Version classiqueVersion mobile

Networked spaces

Caroline Durand
Julie Marchand
Bérangère Redon
et al.

Space and networks

The Red Sea model

Caroline Durand, Julie Marchand, Bérangère Redon et Pierre Schneider

Texte intégral

1The past recent years witnessed the emergence of excellent academic researches focusing on Indo-Mediterranean connectivity (Jaspert, Kolditz 2018), on Red Sea shipping (Agius 2019), on Indo-Roman trade (Tomber 2008; Cobb 2018; De Romanis 2020), on medieval Red Sea (Power 2012), not to mention the previous proceedings of the Red Sea conferences dedicated to connectivity issues (IV: Connected hinterlands, published by Blue et al. 2009; V: Navigated Spaces, connected places, published by Agius et al. 2012). They all provided very informed and erudite developments on these themes. This vitality results from the richness of our documentation which is continuously renewed and reassessed by researchers. Whatever the documents are (written and non-written), they still have much to say about the Mare Erythraeum – viz. the western Indian Ocean and its two gulfs – which formed a system of spatial links, a networked space. We hope this volume is the continuation of the previous publications.

2To give an overview of the topics, this book is organized in thematic chapters and brings together 33 articles. The first ones examine Portuguese sources (D. Couto) and European travelogues (R. Margariti), showing how valuable such documents are for those studying the Red Sea and its related spaces. The second part is devoted to archaeological and historical inquiries into ancient ports, from the Pharaonic ports of Ayn Sukhna and Wadi el‑Jarf (C. Somaglino, P. Tallet), to the Hellenistic port of Berenike (M. Woźniak), and to the Nabataean port of Aynuna (M. Gawlikowski). Then comes a study of Suez (the former al‑Qulzum), located in an intermediary position between Egypt and the Arabian Peninsula (M. Abd el‑Raziq, J. Marchand, C. Somaglino), followed by a study on Suakin (L. Smith et al.), on the Sudanese coast; both contain an analysis of the archaeological material unearthed in the two ports. Islands and insularity form the next section. Islands – here, the Farasan archipelago (S. Marion de Procé; C.J. Robin), and Failaka (M. Gelin et al.) – are very specific nodes which deserve particular attention. Their role as religious places (J. Bonnéric) is also analyzed. The following section introduces the role played by states and power. Two articles focus on the port of Suakin in modern times (S. Taha) and the Beja country in Fatimid times (D. Bramoullé) – both papers also deal with the insertion of hinterland tribes inside maritime network. Three articles are devoted to the Roman period, engaging in various inquiries: M. Cobb and T. Wilkinson address economic issues, J. Simmons discusses the risks of navigation, P.‑L. Gatier examines the relationship between locals and Roman power in the province of Arabia. The fifth section concentrates on how archaeological material casts light on the complexity of commercial networks. Some papers deal with natural resources (gold and hard stone: I. Goncalves; emeralds: J. Oller Guzmán; soft stone: J. Marchand; frankincense: S. Le Maguer‑Gillon); production and consumption are the topics of the others (J. Gates‑Foster; F. Lesguer). Connectivity and mobility form the core of the following part. Some authors reflect on the logistics of exploration, exploitation and trade in the Egyptian coastal area (Y. Gourdon; M. Crépy, B. Redon), and in northwestern Arabia (L. Nehmé). Educational practices also illustrate the connectivity of the internal network of forts in the Egyptian Eastern Desert under Roman rule (J. Lougovaya). In the same area, an Indian coin was unearthed, which adds a small but important piece of information to the Indian-Egyptian connections (S. Bhandare, H. Cuvigny, T. Faucher). The last section is devoted to recently explored areas whose importance in Red Sea networks can now be assessed properly: Somaliland (A. Fernández Fernández, A. González-Ruibal, J. de Torres; J. de Torres), Tigray (A. Benoist et al.; J. Loiseau et al.) and the Horn of Africa (L. Sernicola, C. Zazzaro; J. Phillips).

3The variety in themes and topics reviewed above comes with unity, in the sense that the chapters of this book share several key ideas on which we would like to expand now. First, the Red Sea was, and had always been – to the best of our knowledge – a space of networks. These networks of various scales were made of individual, familial, social, political, cultural, commercial ties, which by nature never stopped shifting and changing. On the other hand, they developed on the basis of relatively stable patterns and structures which we can detect and analyse. More precisely, even if climatic conditions have slightly changed, it can be said without exaggeration that the Red Sea environment (water resources, vegetation cover, sea level…) have not drastically changed over the last 5,000 years. As a consequence, such stable ecological conditions imposed invariable constraints to the networks that criss-crossed this region: water points in the desert, natural corridors to the sea, coves easy to accost and serving as ports, islands and straits acting as landmarks, stopover places and choke points, etc. In contrast, the human factor has evolved. It is then critical to see how – in the course of History – these stable conditions impacted the pattern of connections: how were they integrated into the networks? To what extent did they influence their evolution and their structure, and – since we are particularly dealing with space – their spatial organisation?

4Most articles show that peoples had an important and sophisticated knowledge of the Red Sea area, especially of its ecological constraints. Ancient Greco-Roman descriptions – the Periplus of the Erythraean Sea, Strabo, Pliny (see Robin; Simmons; Phillips, among others) –, as well as medieval and modern chronicles and geographical accounts like those of Nāṣir‑ī Khusraw, Ibn Jubayr (Bramoullé) or Afonso de Albuquerque (Couto), show a striking awareness of these remarkable points – sometimes more attention was paid to them than to local populations and resources. This practical and theoretical knowledge was acquired partly directly, partly through all kinds of intermediaries: long-distance sailors, soldiers, local fishermen, explorers, etc. For instance, the Portuguese official Afonso de Albuquerque (Couto), and the German naturalists Wilhelm Hemprich and Christian Ehrenberg (Margariti) were able to get an accurate knowledge of the Red Sea geography and ecology. It was made of their own field experience and information collected from locals, thanks to a network of trusted contacts dispatched throughout the harbours and small anchorages of the Red Sea.

5Information and expertise could be passed on to others in many ways. The famous story of Eudoxus of Cyzicus who learnt from a shipwrecked Indian how to sail the Indian Ocean using the monsoon winds points to the importance of orally-transmitted knowledge. We also hear of “mnemonics, rhymes, and songs, […] all tools of choice by pre-modern Indian Ocean sailors to remember particular headings and wind regimes” (Simmons). While we lack evidence for the most ancient times, there is no doubt that information networks had been working since Antiquity especially in order to minimize the risks of navigation in the Red Sea or to cross its adjacent deserts. This may explain the presence of guides on an elephant hunting expedition organised by a Ptolemaic king in the 3rd century BC (Crépy, Redon). An exceptional document presented by J. Lougovaya gives a good example of how empirical knowledge was shared, in the form of a poem describing fort after fort, well after well, the quality of the water encountered by Roman travellers on their way from the valley to the port of Berenike. The sharing of information between merchants is also well attested in the exchanges archived at the Geniza of Cairo which are discussed by D. Bramoullé. To sum up, the life of Red Sea (and Indian Ocean) networks depended on multiple levels of knowledge (practical, theoretical, empirical…) working together.

6This leads to a major point which is the importance of “local peoples”, in other words those by whom the Red Sea area had been settled since time immemorial – most of them were not organised in kingdoms. The local tribes’ names are not often provided by our sources and generally little is known about them: some lead a nomadic or semi-nomadic life (the Blemmyes, the Megabaroi, the Trogodytes…), other were settled close to the seashore where they would find their food (the Ichtyophagoi, or Fish Eaters). The point is that these long-overlooked peoples are now receiving more and more attention; as a result, the role they played – or not – in the regional networks is more and more recognized. Indeed, some of them were probably disconnected from the commercial flows and traffic routes and lead a totally autonomous life, such was probably the case of several “Ethiopian” tribes settled in the Red Sea’s hinterland (Diodorus of Sicily, 3, 24‑31). However, most “local peoples” had contacts and interactions with the others, meaning that they played various roles in the local and regional networks. For instance, it must be stressed that the supply chain which fuelled exchange networks functioned through the agency of the local tribes (for instance, tortoiseshell was provided by the Fish Eaters who were able to catch turtles). We also see – especially when dominant powers were decaying – “local peoples” having control of state infrastructures. For instance, the Gebel Zabara and Sikait emerald mines, which may have been in use as early as in the Ptolemaic period, were exploited by “Arabs” according to Strabo (17, 1, 45) at the beginning of the Roman presence in Egypt – this ethnic name may apply to tribes living between the Nile Valley and the seashore. In later times, after Romans withdrew from mining activities, they fell under the control of people related to the Blemmyes (whose presence on the site is evidenced by material remains characteristic of the Eastern Desert dwellers culture [Oller Guzmán]). Previously, the same Blemmyes had been incorporated into the Roman system of control of the Eastern Desert as demonstrated by H. Cuvigny with the example of bread distributions made to nomads in the fort of Xeron (Cuvigny 2014). Eventually comes the case of Somaliland. This previously little prospected area was almost an empty space on the maps displaying the Indian Ocean networks until 2015, when the Incipit‑CSIC project was launched. It has now been clearly established that the populations of the region participated in the Indo-Roman trade (Fernández Fernández, González‑Ruibal, de Torres). In later times, from the 13th to the 16th century, when Somaliland was controlled by Muslim kingdoms, the coast may be compared to “a complex world of coastal contact points, trade routes and urban sites which structured a sophisticated framework for trade” where local nomadic and sedentary populations met with merchants from outside the region interested in trade with the Horn of Africa (de Torres). To conclude, the agency of local peoples was critical in the sense that it truly supported the large-scale networks.

7In addition, their unrivalled knowledge of their own country was an asset which drew the attention of major polities. For example, the Sekhetiu, whose name means “the men of the fringes”, were employed by Amenemhat III in his expeditions to the Sinai. The Medjay, another nomadic population, were also recruited by the Egyptian power to patrol the desert, to the extent that in the New Kingdom period these troops were called “Medjay” (Goncalves). Later, the “Arabs of the desert” and the Blemmyes seem to have been partially involved in the road system developed by the Ptolemies in the Egyptian Eastern Desert, perhaps as a result of their knowledge of water resources and expertise in camel husbandry (Crépy, Redon). The Romans also relied on local peoples to secure their presence and mobility, as P.‑L. Gatier demonstrates in his study of the ethnos, (i.e. “people”) of the Thamudenians, in the province of Arabia: the latter built a temple in Ruwāfa under the reign of Marcus Aurelius and Lucius Verus, whose dedication discloses a “policy of integration into the provincial system”. This also occurred in other places: the reconstruction of the sanctuary of Wādī Matar on the Farasan archipelago – which belonged to the Sabaean sphere of influence (Robin) – during the 2nd century AD, at a time when Roman presence is clear, must be regarded as a similar policy of integration and cohabitation (Marion de Procé). R. Ast and J. Rądkowska have recently published two inscriptions that were presented at this Red Sea conference and published in Ast, Rądkowska 2020, one of which was made by a Blemmye translator, “For the preservation of our Lord Isemne, King of the tribe of Blemmyes” (ὑπὲρ σωτηρίας τοῦ κυρίου Ἰσεμνη βασιλέως ἔθνους Βλεμμύων) – the word ethnos recalls the Thamudenians’ dedication at Ruwāfa. In more recent times, the Beja and the tribe of the Banū al‑Kanz – local potentates from Aswan to the Red Sea – helped the Fatimids establish their power on the Red Sea coastal area (Bramoullé). They gained such importance that the Fatimids even gave them the right to levy customs duties in the ports under their control, at least until the 12th century. S. Taha also shows that the vitality of the port of Suakin during the 16th‑20th centuries depended as much on the Beja agency as on the international exchange networks connected to this place.

8Thus, this shared knowledge helped sail the Red Sea and cross the lands leading to it. It also helped choose the best stopover points, especially ports and harbours. Most articles insist on the qualities of the places serving as ports. For instance, the two ports of Ayn Sukhna and Wadi el‑Jarf presented by C. Somaglino and P. Tallet exemplify this point. Both were founded during the Old Kingdom to reach the Sinai and its mining resources (copper and turquoise); they were established in carefully selected coastal places, well supplied with drinking water and located close to mountainous foothills where galleries were dug to shelter the ships between two expeditions. As for the Roman sites on the Farasan archipelago, drinking water, cultivable land and building material were available nearby; in addition, as S. Marion de Procé demonstrates, higher places were given preference, because of the better visibility they provided. Let us also mention the ports of Aynuna (Gawlikowski) and al‑Quṣayr (Nehmé), both suitable candidates for the Leukè Komè of the classical sources, and the harbours of Aydhāb and Suakin, in Sudan. The former were ideally located at the outlet of major routes reaching the heart of the Arabian Peninsula and particularly the Nabataean area. The latter were two excellent starting points for those heading towards East Africa. Controlling well located places was often of crucial importance for states in terms of strategy and ideology, especially when the heart of the concerned states was located far from these nodes. This is perfectly illustrated by D. Bramoullé’s study, showing that the possession of Aydhāb was essential for the Fatimids in their war against the Sunni powers, not only to get luxury products but also, and above all, to control the pilgrimage routes to Mecca. But it should be noted that this control is sometimes only linked to the occupation that takes place on the site, without any large-scale stakes, and the knowledge of local resources is then exploited by other networks, whose vocation is not to trade or circulate or control the space. This is the case of the islands of the Persian Gulf, which, from the 6th‑7th century onwards, host monasteries, whose members were attracted by the geographical isolation of the islands, of course, but also by the resources that they provide, which were known and exploited in other ways before, within the framework of other networks (Bonnéric).

9Ancient networks left imprints of their activity in space. These are obviously not as impressive as the traces made by modern connectivity (such as the Suez Canal). However, these traces are not negligible and their diversity is striking. The most visible ones are the structures built to allow the activity of networks, such as harbors and fortifications (Gelin et al.; Marion de Procé; Woźniak; Simmons), road stations (Crépy, Redon; Nehmé), watchtowers (Simmons), or warehouses (Gawlikowski). Such features characterise networks sustained by states. Let us mention for instance the Old Kingdom road system implemented by the pharaohs Snefru and Kheops in the Eastern Desert of Egypt (Somaglino, Tallet), or the facilities devised and set up by the Ptolemies and then the Roman power to secure water points and relays in the same region (Crépy, Redon), to safely connect the Nile Valley to the Red Sea ports (Woźniak; Abd el‑Raziq, Marchand, Somaglino) and to provide people with essential supplies (Gates‑Foster). Similar traces are evidenced in Farasan (Marion de Procé) or Failaka (Gelin et al.), two islands which hosted Roman and Seleucid military. L. Nehmé, on the basis of several surveys conducted recently, describes a mobility system, this one linking the two sides of the Red Sea. Ceramic material turned up in the Nabataean port of al‑Quṣayr, showings links with the opposite side of the Red Sea (Myos Hormos and Berenike). Her paper also discloses the complexity of the communication and exchange pattern in the Arabian desert, where routes connecting Hegra to the sea have been discovered. We could compare it with the network of roads established under the Ptolemaic and Roman powers in Egypt’s Eastern Desert. Besides this noticeable remains, faint traces are also found, such as graffiti and inscriptions left by travelers (Nehmé; Simmons). Most importantly, recent fieldwork conducted on the coast of Somaliland brought to light traces proving the passage of merchant ships between the 1st and 3rd centuries AD: among the items found in the graves of nomads buried at the site of Xiis (Heis), imports account for 99% of the material unearthed (Fernández Fernández, González‑Ruibal, de Torres). Finally, other networks may be termed elusive, in the sense that they have left almost no material trace, so that we are restricted to tracking them and even imagining their existence. This is the case of the network through which Islam spread in the Horn of Africa: its existence is revealed by the modest funerary stelae published by J. Loiseau et al. in eastern Tigray. In the same way, one can follow the expansion of Islam by tracking the spread of soft-stone craftsmanship on both sides of the Red Sea (Marchand) – manufactured soft-stone products were in great esteem in the Arab and Muslim world. In Failaka, the presence – among other things – of pig bones reveals that Greco-Macedonians settled on the island, showing the Seleucid interest in the Persian Gulf. In a nutshell, fieldwork and surveys disclose more and more branches of this network system. That said, some aspects of these networks remain more or less out of reach. For instance, J. Phillips reflects on the cultural consequences of foreign imports in the Aksumite kingdom, in an attempt to assess their impact in terms of appropriation, importation, copying, incorporation, interpretation. She comes to the conclusion that she can hardly grasp such cultural features.

10In this array of chapters with topics ranging from Pharaonic to modern times, great attention has been paid by authors to the shifts and changes that affected networks in the course of time. Indeed, contacts and exchanges sometimes grow and flourish, sometimes decline and shrink; in addition, any revival implies geographical or structural shifts. This topic is particularly evidenced by L. Sernicola and C. Zazzaro’s study. In their inquiry into the Mai Negus/Haselo valley over the long term, they examine how this place became part of local and regional networks whose cores were Aksum and Adulis, from the beginning of the Roman period to the end of the first millennium AD. However, this study also shows that the region participated in exchange networks before the peak of the Indo-Roman trade, and that a certain revival occurred in the 16th and 17th centuries. S. Taha and L. Smith et al. describe how Suakin’s destiny changed from the 13th to the 20th century, as a result of various factors, such as the connection with Beja networks, the organisation of the Mecca pilgrimage, the political and commercial rivalries between Ottomans, Portuguese and English powers, the vitality of trade with the Indian Ocean and China… Further north, similar phenomena took place. The third millennium BC was clearly a dynamic period in terms of connectivity, as demonstrated by the excavations at Ayn Sukhna and Wadi el‑Jarf. In the late 1st millenium BC, under the Ptolemaic rule, written and archaeological evidences cast light on a renewed network linking Egypt to Eastern Africa, the Arabian Peninsula and the Indian world, via the Eastern Desert. This process has been made clearer by the recent excavations at Berenike, where more and more Ptolemaic levels are being unearthed. At the bottom of the Red Sea, the vitality of the port of Clysma/Suez has been established thanks to the reassessment of the material from ancient excavations by M. Abd el‑Raziq, J. Marchand and C. Somaglino. On the other side of the Red Sea, the Nabatean networks were also subject to important changes in the course of time (Nehmé; Gawlikowski). Let us see the case of the Horn. The region of Wakarida (Ethiopia) was not part of the Red Sea networks during the first millennium BC, nor even at the beginning of the Aksumite period (late 2nd to early 4th centuries AD). It is not until the 4th century that this small agricultural territory opened up to trade with the kingdom of Aksum, probably following the founding of an urban center, before it declined at the end of the Aksumite period – the ties were probably not strong enough to survive this episode. Finally, the article by M. Cobb and T. Wilkinson on the famous Muziris Papyrus offers a different kind of analysis, as it rests on written evidence. The authors put into perspective the figures of Eritrean trade during the Roman period. They stress the fact that exchanges shifted over a long time but may also fluctuate in a short period of time.

11To conclude, many features of the networks that criss-crossed the Red Sea and the Indian Ocean have already been well described and analyzed, thanks to the previous Red Sea conferences. As for the 9th edition, it successfully pursued the goal devised by the projects founders and we are grateful to the participants for their remarkable contributions. Most importantly, however, the RS9 showed that a number of questions remain in the shadow and are still waiting for academic attention. In particular the Lyon conference shed light on “local peoples” and on the Horn of Africa – especially Somaliland. We are confident that the next meetings will continue along the same lines.

12Finally, we would like to deeply thank the following people: Cécile Clavé, Marcelle Alarcon, Ferouze Guitoun, Alicia Treppoz-Vielle who have been very dedicated in the organization of the RS9 conference; Caroline Develay and Thibaud Fournet who provided the visual documentation and created the posters and the program; Prof. Stéphane Gioanni, Director of the HiSoMA research center, and Prof. Yanick Ricard, Research Vice-President of the ENS‑Lyon, who have been committed to supporting the conference; Dionisius Agius for his wise and thoughtful advices; and of course the twelve members of the Scientific Committee, namely Dionisius Agius, Marie‑Françoise Boussac, John P. Cooper, Jean‑Charles Ducène, Jennifer Gates‑Foster, Elizabeth Lambourn, Laïla Nehmé, Jérémie Schiettecatte, Steven E. Sidebotham, Pierre Tallet, Chiara Zazzaro and Iwona Zych, who have played a critical role during the conference and in the publication of these proceedings.

  • 1 This project has received funding from the European Research Council (ERC) under the European Unio (...)

13We are also grateful to the institutions which have supported the conference: the ERC (European Research Council), which funds the Desert Networks project,1 the University of Artois, the HiSoMA research center, which is part of the “Maison de l’Orient et de la Méditerranée”, the “École normale supérieure” of Lyon, the IDEX of Lyon and the Confluences Museum. The Desert Networks project has also supported the publication of the proceedings that we are pleased to publish in the MOM Éditions series. The book has been prepared by us, the bibliography has been proofread by Bénédicte Khan, and the English language by Bryony White, whom we would like to thank. Finally, our most sincere thanks go to the authors who have participated to this publication and who made it possible to publish a rich, varied and, we hope, useful work.


Agius 2019: D.A. Agius, The life of the Red Sea dhow: a cultural history of Islamic seaborne exploration, London, I.B. Tauris, 2019.

Agius et al. 2012: D.A. Agius, A. Trakadas, C. Zazzaro, J. Cooper (ed.), Navigated spaces, connected places: proceedings of Red Sea Project V, held at the University of Exeter, 16-19 September 2010, BAR International Series 2346, Oxford, Archaeopress, 2012.

Ast, Rądkowska 2020: R. Ast, J. Rądkowska, “Dedication of the Blemmyan interpreter Mochosak on behalf of King Isemne”, ZPE 215, 2020, pp. 147‑158.

Blue et al. 2009: L. Blue, J. Cooper, R. Thomas, J. Whitewright (ed.), Connected hinterlands: proceedings of Red Sea Project IV held at the University of Southampton, September 2008, BAR International Series 2052, Oxford, Archaeopress, 2009.

Cobb 2018: M. Cobb, Rome and the Indian Ocean trade from Augustus to the early third century CE, Boston, Brill, 2018.

Cuvigny 2014: H. Cuvigny, “Papyrological evidence on ‘Barbarians’ in the Eastern Desert of Egypt (end 1st cent.-mid 3rd cent. CE)”, in J.H.F. Dijkstra, G. Fisher (ed.), Inside and out. Interactions between Rome and the peoples on the Arabian and Egyptian frontiers in Late Antiquity (200-800 CE), LAHR 8, Louvain, Peeters, 2014, pp. 165‑198.

De Romanis 2020: F. De Romanis, The Indo-Roman pepper trade and the Muziris Papyrus, Oxford, Oxford University Press, 2020.

Jaspert, Kolditz 2018: N. Jaspert, S. Kolditz, Entre mers – Outre-mer: spaces, modes and agents of Indo-Mediterranean connectivity, Heidelberg, Heidelberg University Publishing, 2018.

Power 2012: T. Power, The Red Sea from Byzantium to the Caliphate AD 500-1000, Cairo, The American University in Cairo Press, 2012.

Tomber 2008: R.S. Tomber, Indo-Roman trade: from pots to pepper. Debates in archaeology, London, Duckworth, 2008.


1 This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant Agreement no. 759078).


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search