), statuary, sarcophagi and basins. Large fort with associated settlement in the main wâdi, but with additional small, unenclosed settlements (villages) located near the quarries, as these are located higher up the mountains. Occupation during early-1st-early-3rd c. AD and again during the 4th-early 5th c. AD. During this later period most activities were centred on the Lykabettus quarry and village, with links to Badia (see there). Excavations 1994-1998.
- Archaeology: Maxfield and Peacock 2001b, Peacock and Maxfield 2007. Archaeobotany: Van der Veen and Tabinor 2007. Archaeozoology: Hamilton-Dyer 2007.

Myos Hormos – Qusayr al-Qadim (MH)
- A port on the Red Sea coast of Egypt, just north of where the Wâdi al-Hammâmât reaches the coast; the site is today referred to as Qusayr al-Qadim, just 8km north of modern Qusayr. One of two main ports for the trade with India during the Roman period. The site is located on an old coral reef just north and east of a now silted lagoon. Occupied early 1st-early/mid3rd c. AD, though probably founded earlier, during the Ptolemaic period. Re-occupied during the medieval Islamic period, then known as Kusayr, see there. Excavations 1999-2003 (and previously 1978-1982, Whitcomb and Johnson 1979, 1982).
- Archaeology: Peacock and Blue 2006, 2011; see also Blue 2018. Archaeobotany: Van der Veen 2011; see also Van der Veen and Morales 2015, 2017. Archaeozoology: Hamilton-Dyer 2011.

Shenshef (SS)
- Large settlement just 21 km south-west of Berenike. Exact function not certain, but likely a satellite settlement of Berenike. Occupied 5th-6th c. AD. Excavations 1994-2001.
- Archaeology: Sidebotham 2011; Archaeobotany: Cappers 2006, Archaeozoology: Van Neer and Ervynck 1999.

Tiberianè – Barud (TI)
- Small quarry and settlement just 10km south-east of Mons Claudianus, and administered from there. Occupation mid-2nd century AD. Excavations: small trial excavation in 1992.
- Archaeology: Peacock and Maxfield 1997, Maxfield and Peacock 2001a. Archaeobotany: Van der Veen 2001. Archaeozoology: Hamilton-Dyer 2001.

Xeron Pelagos, Xeron – Jirf (XE)
- Way-station (praesidium) on the Coptos to Berenike road. Occupation late 1st-mid/late-3rd century AD (abandonment ca. AD 269/271, Brun 2018). Excavations 2010-2013.
- Archaeology: Brun 2014-2015. Archaeobotany: Bouchaud unpublished; see also Bouchaud and Redon 2017. Archaeozoology: Leguilloux unpublished.

Appendix 2

List of all plant foods recovered from the Eastern Desert sites discussed in the text (desiccated and charred remains), and total number of identifications. References: a. Cappers 2006; b. Van der Veen 2011; c. Van der Veen 2001; d. Van der Veen and Tabinor 2007; e. Newton in prep.; f. Newton unpublished; g. Bouchaud unpublished; h. Tengberg 2011; i. Fadl 2013, el-Hadidi and Amer 1996, el-Hadidi and el-Fayoumi 1996, el-Hadidi et al. 1997; j. Smith 2014. For site name abbreviations, see Table 1, plus: SS=Shenshef, AS=Abu Sha’ar, BF= Bi’r Umm Fawâkhir, and KU= Kusayr.

Image 100000000000049D00000347BF8FE8C792E17A5C.jpg
Image 100000000000049D00000348FA4AC34E05D110AE.jpg
Image 100000000000049D00000337CEC90C5D21A6BCEF.jpg

Notes de bas de page

1 We cannot completely rule out the possibility that some of the seeds of the cabbage genus (Brassica spp.) derive from wild cabbage plants growing naturally in the desert, while those of beet (Beta vulgaris) may derive from the wild form that is a common arable weed. For both, however, the frequency and contexts of the finds encourage us to regard them as derived from cultivated plants, especially as cabbage and beet are some of the most frequently mentioned vegetables in the ostraca (in the private letters).

Précédent Suivant

Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.