Navigation – Plan du site

AccueilNuméros16Dossier: Nommer les dieux en Syri...Roman Soldiers in the Religious, ...

Dossier: Nommer les dieux en Syrie gréco-romaine. Paysages onomastiques et dynamiques religieuses

Roman Soldiers in the Religious, Social, and Spatial Network of the Hauran

Les soldats romains dans les réseaux religieux, sociaux et spatiaux du Hauran
Francesca Mazzilli

Résumés

L’article explore le profil des soldats romains dans les sites de culte de l’Hauran par le biais de network and spatial analysis. Le but sera atteint par l’étude des interactions entre les agents sociaux des temples et les divinités, de la même manière que la proximité des sites cultuels et entre les sites et les routes. Malgré la grande quantité de dédicaces, surtout dans les sites de culte le long des routes romaines, l’échelle du principal patronage (agentivité ? exploitation des sites de culte ?) des soldats ne différait pas de celle des autres membres de la communauté. Les soldats appuyaient les programmes de construction religieuse déjà existants dans les villages, mis en place par les communautés locales, et faisaient des dédicaces dans des sites de culte où individus, bouleutai et anciens combattants souvent contribuaient, mais il n’y a pratiquement aucune trace de collaboration directe entre ces groupes et les soldats. Même si les soldats faisaient des dédicaces aux divinités qui n’étaient pas souvent adorées par les bienfaiteurs de l’administration locale, néanmoins ils adoraient la divinité précédemment vénérée, Zeus Kyrios.

Haut de page

Texte intégral

Acknowledgements

I would like to thank Professor Corinne Bonnet and Giuseppina Marano for their support for this publication. I also would like to thank them and Dr Aleksandra Kubiak-Schneider for inviting me to the workshop «Naming the Gods in Graeco-Roman Syria» at the University of Toulouse. I am also grateful to the anonymous peer-reviewers, Eivind Seland, and Tomáš Glomb, for their precious comments and improvements to the text. Thanks for Gwendolyne Knight for proofreading the article. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No. 892166.

Introduction

  • 1 Pollard 2000, 1.
  • 2 Ibid.
  • 3 These figures are the result of the work published by Sartre-Fauriat (2005, 118-119).
  • 4 Sartre-Fauriat 2005, 130-131.

1Soldiers contributed to urbanisation and institutional identity across the Roman Empire, including in the Near East.1 The role of soldiers in Syria and its vicinity (Mesopotamia and Osrhoene) is an optimal subject of investigation because its «evidence is unique in its quantity and nature.»2 This is especially the case in the Hauran (southern Syria), where a higher percentage of inscriptions commissioned by soldiers dating from the second to the fifth century AD have been recorded than in other parts of Syria. While 18.30% of the inscriptions in the Hauran were dedicated by civilians (585 dedicatory inscriptions out of 3195, the total number of all inscriptions recovered in the region), soldiers commissioned 8.2% of the inscriptions recovered in the region (128 dedications, 130 funerary inscriptions, 4 honorific inscriptions), versus 5.1% of the total epigraphic texts found in the Roman province of Arabia and 3.3% in the province of Syria. This figure indicates that soldiers were more prominent members of the financial elite in the Hauran than in Arabia and Syria.3 According to onomastic evidence, soldiers who commissioned inscriptions in the Hauran were locally recruited from within the region, and they invested in their community both during their service and once they retired.4

  • 5 Sartre-Fauriat 2005, 125-127; 2015, Mazzilli 2018, 88-94.
  • 6 In these instances, inscriptions were in Latin, and it has been already discussed that this languag (...)

2Soldiers’ impact in the region also extended into the religious sphere.5 Soldiers often dedicated inscriptions to specific gods associated with the army, such as Mithras and Jupiter Ammon, and to what Sartre-Fauriat has called imported deities, and Mazzilli calls deities with Graeco-Roman names, like Apollo.6

3Although the impact of soldiers on religious life in the Hauran has been widely discussed, this paper aims to investigate it more thoroughly through a multi-levelled quantitative analysis of inscriptions. This encompasses gradually integrating the following additional contextual elements: temple benefactors (individuals, priests, and especially those who presented themselves having a specific role in the community, such as veterans, bouleutai, village magistrates, and village communities acting collectively), the deities from cult sites, and the location of cult sites.

4In this paper, soldiers’ patronage of cult sites is weighted against that of individuals, priests, veterans, bouleutai, village magistrates, and village communities as a whole. Then, deities worshipped by soldiers are compared with those venerated by other benefactors, and with gods worshipped at cult sites more generally. The distribution of cult sites, the dedications commissioned by soldiers and other benefactors, as well as the spread of the main deities throughout the region are analysed by examining the proximity between cult sites, and between these sites and the road network. This spatial analysis assesses whether the proximity between cult sites, and between cult sites and road networks, could have facilitated the spread of central deities or the circulation of benefactors.

  • 7 For example, Glomb et al. 2018; Glomb et al. 2020; Glomb 2021.
  • 8 Mazzilli 2023. For example, see ERC Mapping Ancient Polytheisms Project (Bonnet 2022a; 2022b).

5This multi-layered analysis is achieved through a twofold network and spatial analysis. Network and spatial analyses are valuable techniques for systematic analysis and provide a clear visualisation of relationships between soldiers and benefactors, benefactors and gods, and benefactors or deities and the landscape.7 This twofold methodology facilitates an explication of the interplay between religion (here represented by the names of deities), social agents (benefactors), and landscape (proximity between cult sites, and between cult sites and road networks); this interplay has been growing in popularity as the methodology of investigations in the study of religion in the pre-Roman and Roman eras.8

The data

  • 9 Mazzilli 2018.
  • 10 IGLS XIII.1-2; XIV.1-2, XV.1-2, XVI.1-2, XVI.forth.
  • 11 Hauran VI.

6The data in this study consists of 196 inscriptions that come from or are associated with six urban cult sites and 70 rural cult sites from the pre-provincial to the provincial period (roughly the second century BC to the third century AD) (Appendix, Table 1, Fig. 1). The compiled data were collected from the gazetteer of rural cult sites in the Hauran,9 as well as more recent publications on epigraphy, such as IGLS,10 and on pre-provincial sites from the region.11

Fig. 1: Map of cult sites from the pre-provincial and provincial period within the road network. Data from Appendix, Tables 1, 2, 3

Fig. 1: Map of cult sites from the pre-provincial and provincial period within the road network. Data from Appendix, Tables 1, 2, 3

Bing Virtual Earth Basemap adapted by Author 2022

  • 12 The preponderance of rural cult sites is due to the fact that the region seems to be based on a vil (...)
  • 13 For further information about the site typology see Methodology Section.

7Urban cult sites refers to cult sites in cities and rural cult sites to those situated in a non-urban context, often in villages.12 A cult site can include more than one sanctuary, and can be associated with the same ancient settlement, or a modern one.13 As not all of the cult sites have an ancient name that has survived to the present day, their modern names, using the spelling from the most recent academic publications, have been used instead.

  • 14 There are a couple of examples of inscriptions by soldiers at cult sites from the pre-provincial pe (...)
  • 15 Josephus, AJ 17.28.
  • 16 Millar 1993, 95-96.

8Inscriptions from the provincial period are the main source for this study, as its subject is Roman soldiers, whose presence in the Hauran is mainly attested from this period.14 Provincial period refers to the time when the territory was part of the Roman Empire. The north (lava terrain of modern-day Leja, ancient Trachon) and centre (north of Djebel al-Arab, north of ancient Auranitis) of the Hauran were permanently annexed to the Roman province of Syria in AD 93/94,15 whereas the southern part of the region (south of Djebel al-Arab and Nuqrah, south of ancient Auranitis) became part of the Roman province of Arabia a few years later, in AD 106.16 Therefore, the analysis in this study is focused on 69 cult sites that were dated to, or still in use in, the provincial period.

  • 17 The dating probability of the dataset is not included in the analysis but may be found in the Appen (...)

9Whereas the date of 51 inscriptions is explicitly stated in the text, for the rest of inscriptions (a total of 145) one must often estimate whether they ought to be dated to the pre-provincial or provincial period. In this paper, the level of confidence for dating inscriptions rates from certain (Level 1) to probable (Level 2), possible (Level 3) and unknown (Level 4).17 Level 1 refers to when the date is known based on the epigraphic text. Level 2 refers to when the exact date of the dedication is not known, but it can be estimated with a certain level of confidence as it is found in a non-multi-period temple, or the god mentioned in the inscription appears elsewhere only in the provincial period. Level 3 refers to when the cult site is multi-phased, or the date of the cult site is unknown, but the date of the dedication can be estimated to be one of the two phases. Level 4 is when the date of the inscription is mere speculation, there is no other clear dating evidence, and the god mentioned in the inscription was worshipped in the region in both the pre-provincial and provincial periods.

  • 18 AAES II, 405; IGLS XVI.1, 191.
  • 19 IGLS XVI.forth, 751.

10As this study focuses on epigraphic records from cult sites, inscriptions that seem to be funerary or honorific inscriptions, as well as those that commemorate civilian buildings have not been included, although a reference to a god was also often included in the funerary and honorific types of epigraphy. There are two exceptions to this exclusion. One is an inscription within the group of honorific texts from the city Canatha (present-day Qanawāt), as it refers to a sacred avenue or promenade involving the presence of Seenoi (inhabitants of Sī‘) and the centurion of a legion.18 It can be linked to the cult centre at Sī‘, to a religious complex at Qanawāt, or both. The second is an epitaph from Nimreh because it informs us about the construction of a Tychaion, temple dedicated to Tyche, in the village.19

  • 20 Sartre-Fauriat 2005, 130-131.

11The reason for looking at inscriptions from cult sites is that it facilitates the examination of soldiers’ participation in public religious places that would be visible in a main place of gathering. Even though soldiers in the Hauran were often individuals of local origin,20 their deliberate choice to identify themselves as soldiers in this public context can be considered evidence of displaying their social advancement, recognised by Rome, and the wealth associated with their position. It would also have been a way to show their pride in belonging to a specific group, the legion, which was under the service of Rome, as well as a sense of honour in their service to the community and to the Empire more broadly. Because of their public service to Rome and their role in the community, they had the financial means to commission dedications in public religious buildings. Similar ideas of pride in their service to the community, and evidence of wealth due to their social role, can be said for individuals or groups who presented themselves as veterans, bouleutai, village magistrates and village communities as a whole.

  • 21 Sartre 1991, 328, 252.
  • 22 Sartre 1987, 244.
  • 23 Sartre 1993, 113-135.

12In this context, although individuals have been included in the discussion, veterans, bouleutai, village magistrates, and village communities acting as single units are the most appropriate points of comparison. An analysis of their contributions to cult sites can offer a better assessment of the position of soldiers in the community, as a cult site was a main public gathering place in the local community. Veterans (οὐετρανοί) were usually soldiers who had retired after their service in the Roman legions; they were often granted land because of their service to Rome.21 Α bouleutes (βουλευτής) was a civic councillor, often identified as a member of the municipal council of the city.22 In scholarly discourse, the term has often only been transliterated from ancient Greek, perhaps because of complications in its translation ; it is not impossible that bouleutes can also refer to the councillor of a village.23 There is only one case in the Hauran where the city itself commissioned an altar at Bosra, and it has been differentiated from dedications by bouleutai.

  • 24 I have elected to use modern names, as we do not have names of ancient villages and sites for all t (...)
  • 25 IGLS XV.1, 90 for Zebīreh and IGLS XV.1, 172 for Ezra‘.
  • 26 MacLean Harper 1928, 105-116; Sartre 1993, 120-121.
  • 27 Sartre-Fauriat 1999, 223-240.

13Village communities were groups of individuals that collectively called themselves koinon (τὸ κοινόν), meaning community either generically, in the sense of a village, or specifically, with reference to the name of a village (e.g. the community of Agraina, modern-day Lubbein ; Damatha, modern-day Dāmā-Dāmit al-‘Aliyyah ; or Saura, modern-day Ṣūr al-Lejā’), or to inhabitants of a village (e.g. Seenoi, from Sī‘; Phainesiens, from the Roman village Phaina, modern-day Mismiyyeh; or Airesioi from Aere, modern-day Sanamein).24 The Iachphirenoi under the domain of Habibenoi and Bassos at Zebīreh and the Saameniens of Zorava at Ezra‘ are not included in the village community category because it is unclear if they refer to a group of people belonging to a village or a tribe and they do not appear to belong to the village community as a single unit.25 Koinon is not very different from the term kome (κώμη), meaning village, but the former stresses the organic aspect of a village’s community,26 and the autonomy involved in building amenities or prestige buildings thanks to public funds.27

  • 28 McLean Harper 1928,123-127.

14Village magistrates were individuals belonging to the village administration; they included pistoi (πιστοί, literarily meaning reliable persons) episkopoi (ἐπίσκοποι, literarily meaning supervisors/inspectors), strategoi (στρατηγοί) and pronoetai (προνοηταί, both mean officials and supervisors). They were all involved in public works or buildings in southern Syria.28

15Included in the group individuals are single or multiple benefactors who presented themselves in inscriptions using their personal names. Those who called themselves priests (ἱρείς) have been considered separately from this group, as they have a specific religious role in the community.

  • 29 Bonnet et al. 2018; Belayche 2017.
  • 30 Network analysis with each part of the divine formula (e.g., theonym, epithets) seems a valuable me (...)
  • 31 Kubiak-Schneider 2021.
  • 32 Sartre-Fauriat 2007, 18
  • 33 IGLS XIII.2, 9571.
  • 34 Sourdel 1952, 69-74.
  • 35 IGLS XVI.2, 304, 307a.
  • 36 Will 1985, 105-125; IGLS XVI.2, 304.
  • 37 See further discussion on the meaning Mazzilli 2018, 66-67; Mazzilli 2021.
  • 38 IGLS XVI.2, 307a.

16When considering the names of deities in this analysis, the complete divine formula has been regarded as a single entity whenever possible, such as «Helios, the god Megistos». Using the complete divine formula can help to keep what constituted the theological identity of the gods intact; that is, not only the name of a god, also known as a theonym, but also the attributes and epithets associated with them, as each component is equally essential.29 Therefore, grouping deities into one category based on their shared theonym can be misleading. For instance, in the case of Zeus and his epithets, such as Keuranos and Kyrios,30 each has been differentiated in this analysis because, although they are all under the same umbrella as leading celestial gods, they had different meanings.31 This can also be said for other deities, such as Tyche and Athena. Tyche is a common deity with a Greek name in the Near East, including in the Hauran, as she is the protector of the village communities and local individuals.32 However, by adding the epithet Samethos to Tyche, as in the case of the cult site at Deir (South), Tyche is specifically referred to a tribe or place called Samethos.33 In the Hauran, the Semitic warrior deity Allat was called Athena in Greek inscriptions.34 However, this is not always the case. Athena can, for example, be considered a Greek deity in the list of deities depicted in the relief at Suwaidā’,35 with a Greek theme.36 Athena is, after all, still a deity with a Greek name.37 In six out of 14 inscriptions dedicated to Athena from this study’s corpus, the deity has the additional epithet Kyria (the lady), which encourages the identification of Athena as a local deity.38 However, this epithet is not exclusively associated with Athena in inscriptions from the Hauran. From the corpus of this study, Kyria is the epithet of the deity Atargatis in one instance, but also of Tyche Patria in three inscriptions from Bosra (Appendix, Table 2).

  • 39 This has been extensively discussed elsewhere; see, for instance, Sartre-Fauriat 2015; Mazzilli 201 (...)

17The aim of the paper is not to discuss where these deities came from and which cultural identities they could embody,39 but to identify relationships between who worshipped these deities from the administrative community, and to draw some conclusions about the role of soldiers in this interlinked religious and social network.

  • 40 For the use of ‘gods of individuals’ who have alternatively also been called ‘personal gods’, see P (...)

18Because of their common local ties, gods bearing the name of an individual, a tribe, or a toponym of the region have been sorted into the following three groups: gods of individuals,40 toponymic deities, and deities followed by the name of a tribe.

  • 41 IGLS XIV.1, 349; IGLS XVI.2, 382d.

19The first group comprises deities where the generic name of a god is usually followed by the genitive case of the name of an individual. We cannot be certain that the saint-god Askaras from Tall Aḥmar and the ancestral god Akeiras from Dā’il can be under this group as Askaras and Akeiras could be the names of a dedicator or of a god, but there is no parallel.41

  • 42 Kaizer 2007, 449.

20The second group comprises deities with a Greek theonym followed by a regional toponym, usually a place or a village (e.g., Athena of Arra, Athena Kyria Gozmaie, Tyche of Sakkaia, Zeus of Aire, and Zeus of Phaina). Toponymic deities appear in Roman Syria.42 Examples include Zeus Damaskenos (Zeus of Damascus), Jupiter Optimus Maximus Heliopolitanus, from Heliopolis. Both deities appear in the Hauran. As they are deities with a toponym outside the Hauran, they have not been considered in the group of toponymic deity.

21Deities followed by the name of a tribe consist of a Greek theonym followed by the name of a tribe (e.g., Tyche of Samethos and Kronos of Sokarathoi), but they are difficult to identify. Kronos of Azobos and Kronos of Sokarathoi, for example, can represent either a toponymic deity or a Greek theonym associated with the name of an individual or a tribe. In cases where their association with a tribe or a name is not clear, both are outlined in the analysis (Appendix, Table 2, Fig. 6).

Methodology

22This article undertakes a two-step analysis that examines the role of soldiers at cult sites in relation, firstly, to other benefactors, and, secondly, to the gods worshipped by the benefactors and the gods worshipped at cult sites.

  • 43 Brughman, Peeples, 2017, 2; Knappett 2011, 9; Fulminante 2014; Knox et al. 2006.
  • 44 Da Vela 2015, 229, for examples of two-mode networks and their meaning.

23The investigation of the role of benefactors is undertaken through network and spatial analyses. Network analysis is an analytical application used to understand how a set of links work.43 It is a valuable tool to deal with relational phenomena, whereby cult sites (nodes) and the links (ties or edges based on having the same type of attributes) are the constitutive elements of networks. Visone 2.18 software is used to create a two-mode network.44 It visualises the relationships between different kinds of benefactors (attributes), and between benefactors (visualised as squares in the graphs) and cult sites (circles in the graphs) where both are nodes of the network, creating clusters of sites associated with one or more attributes (Fig. 2, 6, 7, 8).

24To understand the role of benefactors at cult sites more thoroughly, three other elements have been integrated into the analysis: the frequency of the presence of each type of benefactor, the type of patronage of cult sites by benefactors and the cult site typology. The first element represents the number of inscriptions by weighting the edge (visible through the thickness of the line). The frequency of a temple’s patronage by benefactors is evaluated by differentiating between when the benefactor made a major contribution to the cult site (e.g., a temple or its door) and when he only commissioned a dedication or an altar (Fig. 2). This differentiation shows the level of contribution to religious and public life by members of the community, and also acts as an indication of their wealth and influence in society. The level of benefactors’ participation in public life also depends on the size of the cult site that they supported financially. The variation in size between cult sites is included in the analysis of benefactors, and is divided into six broad categories based on archaeological and epigraphic evidence. Type 1 comprises complex religious centres with more than two temples within the same sacred enclosure. It is followed by Type 2, which consists of substantial religious centres that have a wide circumscribed sacred area (based on remains of a sacred precinct wall or the mention of a temenos in an inscription), or two temples (according to archaeological or epigraphic evidence). Type 3 includes sanctuaries that are significant because of the variety of deities and the large quantity of inscriptions (between eleven and twelve) recovered from the pre-provincial and provincial periods, but with no current standing remains nor written evidence to inform us of the presence of a temenos or the number of temples. Type 4 comprises sanctuaries without a temenos where their layouts can be identified, and which also have a smaller quantity of inscriptions than in Type 3. In Type 5 are sanctuaries whose layout cannot be identified, but scattered remains of statues, decorative motifs, or inscriptions can still offer some interesting insight about the cult site. Type 6 includes sanctuaries that do not offer any information about the actual cult site apart from their presence or possible existence as suggested by scholars. It is important to point out that this typology is biased, based on the preservation of cult sites and their inscriptions; however, this is the case of any source of investigation into the ancient past. Therefore, it should not be considered absolute, but still a valuable component in this analysis in order to have a more comprehensive understanding of the role of soldiers at cult sites.

  • 45 Rivers 2016, 133.
  • 46 Broodbank 1993; Davis 1982; Hunt 1988; Terrell 1977; Collar 2013.

25Next, a spatial analysis of benefactors has been undertaken in order to understand the impact of the Roman army on the territory through the proximity of cult sites, and the proximity between cult sites and the road network. The proximity of sites is assessed by proximal point analysis (PPA). This analysis uses geographical space to identify the nearest neighbours or sites within a given dataset.45 Studies of the ancient Mediterranean that use this application often consider the proximity of the closest three sites,46 as does this analysis. PPA is a valuable tool that permits the identification of clusters according to the proximity between sites, and which might not be apparent by visual inspection. PPA does not take in consideration the type of the terrain or routes. It cannot be considered a controversial methodology in the Hauran, as sites are easily accessible: the terrain is mainly between 450 and 600 km away from the mountainous area of modern-day Djebel al-Arab, in the eastern part of the region, which has a gentle slope with a maximum height of 1700 m. Only a handful of cult sites reached 1250-1400 m. Furthermore, the Hauran is a small area, roughly 100 km by 100 km ; thus, it would be easy to cross it, and there are no huge distances between cult sites (Fig. 1, 3). PPA is a valuable tool, as proximity between sites would have facilitated not only the movement of religious traditions, but also the movement of people. Furthermore, the proximity between sites is also assessed by identifying the closest three sites to another one using roads, when relevant. The calculation of PPA is provided by plugins from QGIS software. Distance Matrix is used to identify the three closest cult sites to one another, whereas QNet ODMatrix is the plugin used to identify the three closest cult sites to one another using roads. Their results are visualised through Point Connector. The distribution of benefactors of cult sites in relation to the Roman road network is examined by analysing the cult sites within 5 km of the roads using Visualist.

  • 47 Bauzou 1985; 2003.
  • 48 Witcher 1997; Kolb 2015, 649. For display of power over landscape see Purcell 1990, 15, 23.

26Roman roads were built in the provincial period, corresponding with the time when evidence for the presence of Roman soldiers was greatest (Fig. 1). In this analysis, ‘Roman roads’ indicates roads built in the second-third century, according to milestones and archaeological evidence; however, this does not mean that these roads were not used earlier on. Secondary roads were subordinate, most likely pre-existing, roads known to be used in the Roman period. Hypothetical roads do not have Roman milestones, and their construction and use in the Roman period is not certain (Fig. 1).47The absence of known roads does not imply there were no routes between particular sites, but it implies that these routes did not belong to the infrastructure system built when the region was under Roman dominion. The road network can be seen as symbol of Roman power over the landscape.48 For this reason, it is quite important to discuss the relationship between cult sites and roads, especially considering that a total of 45 out of 69 provincial cult sites were within a 5 km radius of a Roman road (Fig. 1).

27The second step of the analysis focuses on the gods worshipped by benefactors at the cult sites and deities from cult sites through a two-mode network. The edge stands for the action of a benefactor making a dedication to a deity at a cult site; it varies depending on who the benefactor was (identified by the colour of the edge, Fig. 6). It aims to create a link between benefactors and the gods they worshipped. A spatial analysis of the cult sites where the main deities worshipped at those sites, and the deities worshipped by soldiers, reveals the deities’ distribution in relation to the road network and the proximity of cult sites using PPA.

Benefactors: a network analysis

28The two-mode network of benefactors can help to deconstruct the role of soldiers in more detail, rather than merely indicating their major impact on the public religious life of the region based on the large quantity of their written dedications (37 inscriptions), exceeded only by those commissioned by individuals (with a total of 56 inscriptions from the provincial period).

  • 49 Speidel 1984, 691-692.

29The weight of the edge helps us to visualize the observation that individuals, followed by soldiers, were the group of benefactors who made more multiple dedications than other benefactors at the same cult sites. However, while individuals made multiple dedications in cities (Bosra, Dionysias, which is present-day Suwaidā’, and Canatha, which is present-day Qanawāt), soldiers and individuals made an equal number of dedications only at Bosra. Individuals outnumbered soldiers at Canatha and predominated at Dionysias. This discrepancy between cities exists because Bosra was the only city where the Legio III Cyrenaica was stationed.49

  • 50 IGLS XV.1, 17, 18, 19; Sartre-Fauriat, Sartre 2014, 36.

30The largest quantity of dedications by soldiers was in the village Mismiyyeh, which had a smaller number of dedications by individuals (4 versus 2). This is not surprising considering that a garrison of the detachment of the Legio III Gallica and Legio XVI Flavia Firma was situated in Mismiyyeh.50

31When looking at the site typology, we may notice that soldiers often made dedications at complex religious centres. At four cult sites within this group, soldiers funded the main building or a major structural component: a centurion completed the sacred enclosure and consecrated the Tychaion at Ṣanamein in AD 191; centurions commissioned sanctuaries and statues at Mismiyyeh (the ancient village of Phaina); a Roman soldier rebuilt a building, possibly a temple, at Mseikeh; and another soldier commissioned a door of the North Temple at ‘Atīl.

32By including the type of contribution that benefactors made, it is possible to suggest that soldiers might not have played an extensively predominant role at cult sites across the Hauran, since village magistrates, village communities, individuals, and veterans were patrons of a larger number of cult sites, although these sites were often of a smaller size than those where soldiers were main benefactors (Fig.2).

Fig. 2: A weighted graph showing relationships between benefactors (squares) and cult sites (circles) according to inscriptions from the provincial period. The colour reflects the benefactor category, and when the circle, standing for the cult site, is wholly coloured according to one of the benefactor categories, it means that the benefactor commissioned the temple or a main component of the building. Data from Appendix, Table 2 (Author 2022)

Fig. 2: A weighted graph showing relationships between benefactors (squares) and cult sites (circles) according to inscriptions from the provincial period. The colour reflects the benefactor category, and when the circle, standing for the cult site, is wholly coloured according to one of the benefactor categories, it means that the benefactor commissioned the temple or a main component of the building. Data from Appendix, Table 2 (Author 2022)

33While it is common to find the copresence of different categories of benefactors at the same cult site (31.2%, 24 out of 69 cult sites), soldiers appear to be the only main patrons at ‘Atīl and Ṣanamein, whereas bouleutes, individuals, or veterans were the main patrons at cult sites in Qanawāt, Ṣūr al-Lejā’ and Ḥibrān, implying a secondary role for soldiers at these sites (Fig. 2).

  • 51 Sartre 2017, 175.
  • 52 For Canatha see Sartre 1981; Denzter 1986, 396-67; for Bosra see Bowersock 1983, 73 ff.; Sartre 198 (...)
  • 53 Speidel 1984, 691-692.

34A bouleutes with his wife and son made a pronaion, and another bouleutes commissioned a base or a part of the sanctuary, according to an inscribed column base at Qanawāt. A similar type of dedication was also made by an individual in the sanctuary of Zeus Megistos. In addition, it has been suggested that a civic sanctuary was built at Bosra because an altar was commissioned by the city itself,51 but no explicit evidence has indicated that the city was the patron of a temple. The financial contribution by bouleutes or the city to religious public buildings at Bosra and Canatha (Qanawāt) can be explained by the fact that the sites were the two main long-term urban centres of the region.52 However, bouleutai made more than one major contribution to the sanctuary at Qanawāt, whereas soldiers commissioned multiple dedications at Bosra. This distinction might have existed because a Roman legion was stationed at Bosra.53

  • 54 For the Herodian garrison at Ṣūr al-Lejā’in the pre-provincial period see IGR III, 1144; PAES III, (...)

35Veterans seemed to have made primary contributions to relatively substantial cult sites at Ṣūr al-Lejā’ and Ḥibrān by commissioning a doorway, although the presence of soldiers, including dedications in sanctuaries, is also attested at both sites since the pre-provincial period.54 Both locations are long-term significant cult sites, based on the variety of deities, the archaeological remains and the number of inscriptions (site typology, Appendix, Tables 1, 2 and 3).

  • 55 Segal 2008, 110; Sartre-Fauriat, Sartre 2014, 37.
  • 56 IGLS XV.1, 5.
  • 57 Sartre-Fauriat, Sartre 2014, 37.
  • 58 IGLS XV.1, 11.

36Soldiers rarely acted as the main patrons at the same cult site where other benefactors made a major contribution, except for Mismiyyeh and Mseikeh (Fig. 2). In the first case, village inhabitants first consecrated a temple or part of it, potentially the prostyle temple identified in the 19th century but dismantled to build the Ottoman barracks in northern edge of the village;55 here, statues of Isis and Pax were commissioned by soldiers. A few years later, an architrave of a temple was built under the authority of a centurion, and another centurion consecrated a temple and a statue. Additionally, a Latin inscription mentions a templum and a statue at the time of the emperor Commodus.56 It is unclear which sanctuary, or indeed how many sanctuaries, these inscriptions refer to. They may refer to the remains of a second temple identified in the west of the village by the architect Kalos.57 Alternatively, the Latin inscription mentioning a templum and a statue may refer to the temple with other imported deities like Isis and Pax.58

  • 59 IGLS XV.1, 17, 18, 19.
  • 60 Sartre-Fauriat, Sartre 2014, 36.
  • 61 IGLS XV.1, 13.
  • 62 Sartre-Fauriat, Sartre 2014, 34.

37Mismiyyeh (ancient Phaina) can be considered a major religious centre for soldiers, as it has the highest quantity of dedications by soldiers in cult sites in the Hauran (Appendix, Table 2, and Fig. 2). This can be explained by the presence of a garrison of the detachment of two legions in the village.59 Its religious importance in the territory is indicated by the presence of a god named after the village (Zeus of Phaina) at a cult site at Dāmā-Dāmit al-‘Aliyyah in the southern Leja. It can be presumed that the temple commissioned by village inhabitants to Zeus Megistos was implicitly dedicated to Zeus of Phaina.60 Mismiyyeh was the Roman village Phaina, which became a metrokomia, meaning «mother» village, during the reign of emperor Commodus.61 This new title implied that it was a key settlement within the administrative structure of the region, and had a more important role and more responsibilities than a village. The importance of this village in broader terms in the provincial period is suggested by the fact that the village community of Phaina was attested in Rome.62

  • 63 IGLS XV.1, 140a.

38Information about the cult site and the settlement of Mseikeh is scantier than Mismiyyeh. However, according to the epigraphic record it is possible to say that a veteran built a temple, possibly with an enclosed courtyard,63 and a soldier rebuilt a possibly sacred building, although there is more than one inscription dedicated by a soldier. We can assume that Mseikeh was not just a small cult site, due to the copresence of two different types of dedicators and the worship of more than one deity in the same village. The deities include the god of Loaithemos (i.e., part of the group gods of individuals), as well as Zeus, an ancestral god. An additional temple or a group of statues was dedicated to the imperial cult (Appendix, Table 2, and Fig. 2, 6).

39It is interesting to note that in both Mismiyyeh and Mseikeh, soldiers were not the initial benefactors: they either carried on the religious building programme in the village started by the inhabitants, in Mismiyyeh, or they rebuilt a building, most likely a religious structure, in Mseikeh. This suggests a certain sense of continuity in maintaining and supporting the religious building programme of the local community on the part of the soldiers. It is the same for the Tychaion at Ṣanamein that was completed by a centurion. Nevertheless, these examples do not show a joint collaboration between soldiers and other benefactors, witnessed in the same epigraphic record, although they might have made a dedication or contributed to a building at the same cult site.

  • 64 Wadd., 2367; AAES III, 428b; IGLS XVI.1, 263.

40An exception is an inscription found in Qanawāt where Seenoi (inhabitants of Sī‘) were might also have contributed to the inscription written in honour of the sacred avenue/promenade with a centurion of the legion. It is tempting to suggest that this inscription might refer to the sacred way attested at the major cult centre at Sī‘, as the inscription was commissioned by inhabitants of the settlement roughly 2 km from the city Canatha (Qanawāt), and their name appears in another inscription from Sī‘.64

41Civilians appeared to have made joint inscriptions with other benefactors in a couple more instances than did soldiers. Bouleutai made two dedications with their family members at Qanawāt and Mušannaf, implying that their dedications were more likely a private commission than an official one. Temple treasurers, consisting of a group of individuals, including a veteran, commissioned a dedication or a sanctuary at Sahwet al-Khoḍr. In addition, a soldier who made a dedication at Sulaym was also a priest. These two inscriptions show that veterans and soldiers were integrated in the life of the sanctuaries and were in charge of their business and religious matters.

42The large quantity of dedications by soldiers in cult sites is undeniable, especially at substantial cult sites. However, their role as temples’ main patrons did not outnumber other main benefactors in the region. Soldiers also contributed to the religious building programmes of villages started by other members of the community, without directly collaborating with other benefactors.

Benefactors: a spatial analysis

43Regarding the spatial analysis of benefactors, it is apparent that soldiers were mainly benefactors of cult sites situated on Roman roads or in their immediate proximity (within 5 km) (Fig. 3). This outcome is even more striking when contrasted with cult sites where other benefactors made financial contributions, which were often not located on the road (Fig. 3). Cult sites with soldiers as benefactors covered the main road crossing Leja, and the road bordering the eastern edge of Leja, all the way to Bosra. In particular, soldiers also made multiple dedications at Dionysias (present-day Suwaidā’) and Bosra, cities on the main road; however, their largest quantity of dedications was in the village of Mismiyyeh, on the road crossing Leja and leading first to Suwaidā’, and then Bosra.

Fig. 3: Map showing the cult sites with different benefactor categories in the road network. Data from Appendix, Table 2

Fig. 3: Map showing the cult sites with different benefactor categories in the road network. Data from Appendix, Table 2

Bing Virtual Earth Basemap, adapted by Author 2022

  • 65 IGR III, 1144; PAES III, 797/1; Brünnow, von Domaszewski 1909, 309; IGLS XV.1, 103.
  • 66 Hauran VI, 508-509.

44The sanctuaries where Roman soldiers commissioned a building or a major structural component were placed on a Roman road (Mismiyyeh, Ṣanamein, ‘Atīl), apart from the temple at Mseikeh, which was not situated on the Roman road. Mseikeh, together with Ṣūr al-Lejā’, is one of the main cult sites with soldiers as benefactors that were not placed near the road network (Fig. 2-3). However, in the village at Ṣūr al-Lejā’, roughly 6 km from Mseikeh, soldiers were attested in the pre-provincial period before the construction of Roman roads, based on a big lintel commissioned by soldiers from Agrippa’s army.65 It has been suggested that this settlement would have been part of the route in the southern Leja, based on the occupation of Ṣūr al-Lejā’, Jaddal, Dāmā-Dāmit al-‘Aliyyah, and ‘Arīqah during the pre-provincial period. There is, however, no archaeological evidence of tracks alongside these sites.66

45In the cult sites at Ṣūr al-Lejā’ and Mseikeh, veterans seem to be the main benefactors. By applying PPA without roads, we are able to see that these two sites were part of a small cluster of cult sites where veterans were dedicators; namely, the sites of Muṭa‘iyyeh, Shaqrā, Mseikeh, and Ṣūr al-Lejā’ in Leja (Fig. 4). The cult site of ‘Arīqah, where a veteran was a main temple benefactor, did not belong to any clusters, but it was still within a range of less than 20 km of this small cluster. In Djebel al-Arab, veterans appeared to be the main benefactors at cult sites in Sahwet al-Khoḍr and Ḥibrān, which were less than 7 km apart. In the first case, the veteran was a temple’s treasurer, and possibly contributed to the construction of a sanctuary. Veterans also commissioned a temple door at the second site, where there were another two dedications by a veteran. The proximity between the two cult sites could have influenced the significant role of veterans at these two temples, which can also be suggested for the other cluster in southern Leja, especially at Mseikeh and Ṣūr al-Lejā’, which were next to each other and where veterans were the main patrons of temples.

Fig. 4: Map showing PPA of cult sites where benefactors commissioned dedications. Data from Appendix, Table 2

Fig. 4: Map showing PPA of cult sites where benefactors commissioned dedications. Data from Appendix, Table 2

Bing Virtual Earth Basemap adapted by Author 2022

46Village communities were patrons of three temples in a cluster of sites close to each other in southern Leja, but one of the three examples was not directly connected with the other two (Fig. 4). Inhabitants of the village also contributed to the erection of the temple at Mismiyyeh, situated on the road, although soldiers were the predominant social agents in the religious life of the village (Fig. 2-3). Village magistrates were often benefactors of cult sites not situated on a Roman road, especially the north-south road crossing the Hauran. At the same time village magistrates did not seem to form clusters based on proximity between sites. Cult sites with soldiers as benefactors situated in the Djebel al-Arab, the eastern edge of Leja, and in the middle of Leja seemed to form small clusters of sites, according to PPA using roads and not simply PPA (Fig. 4-5). However, they seemed to form clusters of few sites, and this figure is not as striking as the proximity to roads of cult sites that had soldiers.

Fig. 5: Map showing PPA of cult sites, using roads, where benefactors commissioned dedications. Data from Appendix, Table 2

Fig. 5: Map showing PPA of cult sites, using roads, where benefactors commissioned dedications. Data from Appendix, Table 2

Bing Virtual Earth Basemap adapted by Author 2022

47Bouleutai appeared at four cult sites (Bosra and Canatha, modern-day Qanawāt, as well as Mismiyyeh and Mušannaf) on the road where soldiers also made financial contributions. In addition, bouleutai were benefactors of the cult site at Khabab, which was not situated near a Roman road, and at Shaqrā, which was placed near a hypothetical road. In both cases, no soldiers appear to have been dedicators (Fig. 3). In the case of Khabab, it is unknown which city the bouleutes was from. At Shaqrā, the bouleutes came from the city Philippopolis (present-day Shahbā), a roughly 38-km distance away ; this can indicate that it was a private individual dedication, like the others in the cities, although in the former there is no reference of joint inscriptions with family members; however, this can be due to the fact that the inscription is fragmentary. Individuals appeared to be benefactors of cult sites spread across the Hauran, on roads as well as farther away from them.

  • 67 Southern 2007, 103; Kolb 2015, 664.

48Although soldiers and other benefactors coexisted within the same cult site in some instances, the spatial analysis of soldiers and other benefactors as temple dedicators suggests that soldiers were stationed along the roads and were more mobile than veterans and village communities who, instead, made dedications in clusters of sites next to each other and not often near the road. Although soldiers’ dedications concentrated where they were stationed, at Mismiyyeh and Bosra, they also appeared in cult sites on the road connecting these two stations and other roads. Roads were important passages used by soldiers.67

Benefactors and gods: a network analysis

49By including an additional attribute within the network analysis (i.e., the gods), it is possible to discern the diversity and plurality of deities worshipped at cult sites where soldiers made dedications. This implies copresence between deities worshipped by the Roman army and other benefactors, and, ultimately, respect on the part of soldiers for other benefactors, their deities, and other cults more generally (Fig. 6). Tyche and gods of individuals seem to be quite popular deities at cult sites. However, the cult of Tyche cannot be linked to soldiers, or any specific benefactor category, whereas gods of individuals seem to be linked with village communities. A soldier only commissioned a temple to Tyche at Ṣanamein, and another soldier a door of a temple at ‘Atīl dedicated to the god of an individual having an additional epithet, Theandrios.

Fig. 6: Graph showing the relationship between benefactors (colour of the line, and if it is grey it means that the inscription’s commissioner does not appear), their gods, and cult sites. (IOMH is the usual abbreviation for Jupiter Optimus Maximus Heliopolitanus, I have used JOM for Jupiter Optimus Maximus). Data from Appendix, Table 2 (Author 2022)

Fig. 6: Graph showing the relationship between benefactors (colour of the line, and if it is grey it means that the inscription’s commissioner does not appear), their gods, and cult sites. (IOMH is the usual abbreviation for Jupiter Optimus Maximus Heliopolitanus, I have used JOM for Jupiter Optimus Maximus). Data from Appendix, Table 2 (Author 2022)
  • 68 Kaizer 2007, 449.
  • 69 Sartre 2017, 175; IGLS XVI.2, 311a.
  • 70 Seyrig 1941, 44, Pl.4 No. 2; Sourdel 1952, 89; Pollard, Berry 2012, 158.

50Soldiers, on the other hand, seemed to commission dedications to Zeus Kyrios and gods that have Jupiter Optimus Maximus or Ammon in their divine formula (Fig. 6). Jupiter Optimus Maximus was foremost a god of Rome,68 but also a god of soldiers.69 Jupiter Ammon was the main god of the Legio III Cyrenaica, based at the city Bosra in southern Hauran.70

  • 71 Sartre-Fauriat 2015, 307.
  • 72 Will 1952, 67-68 ; Turcan 1989, 238-239; IGLS XVI.1, 272a.
  • 73 Beck 1996, 178; Gordon 2009, 379-450.
  • 74 IGLS XVI.1, 156.

51Except for a Greek stele from Suwaidā’ dedicated to Hammon (which would have been Jupiter Ammon in Latin), inscriptions dedicated to gods with Jupiter Optimus Maximus or Ammon in their divine formula were in Latin. Although soldiers used Greek, as shown by epigraphy from cult sites in the Hauran, Latin appeared to be used only by soldiers or foreign functionaries in the Hauran,71 suggesting that these inscriptions and the deities mentioned in these inscriptions were related to soldiers and can be go under the umbrella of military gods. An inscription to add to this group of deities is the one from Sī‘ on the inscribed relief of Mithras, which mentions Sol Invictus.72 Mithras was widely but not exclusively worshipped by soldiers across the Empire.73 This inscription was also in Latin. Sol Invictus and Mithras were not worshipped by non-military benefactors according to epigraphic evidence from cult sites in the region. Gods with Jupiter Optimus Maximus or Ammon in their divine formula were only worshipped by soldiers at cult sites located in the Hauran (Fig. 6). Exceptions are inscriptions where the name of the god is the only text inscribed on the relief of Mithras from Sī‘, and the Greek stele from Suwaidā’ (Fig. 6). The dedication to Jupiter Optimus Maximus from Qanawāt was commissioned by an individual named Marcus Aurelius. Sartre-Fauriat and Sartre have suggested that he could have been a soldier or imperial official, given that it was a Latin inscription,74 and the role of this person might be missing because the inscription is fragmentary.

  • 75 Sourdel 1952, 19-31; Mazzilli 2018, 61-64.
  • 76 Sourdel 1952, 44 fn. 5.
  • 77 Regarding discussion about clusters in network analysis, see Newman 2006.

52Dedications to Zeus Kyrios were, on the other hand, not exclusively commissioned by soldiers; the commissioners were individuals, and in one instance by a veteran together with other individuals. Zeus Kyrios has been associated with the pre-existing cult of Baalshamin, as Baalshamin was the main celestial god worshipped in the Hauran.75 Only Aramaic dedications to this Lord of Heaven have been found in the region (Appendix, Table 2). In one instance, a temple was dedicated to the god Lord (Bel) at Shaqqā, but the inscription was in Greek.76 Epithets, such as Kyrios, do not always appear with Zeus in the Hauran, and Zeus is also associated with other epithets like a toponym or Megistos (Appendix, Table 2, Fig. 7). A two-mode network between deities having Zeus in their divine formula or an epithet usually associated with Zeus, and the cult sites where these deities were worshipped, shows that Zeus Kyrios, or Zeus Megistos, was also venerated at the cult sites where Baalshamin was worshipped, creating a cluster through modularity (Fig. 7).77 There is no connection, on the other hand, between cult sites where Baalshamin was worshipped and the sanctuaries where the deity was named generically as a god and given the epithet Keraunos or Megistos: they belong to different clusters from the one with Baalshamin (Fig. 7).

Fig. 7: A weighted graph showing the cult of Baalshamin at cult sites in relation to the cult of Zeus, Zeus with various epithets, and these epithets preceded by the generic god-term. Data from Appendix, Table 2 (Author 2022)

Fig. 7: A weighted graph showing the cult of Baalshamin at cult sites in relation to the cult of Zeus, Zeus with various epithets, and these epithets preceded by the generic god-term. Data from Appendix, Table 2 (Author 2022)

53Zeus was already worshipped with these epithets in the pre-provincial period (Fig. 8). A comparison between pre-provincial two-mode networks and the networks between benefactors and their deities in the provincial period enables us to easily see that soldiers, and occasionally veterans, adopted the pre-existing cult of Zeus, specifically Zeus Kyrios (Fig. 6, 8).

Fig. 8: A weighted graph showing the cult of deities worshipped at the pre-provincial cult sites. Data from Appendix, Table 2 (Author 2022)

Fig. 8: A weighted graph showing the cult of deities worshipped at the pre-provincial cult sites. Data from Appendix, Table 2 (Author 2022)

54The adoption of the cult of the pre-provincial god by soldiers complemented their main patronage, which consisted of supporting pre-existing religious programmes of villages.

55Additionally, according to the network between benefactors and deities, Zeus Megistos was often, but not always, venerated at sanctuaries where soldiers acted as dedicators (Qanawāt, Hit, Sī‘and Ḥibrān) (Fig. 6). Furthermore, this network shows the correlation between Zeus Megistos and military gods and between Zeus Megistos and soldiers. For example, where there is a dedication to Zeus Megistos at Qanawāt, there is also an inscription dedicated to Jupiter Optimus Maximus and Sol Invictus, respectively. A temple was dedicated to Zeus Megistos at at Mismiyyeh where there are multiple dedications by soldiers, including statues of Isis and Pax. These figures imply the acceptance and, possibly, the worship of Zeus with the epithet Megistos by Roman soldiers, although there is no direct written evidence for this cult having been adopted by soldiers.

56Therefore, the network between benefactors and their deities in the provincial period, together with the pre-provincial religious network, shows that soldiers «brought» new deities (military gods) into the region. It also shows that they rarely worshipped deities mainly venerated in the provincial period (Tyche and a god of an individual, in one instance), while worshipping the pre-existing deity Zeus Kyrios, who was also worshipped by other benefactors in the provincial period (Fig. 6, 8).

Benefactors and gods: a spatial analysis

  • 78 IGR III, 1144; PAES III, 797/1; Brünnow, von Domaszewski 1909, 309; IGLS XV.1, 103.

57The link between soldiers and military gods can be strengthened even further by the fact that military gods and Zeus Kyrios were worshipped along Roman roads or in their immediate proximity, except in Ṣūr al-Lejā’ (Fig. 9), which was a village in southern Leja with a long-term military presence since the pre-provincial period.78

Fig. 9: Map showing the cult sites where the main deities were worshipped in the road network. Data from Appendix, Table 2

Fig. 9: Map showing the cult sites where the main deities were worshipped in the road network. Data from Appendix, Table 2

Bing Virtual Earth Basemap adapted by Author 2021

  • 79 IGLS XV.1, 7.

58Of interest is that all the inscriptions dedicated to Zeus Kyrios and Megistos in the provincial period are in Djebel al-Arab, except for the inscription in Mismiyyeh (Fig. 9-10). In the last case, it is inscribed on a cult statue’s base. It has been suggested that in this temple, Zeus Megistos can refer to Zeus of Phaina, named after the village, as an altar dedicated to this god was recovered at a different cult site in Leja, Dāmā-Dāmit al-‘Aliyyah.79 Dedications to Zeus Kyrios and Megistos were also mainly found at cult sites on Roman roads or in their immediate proximity, except in Btheineh and Sahwet al-Khoḍr (Fig. 10). In the first case, a sanctuary in honour of this deity might have been built, but due to the fragmentary nature of the inscription this remains uncertain. In the second case, temple-treasurers, including a veteran, commissioned a building, most likely a temple to this god. Therefore, there seems to be a connection between Zeus with epithets (not just Kyrios, but also Megistos) and Roman roads, confirming the link between Roman roads and soldiers.

Fig. 10: Map showing PPA of cult sites where the main deities were worshipped. Data from Appendix, Table 2

Fig. 10: Map showing PPA of cult sites where the main deities were worshipped. Data from Appendix, Table 2

Bing Virtual Earth Basemap by Author 2022

59In contrast to gods associated with soldiers, gods of individuals were often not worshipped at cult sites situated on Roman roads (Fig. 9). The cults of gods of individuals, seem, instead, to have mainly coalesced into two broad clusters: southern Leja and the north of Djebel al-Arab, except for Dā’il, but the name of this god and his belonging to this group are dubious, like the example from Tall Aḥmar (see Methodology Section) (Fig. 10). The cluster in Leja matches neither with the clusters of veterans, nor with the group of pre-provincial settlements along the hypothetical road in southern Leja; dedications to gods of individuals were made during the provincial period. When looking more carefully at the clusters of gods of individuals, it is possible to notice a cluster of four sanctuaries next to one another in southern Leja, where a god named after the individual Aumos was worshipped (Appendix, Table 2). This deity was also venerated at Shahbā, on almost the same longitude as the sites in the southern part of Leja (Fig. 10).

60The detachment between the cult of Tyche and soldiers is further emphasized by the fact that Tyche was not associated with soldiers as often as other deities worshipped in sanctuaries situated on known Roman roads (Fig. 9). At first glance, the cult of Tyche seems to spread across the Hauran, without concentrating specifically in one part of the region. By undertaking PPA, it is possible to notice the formation of clusters of cult sites where Tyche was worshipped. A cluster of five cult sites can be identified in the southwestern part of Leja, and one of four sites in the north of Djebel al-Arab. Tyche was worshipped in two sanctuaries immediately next to each other in the south and southwest of the region (Fig. 10). These figures can suggest that the cult of Tyche was often spread by cult sites next to each other; thus, the vicinity of cult sites could have played an important role in expanding the cult of Tyche.

Concluding remarks

61On the basis of the multi-levelled network and spatial analyses of temple benefactors from local administration, as well as gods worshipped in cult sites, undertaken in this article, Roman soldiers seemed to have a distinctive profile within the religious environment of the Hauran. They differed from other benefactors, especially those from the local administration, in terms of the location of their cult sites as well as their deities. Roman soldiers were active members of the religious community of cult sites, especially along Roman roads. They commissioned an extraordinary number of dedications to a variety of deities, often, but not exclusively, to Zeus Kyrios and military gods. The primary exceptions to this trend were a couple of cult sites not on a road network, such as Ṣūr al-Lejā’ and Mseikeh in southern Leja, where veterans, nevertheless, played a prominent role in their temples, alongside soldiers, especially in the case of Mseikeh.

62In contrast with soldiers, cult sites where village community as a single unit and village magistrates, and, to a certain extent, veterans, were benefactors of cult sites that were often not situated on Roman roads. This can also be said for the cults of gods of individuals and of Tyche, which were not frequently deities associated with soldiers. Village communities and cults of gods of individuals and Tyche formed small clusters in the region. The formation of these small clusters can suggest that proximity between cult sites might have played a role in the movement of veterans and village communities, their financial involvement in cult sites and the spread of the cults of Tyche and gods of individuals. Nevertheless, both the cult of Tyche and village magistrates as benefactors were still relatively widely spread across the region.

63The contrast between soldiers and their deities on the one hand, and village community and magistrates, and, to a certain extent, veterans on the other, can also indicate that veterans and village communities used and crossed Roman roads less often than soldiers, who were also stationed along the roads where they participated in public religious life. The lack of strong links between soldiers and the cult of Tyche or gods of individuals would also imply different locations for their cults from where soldiers made dedications.

64At the same time, this analysis has highlighted that soldiers did not have a prominent role in the public religious life of the region, as they did not build more cult sites or major parts of a building’s structure than other benefactors from the local administration; rather, they supported pre-existing religious building programme initiated by the local community. They carried on making dedications to the previously venerated deity Zeus Kyrios.

65Furthermore, some major cult sites display copresence of Roman soldiers and other social agents, and indicate respect on the part of soldiers towards other cults, based on the diversity of deities worshipped at cult sites where soldiers made dedications. Bouleutai were often benefactors of sanctuaries on a road where soldiers were also benefactors. It is common to find the copresence of soldiers and veterans as benefactors at the same cult site. Village communities, soldiers, and veterans made dedications to the cult centre at Ṣūr al-Lejā', whereas the village’s inhabitants and a soldier commissioned the temple at Mismiyyeh. Both cult sites were situated in Leja, which highlights the variety of social agents in this part of the region. Even in these two examples (i.e., Mismiyyeh, Ṣūr al-Lejā’), while soldiers worshipped their god, they respected deities, like the personal god, worshipped by a given village community (e.g., in Ṣūr al-Lejā’) or the god named after a village, like Zeus of Phaina (e.g., in Mismiyyeh). The copresence of different social agents and different deities at the same cult sites in Leja has revealed the complexity of religious centres there, possibly triggered by the presence of Roman soldiers and veterans in the area in addition to the presence of local village communities as active agents in their temples.

  • 80 Examples of the study of onomastics or kinships through networks are in Da Vela 2019; Brughmans et (...)

66This article has shown the potential and the validity of applying network and spatial analyses to the study of cult sites, and the need to discuss the naming of deities in relation to their actors (temple benefactors, in this case) and landscape (the proximity between cult sites, as well as between these sites and roads). It has laid the basis of such a study with a twofold methodology that leads us to investigate even further the impact of Roman soldiers on cult sites and, more generically, the role of social agents as well as of the landscape, not only in naming deities, but also in the formation of cult sites. Specifically, the investigation of the role of soldiers as benefactors of cult sites in the Hauran can be extended by integrating a larger dataset into this study, which these sites can offer. As yet unexplored are the iconographic evidence of deities, architectural features of cult sites, and the onomastics and kinships of soldiers and individuals.80 This study of the role of soldiers in cult sites in the Hauran has stressed the significance of investigating the interplay between deities’ names, social agents, and the proximity between cult sites and between these sites and roads, which has been facilitated through the joint network and spatial analysis.

67Appendix

68The dataset used in this research article is deposited at the DataverseNO repository as Supporting Data for: Roman Soldiers in the Religious, Social, and Spatial Network of the Hauran. It includes data tables of sites, inscriptions, benefactors, and gods. Additionally, it includes georeferenced vector data of road networks, cult sites, and PPA analysis. More information can be found in the 00_Mythos_Data_Read_Me.text file.

69See doi https://doi.org/​10.18710/​3UF8MQ

Haut de page

Bibliographie

Alpass 2013: P. Alpass, The Religious Life of Nabataea, Leiden 2013.

Amer, Gawlikowski 1984: G. Amer, M. Gawlikowski, « Le sanctuaire impérial de Philippopolis », Damaszener Mitteilungen 2 (1984), 1-16.

Augier, Sartre 2002: Y. Augier, M. Sartre, « Le dieu de Rabbos, maître du ‘temple périptère’ de Canatha », Damaszener Mitteilungen 13 (2002), 125-130.

Bauzou 1985: T. Bauzou, « Les voies de communication dans le Hauran à l’époque romaine », in Hauran I (1985), 137-166.

Bauzou 2003: T. Bauzou, « Le réseau routier de la Syrie du sud (Hawran, Jawlan) », in Hauran II (2003), 289-310.

Beck 1996: R. Beck, « The Mysteries of Mithras », in J. S. Kloppenborg, S. G. Wilson (eds), Voluntary Associations in the Greaco-Roman World, New York 1996, 176-185.

Berry, Pollard 2012: N. Pollard, J. Berry, The Complete Roman Legions, London 2012.

Bolelli 1986: G. Bolelli, « La ronde-bosse de caractère indigène en Syrie du Sud », in Hauran I (1986), 311-372.

Bonnet 2022a: C. Bonnet, « Introduction. Exploring the Intersection between Divine Names and Places» in C. Bonnet, T. Galoppin , E. Guillon , M. Luaces , A. Lätzer-Lasar , S. Lebreton, F. Porzia , J. Rüpke, E. Rubens Urciuoli (eds), Naming and Mapping the Gods in the Ancient Mediterranean Spaces, Mobilities, Imaginaries, Berlin 2022.

Bonnet 2022b: C. Bonnet (dir.), ERC Mapping Ancient Polytheisms Website: https://map-polytheisms.huma-num.fr/ (2022/11/11).

Bonnet et al. 2018: C. Bonnet, M. Bianco, T. Galoppin, É. Guillon. A. Laurent, et al. « Les dénominations des dieux nous offrent comme autant d’images dessinées (Julien, Lettres 89b, 291 b). Repenser le binôme théonyme-épithète », Studi e materiali di storia delle religioni 84/2 (2018), 567-591. HAL Id : hal-01990135 https://hal.archives-ouvertes.fr/hal-01990135

Bounni 1991: A. Bounni, « La stèle de Qadboun », Mélanges de I'École francaise de Rome 103 (1991), 51-55.

Bowersock 1983: G. W. Bowersock, Roman Arabia, Cambridge, 1983.

Braemer et al. 1999: F. Braemer, J.-M. Dentzer, M. Kalos, P. Tondon, « Tours à noyau chemisé de Syrie du Sud », Syria 76 (1999), 151-176.

Broodbank 1993: C. Broodbank, « Ulysses without sails: trade, distance, knowledge and power in the early Cyclades », World Archaeology 24.3 (1993), 315-331.

Brughmans, Peeples 2017: T. Brughmans, M. A. Peeples, « Trends in Archaeological Network Research », Journal of Historical Network Research 1 (2017), 1-24.

Brughmans et al. 2021: T. Brughmans, O. Bobou, N. B. Kristensen, R. R. Thomsen, J. V. Jensen, R. Raja. « A kinship network analysis of Palmyrene genealogies », Journal of Historical Network Research (2021). DOI: https://doi.org/10.25517/jhnr.v6i1.65

Brünnow, von Domaszewski 1909: R. E. Brünnow, A. von Domaszewski, Die Provincia Arabia, Strasbourg 1909.

Burns 1999: R. Burns, The Monuments of Syria, London 1999.

Cantineau 1932: J. Cantineau, Le Nabatéen, Paris 1932.

Collar 2013: A. C. F. Collar, Religious Networks in the Roman Empire: The Spread of New Ideas, Cambridge 2013.

Davis 1982: J. Davis, « Thoughts on Prehistoric and Archaic Delos », Temple University Aegean Symposium 7 (1982), 23-33.

Da Vela 2015: R. Da Vela, « Applicazione della Social Network Analysis per lo studio della trasmissione culturale e dell’evoluzione delle identità culturali locali durante la tarda età ellenistica », Atti e memorie dell’Accademia Toscana di Scienze e Lettere La Colombaria 79-80 (2015), 205-230.

Da Vela 2019: R. Da Vela, « Interlocking Networks and the Sacred Landscape of Hellenistic Northern Etruria: Capturing Social and Geographic Entanglement Through Social Network Analysis », Open Archaeology 5.1 (2019), 505-518. DOI: https://doi.org/10.1515/opar-2019-0031

Dentzer 1986: J.-M. Dentzer, « Conclusion : développement et culture de la Syrie du Sud dans la période préprovinciale Ier s. avant J.-C. – Ier s. après J.-C.) », in Hauran I (1986), 387-429.

Dentzer 1991: J.-M. Dentzer, « Hellénisation et cultures indigènes dans la Syrie intérieure : l’exemple du sanctuaire de Sî’ » in Ho Hellēnismos stēn Anatolē : praktika 1. Diethnous Archaiologikou Synedriou, Delphoi, 6-9 Noemvriou 1986, Athena 1991, 269-276.

Dentzer 1999: J.-M. Dentzer, « L'espace des tribus arabes à l'époque hellénistique et romaine : nomadisme, sédentarisation, urbanisation », Comptes rendus des séances, Académie des Inscriptions et Belles-Lettres 143.1 (1999), 231-271.

Dentzer et al. 1993: J.-M. Dentzer, P.-M. Blanc, R. Mukdad, A. Mukdad, « Nouvelles recherches franco-syriennes dans le quartier est de Bosra Ash-sham », Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres 137.1 (1993), 117-147.

Dentzer et al. 2007: J.-M. Dentzer, P.-M. Blanc, T. Fournet, « Le développement urbain de Bosra de l'époque nabatéenne à l'époque byzantine : bilan des recherches françaises 1981-2002 », in J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, A. Mukdad (eds), Bosra : aux portes de l’Arabie, Beyrouth 2007, 31-38.

Dentzer-Feydy 1979: J. Dentzer-Feydy, « À propos du temple dit de ‘Dusares’ à Sî’ », Syria 56 (1979), 325-332.

Dentzer-Feydy 1986: J. Dentzer-Feydy, « Décor architectural et développement du Hauran dans l'Antiquité (du Ier s. av. au VIIe s. de notre ère) », in Hauran I (1986), 260-309.

Dentzer-Feydy 1990: J. Dentzer-Feydy, « Les chapiteaux corinthiens normaux de Syrie méridionale », Syria 67 (1999), 633-663.

Dentzer-Feydy 1992: J. Dentzer-Feydy, « Les linteaux à figures divines en Syrie méridionale », Revue Archéologique (1992), 65-102.

Dentzer-Feydy 1993: J. Dentzer-Feydy, « Introduction de l'acanthe dans la sculpture monumentale du Proche-Orient à l'époque greco-romaine », in L. Pressouyre (éd.), L' Acanthe dans la sculpture monumentale de l'Antiquité à la Renaissance, (colloque de Paris, Sorbonne, 1er-5 octobre 1990), Paris 1993, 98-112.

Dentzer-Feydy 1998: J. Dentzer-Feydy, « Remarques sur le temple de Rimet Hazim (Syrie du Sud) », Syria 75 (1998), 201-211.

Dentzer-Feydy 2003: J. Dentzer-Feydy, « Le sanctuaire », in Hauran II (2003), 40-111.

Dentzer-Feydy 2007: J. Dentzer-Feydy, « Les dieux », in J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, A. Mukdad (eds), Bosra : aux portes de l’Arabie, Beyrouth 2007, 41-43.

Dentzer-Feydy 2008: J. Dentzer-Feydy, « Le décor architectural des maisons de Batanée », in Hauran III (2008), 183-232.

Dentzer-Feydy 2010: J. Dentzer-Feydy, « Les sanctuaires païens de type régional en Syrie du sud », in Hauran V (2010), 225-238.

Dentzer-Feydy 2015: J. Dentzer-Feydy, « New Archaeological Research at the Sanctuary of Sî’ in Southern Syria: the Graeco-Roman Divinities invite themselves to Baalshamîn », in M. Blömer, A. Lichtenberger, R. Raja, (eds), Religious Identities in the Levant from Alexander to Muhammed: Continuity and Change, Turnhout 2015, 313-325.

Dentzer-Feydy, Nehmé 2007: J. Dentzer-Feydy, L. Nehmé, « Les dieux avant la province d’Arabie », in J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, A. Mukdad (eds), Bosra : aux portes de l’Arabie, Beyrouth 2007, 18-20.

Donceel, Sartre 1997: R. Donceel, S. Sartre, « Théandrios, dieu de Canatha », Electrum 1 (1997), 21-34.

Dunand 1926: M. Dunand, « Rapport sur une mission archéologique au Djebel Druze », Syria 7 (1926), 326-335.

Dunand 1930: M. Dunand, « La voie romaine du Ledjâ », Extrait des Mémoires présentés par divers savants à l’Académie des Inscriptions et Belles Lettres 13.2 (1930), 521-527.

Dunand 1933: M. Dunand, « La voie romaine du Ledjâ », Mémoires présentes par divers savants à l’Académie des inscriptions et belles-lettres de l’Institut de France 13.2 (1933), 521-557.

Dunand 1934: M. Dunand, Mission archéologique au Djebel Druze : le musée de Soueïda, inscriptions et monuments figurés, Paris 1934.

Dunand 1950: M. Dunand, « Nouvelles inscriptions du Djebel Druze et du Hauran », Archiv orientální 18 (1950), 144-164.

Dussaud 1941: R. Dussaud, « Le saint dieu Paqeidas », Syria 22,3/4 (1941), 295-297.

Dussaud 1955: R. Dussaud, La pénetration des Arabes en Syrie avant L’Islam, Paris 1955.

Dussaud, Macler 1903: R. Dussaud, F. Macler, Mission dans le régions désertiques de la Syrie Moyenne, Paris 1903.

Fiaccadori 1999: G. Fiaccadori, « Nuova dedica a Dusares da Bosra », Felix Ravenna (1999), 145-148.

Freyberger 1989: K. S. Freyberger, « Das Tychaion von aş-Şanamain », Damaszener Mitteilungen 4 (1989), 87-108.

Freyberger 1991: K. S. Freyberger, « Der Tempel in Slim: ein Bericht », Damaszener Mitteilungen 5 (1991), 8-38.

Freyberger 1998: K. S. Freyberger, Die frühkaiserzeitlichen Heiligtümer der Karawanenstationen im hellenisierten Osten, Mainz am Rhein 1998.

Freyberger 2000: K. S. Freyberger, « Qanawat. Der Südtempel (Tempel des Zeus Megistos): Deutung und Funktion im städtischen Kontext von Kanatha », Damaszener Mitteilungen 12 (2000), 155-175.

Freyberger 2010: K. S. Freyberger, « Bedeutung und Funktion der Heiligtümer im städtischen Kontext des antiken Kanatha », in Hauran V (2010), 239-255.

Freyberger 2015: K. S. Freyberger, « Continuity and Change of Religious Life in Southern Syria during the Hellenistic and Roman Periods », in M. Blömer, A. Lichtenberger, R. Raja, (eds), Religious Identities in the Levant from Alexander to Muhammed: Continuity and Change, Turnhout 2015, 283-296.

Fulminante 2014: F. Fulminante, « The Network Approach: Tool or Paradigm? », in K. Felder, S.C. Evans (eds), Archaeological Review from Cambridge 29.1 (2014), 165-176.

Gawlikowski 1990: M. Gawlikowski, « Les dieux de Palmyre », in von H. Temporini, W. Haase (eds), ANRW 18.4, Berlin, 2605-2658.

Glomb et al. 2018: G.T. Glomb, A. Mertel, Z. Pospíšil, Z. Stachoň, A. Chalupa, « Ptolemaic military operations were a dominant factor in the spread of Egyptian cults across the early Hellenistic Aegean Sea », PLoS ONE 13.3 (2018). DOI: https://doi.org/10.1371/journal.pone.0193786

Glomb et al. 2020: T. Glomb, A. Mertel, Z. Pospíšil, A. Chalupa, « Ptolemaic political activities on the west coast of Hellenistic Asia Minor had a significant impact on the local spread of the Isiac cults: A spatial network analysis », PLoS ONE 15.4 (2020). DOI: https://doi.org/10.1371/journal.pone.0230733

Glomb 2021: T. Glomb, « The Spread of the Cult of Asclepius in the Context of the Roman Army Benefited from the Presence of Physicians: A Formal Network Approach », PLoS ONE 16.8 (2021). DOI: https://doi.org/10.1371/journal.pone.0256356

Gordon 2001: R. Gordon, « Trajets de Mithra en Syrie romaine », Topoi 11.1 (2001), 77-136.

Gordon 2009: R. Gordon, « The Roman Army and the Cult of Mithras: a Critical View », in C. Wolff (ed.), L’Armée Romaine et la Religion sous le Haut-Empire Romain, Lyon 2009, 379-450.

Healey 2001: J. Healey, The Religion of the Nabataeans: A Conspectus, Leiden 2001.

Hunt 1988: T. L. Hunt, « Graph theoretic network models for Lapita exchange: a trial application », in P.V. Kirch, T.L. Hunt (eds), Archaeology of the Lapita Cultural Complex: A Critical Review, Seattle 1988, 135-155.

Kaizer 2007: T. Kaizer, « Religion in the Roman East », J. Rüpke (ed.), A Companion to Roman Religion, Malden 2007, 446-456.

Kalos 1997: M. Kalos, « Le site de Saḥr (Syrie du Sud) », Topoi 7 (1997), 965-991.

Kalos 1999: M. Kalos, « Un sanctuaire d'époque hellénistique en Syrie du Sud : Khirbet Massakeb », Topoi 9 (1999), 777-794.

Kalos 1999: M. Kalos, « Un sanctuaire de Mithra inédit en Syrie du sud », Topoi 11.1 (2001), 229-277.

Kalos 2003: M. Kalos, « Le Site de Saḥr al-Lağāt (Syrie du sud) », in K. S. Freyberger, A. Henning, H. von Hesberg (eds), Kulturkonflikte im Vorderen Orient an der Wende vom Hellenismus zur römischen Kaiserzeit, Rahden/Westf. 2003, 157-167.

Knappett 2011: C. Knappett, An Archaeology of Interaction. Network perspectives on Material Culture and Societies, Oxford 2011.

Knox et al. 2006: H. Knox, M. Savage, P. Harvey, « Social Networks and the Study of Relations: Networks as Method », Metaphor and Form. Economy and Society 35.1 (2006), 113-140. DOI:10.1080/03085140500465899

Kolb 2015: A. Kolb, « Communications and Mobility in the Roman Empire », in C. Bruun, J. Edmondson (eds), Oxford Handbook of Roman Epigraphy, Oxford 2014, 649-670.

Kropp 2010: A. J. M. Kropp, « Limits of Hellenisation: Pre-Roman basalt temples in the Hauran », Bollettino di Archeologia on line (International Congress of Classical Archaeology Meetings between cultures in the ancient Mediterranean, Rome 2008), 2010, 1-18 (http://151.12.58.75/archeologia/bao_document/articoli/1_KROPP.pdf).

Kropp 2013: A. J. M. Kropp, Images and Monuments of Near Eastern Dynasts, 100 BC–AD 100, Oxford 2013.

Kubiak-Schneider 2021: A. Kubiak-Schneider, « Maître de l'Univers, du Monde, de l'Éternité : des dieux aux pouvoirs illimités à Palmyre ? », in C. Bonnet (ed.), Noms des dieux : Portraits de divinités antiques, Toulouse 2021, 133-151.

Lebreton 2019: S. Lebreton, « Des dieux et des personnes : une approche par les réseaux (mondes grecs et ouest-sémitiques, ca. 1000 av. n. è. – 400 d. n. è.) », in Cinquième rencontre du groupe Res-Hist (Réseaux & Histoire) « La personne en question dans les réseaux » (Rennes, 17-18 octobre 2019).

Mascle 1944: J. Mascle, Le Djebel Druze, Beyrouth 1944.

Mazzilli 2014: F. Mazzilli, Beyond Religion: Cultural Exchange and Economy in Northern Phoenicia and the Hauran, Syria, PhD Thesis. University of Durham 2014.

Mazzilli 2018: F. Mazzilli, Rural Cult Centres in the Hauran: Part of the Broader Network of the Near East, Oxford 2018.

Mazzilli 2021: F. Mazzilli, « The Pantheon of the Hauran, A Dialectical Dialogue », in F. Mazzilli, D. Van Der Linde (eds), Dialectics of Religion in the Roman World, Potsdamer Altertumswissenschaftliche Beiträge, Stuttgart 2021, 239-274.

Mazzilli 2023: F. Mazzilli, « A Decade of Religious Networks in the Pre-Roman and Roman periods: An assessment of their Methodology », in R. Da Vela, M. Franceschini, F. Mazzilli (eds), Networks as Resources for Ancient Communities, Tübingen 2023.

McLean Harper 1928: G. McLean Harper, « Village administration in the Roman Province of Syria », Yale Classical Studies 1 (1928), 105-168.

Milik 1958: J. Milik, « Nouvelles inscriptions nabatéennes », Syria 35 (1958), 227-251.

Milik 1972: J. Milik, Dédicaces faites par des dieux (Palmyre Hatra Tyr) et des thiases sémitiques à l’époque romaine, Paris 1972.

Milik 2003: J. Milik, « Une bilingue araméo-grecque de 105/104 avant J.-C. », in Hauran II (2003), 269-275.

Millar 1993: F. Millar, The Roman Near East 31 BC- AD 337, Cambridge-London 1993.

Nehmé 2005-2006: L. Nehmé, « Inscriptions vues et revues à Madāʾin Ṣāliḥ », Arabia 3 (2005-2006), 179-225.

Nehmé 2010: L. Nehmé, « Les inscriptions nabatéennes du Ḥawarān », in Hauran V (2010), 451-493.

Newman 2006: M. E. J. Newman, « Modularity and community structure in networks », Proceedings of the National Academy of Sciences of the United States of America 103.23 (2006), 8577–8696.

Parker 2017: R. Parker, Greek Gods Abroad: Names, Natures, and Transformations, Oakland, CA 2017.

Pollard 2000: N. Pollard, Soldiers, cities, and civilians in Roman Syria, Ann Arbor 2000.

Purcell 1990: N. Purcell, « The Creation of Provincial Landscape: the Roman impact on Cisalpine Gaul », in T. Blagg, M. Millett (eds), The Early Roman Empire in the West, Oxford 1990, 7-29.

Rivers 2016: R. Rivers, « Can Archaeological Models Always Fulfil our Prejudices ? », in T. Brughmans, A. Collar, F. Coward, (eds), The Connected Past: Challenges to Network Studies in Archaeology and History, Oxford 2016, 123-148.

Sartre 1981: M. Sartre, « Le territoire de Canatha », Syria 58 (1981), 343-357.

Sartre 1985: M. Sartre, Bostra, des origines à l’Islam, (BAH 117), Paris 1985.

Sartre 1987: M. Sartre, « Villes et villages du Hauran (Syrie) du Ier au IVe siècle », in E. Frézouls (ed.), Sociétés urbaines, sociétés rurales dans l'Asie Mineure et la Syrie hellénistiques et romaines, (Actes du colloque organisé à Strasbourg (novembre 1985) par l'Institut et le Groupe de recherche d’histoire romaine et le Centre de recherche sur le Proche-Orient et la Grèce antiques), Strasbourg 1987, 239-257.

Sartre 1991: M. Sartre, L’Orient romain : provinces et sociétés provinciales en Méditerranée orientale d’Auguste aux Sévères (31 avant J.-C-235 après J.-C.), Paris 1999.

Sartre 1993: M. Sartre, « Communautés villageoises et structures sociales d'après l'épigraphie de la Syrie du sud », in A. Calbi, A. Donati, G. Poma (eds), L'epigrafia del villaggio, Faenza 1993, 117-135.

Sartre 1999: M. Sartre, « Les metrokomiai de Syrie du Sud », Syria, 76 (1999), 197-222.

Sartre 2003: M. Sartre, « Inscriptions grecques de Sī‘ 8 et des environs », in Hauran II (2003), 281-282.

Sartre 2007: M. Sartre, « Le cadre historique », in J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, A. Mukdad (eds), Bosra : aux portes de l'Arabie, Beyrouth 2007, 9-15.

Sartre 2011: M. Sartre, Bostra (Supplément) et la plaine de la Nuqrah. Inscriptions grecques et latines de la Syrie, XIII 2, Beyrouth 2011.

Sartre 2017: M. Sartre, « Panthéons civiques du Hauran », in R. Raja (ed.), Contextualizing the Sacred in the Hellenistic and Roman near East Religious Identities in Local, Regional, and Imperial Settings, Turnhout 2017, 170-180.

Sartre-Fauriat 1999: A. Sartre-Fauriat, « Les notables et leur rôle dans l’urbanisme du Hauran à l’époque romaine », in C. Petitfrère (ed.), Construction, reproduction et représentation des patriciats urbains de l’Antiquité au XXe siècle, (Actes du colloque de 7, 8 et septembre 1998 tenu à Tours), Tours 1999, 223-240.

Sartre-Fauriat 2004: A. Sartre-Fauriat, Les voyages dans le Hawran (Syrie du Sud) de William John Bankes 1816 et 1818, Beyrouth 2004.

Sartre-Fauriat 2005: A. Sartre-Fauriat, « Les soldats, un élite en Syrie à l’époque impériale » in A.J. Los, K. Nawotka (eds.), Elite in Greek and Roman Antiquity, Antiquitas 28, Wroclaw 2005, 117-132.

Sartre-Fauriat 2007: A. Sartre-Fauriat, « Inscriptions inédites pour la Tyche en Syrie du Sud », Mélanges de l’Université Saint-Joseph 59 (2007), 1-20.

Sartre-Fauriat 2015: A. Sartre-Fauriat, « Nouveaux dieux et dieux nouveaux dans le Hauran (Syrie Du Sud) à l’époque romaine », in M. Blömer, A. Lichtenberger, R. Raja, (eds), Religious Identities in the Levant from Alexander to Muhammed: Continuity and Change, Turnhout 2015, 297-311.

Sartre-Fauriat 2017a: A. Sartre-Fauriat, « La vie religieuse sur le Trachôn à l’époque romaine – Les apports de l’épigraphie », in R. Raja (ed.), Contextualizing the Sacred in the Hellenistic and Roman near East Religious Identities in Local, Regional, and Imperial Settings, Turnhout 2017, 181-190.

Sartre-Fauriat 2017b: A. Sartre-Fauriat, « Les inscriptions », in Hauran IV.1 (2017), 327-350.

Sartre-Fauriat, Sartre 2014: A. Sartre-Fauriat, M. Sartre, Le plateau du Trachon et ses bordures. Inscriptions grecques et latines de la Syrie, XV, Beyrouth 2014.

Sartre-Fauriat, Sartre 2016: A. Sartre-Fauriat, M. Sartre, La Batanée et le Jawlān Oriental. Inscriptions grecques et latines de la Syrie, XIV.1, Beyrouth 2016.

Schlumberger 1933: D. Schlumberger, « Les formes anciennes du chapiteau corinthien en Syrie, en Palestine et en Arabie », Syria 14 (1933), 283-317.

Segal 2008: A. Segal, « Religious Architecture in the Roman Near East: Temple of the Basalt Lands (Trachon and Hauran) », in T. Kaizer (ed.), The Variety of Local Religious Life in the Near East in the Hellenistic and Roman Period, Leiden 2008, 97-132.

Segal 2013: A. Segal, Temples and Sanctuaries in the Roman East: Religious Architecture in Syria, Iudaea/ Palaestina and Provincial Arabia, Oxford 2013.

Seyrig 1941: H. Seyrig, « Les inscriptions de Bostra » Syria 22 (1941), 44-48.

Seyrig 1949: H. Seyrig, « Antiquités syriennes : 41. Nouveaux monuments palmyréniens de Baalshamin », Syria 26 (1949), 29-41.

Seyrig 1971: H. Seyrig, « Antiquités syriennes : 93. Bel de Palmyre », Syria 48 (1971), 89-100.

Sourdel 1952: D. Sourdel, Les cultes du Hauran à l’époque romaine, Paris 1952.

Southern 2007: P. Southern, The Roman Army: a Social and Institutional History, Oxford 2007.

Speidel 1984: M. P. Speidel, Roman Army Studies, 1, Amsterdam 1984.

Speidel 1998: M. P. Speidel, « Legio IIII Scythica, its movements and men », in D. Kennedy (ed.), The Twin Towns of Zeugma on the Euphrates: Rescue Work and Historical Studies, Ann Arbor 1998, 163-232.

Starcky 1981: J. Starcky, « Pétra et la Nabatène », Supplément au Dictionnaire de la Bible 7 (1981), 886-1017.

Starcky 1986: J. Starcky, « Les inscriptions nabatéennes et l’histoire de la Syrie méridionale et du nord de la Jordanie » , in Hauran I (1986), 167-181.

Steinsapir 2005: A. I. Steinsapir, Rural sanctuaries in Roman Syria: the creation of a sacred landscape, British Archaeological Reports International Series 1431, Oxford 2005.

Terrell 1977: J. E. Terrell, « Human biogeography in the Solomon Islands », Fieldiana Anthropology 68.1 (1977), 1-47.

Tholbecq 2007: L. Tholbecq, « Hérodiens, Nabatéens et Lagides dans le Hawrân au Ier s. av. J.-C. : réflexions autour du sanctuaire de Ba'alšamîn de Sîa », Topoi 15.1 (2007), 285-310.

Turcan 1989: R. Turcan, Les cultes orientaux dans le monde romain, Paris 1989.

Weber 2003a: T. M. Weber, « Sahr al-Ledja. La statuaire d’un sanctuaire tribal en Syrie du Sud et ses relations roman mésopotamiennes », Topoi Supplément 4 (2003), 349-377.

Weber 2003b: T. M. Weber, « Ein Denkmal der Herodier in der syrischen Basaltwüste », in K. S. Freyberger, A. Henning, H. von Hesberg (eds), Kulturkonflikte im Vorderen Orient an der Wende vom Hellenismus zur römischen Kaiserzeit, Rahden/Westf. 2003, 257-276.

Weber 2006: T. M. Weber, Sculptures from Roman Syria in the National Museum at Damascus: From Cities and Villages of Central and Southern Syria, 1, Worms 2006.

Weber 2007: T. M. Weber, « La sculpture », in J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, A. Mukdad (eds), Bosra : aux portes de l’Arabie, Beyrouth 2007, 44-52.

Wenning 1983: R. Wenning, « Das Ende des Nabatäischen Königreiches », in A. Invernizzi, J.-F. Salles (eds.), Arabia Antiqua. Hellenistic Centres around Arabia, Rome 1993, 81-103.

Will 1952: E. Will, « Nouveaux monuments sacrés de la Syrie romaine », Syria 29 (1952), 60-73.

Will 1985: E. Will, « Un problème d'interpretatio graeca : la pseudo-tribune d'Echmoun à Sidon » Syria 62.1-2 (1985), 105-124. DOI: https://doi.org/10.3406/syria.1985.6877

Witcher 1998: R. Witcher, « Roman Roads: Phenomenological Perspectives on Roads in the Landscape », Theoretical Roman Archaeology Journal 97 (1998), 60-70. DOI: https://doi.org/10.16995/TRAC1997_60_70

Haut de page

Annexe

Abbreviations

AAES II = H. C. Butler, Publications of an American Archaeological Expedition to Syria 1899-1900. II. Architecture and Other Arts, Leiden 1904.

AAES III= W. K. Prentice, Publications of an American Archaeological Expedition to Syria in 1899-1900. III. Greek and Latin Inscriptions, Leiden 1908.

AAES IV= E. Littmann, Publications of the Princeton University Archaeological Expeditions to Syria in 1904-1905 and 1909. IV (a–D). Semitic inscriptions, Leiden 1914.

CIG = Corpus Inscriptionum Graecarum, Berlin 1825-1877.

CIL = Corpus Inscriptionum Latinarum, Berlin 1863.

CIS = Corpus Inscriptionum Semiticarum Ab Academia Inscriptionum Et Litterarum Humaniorum Conditum Atque Digestum Ii, Inscriptiones Aramaicae, Paris 1881.

Hauran I = J.-M. Denzter (ed.), Hauran I : recherches archéologiques sur la Syrie du Sud à l’époque hellénistique et romaine, Paris 1985-1986.

Hauran II = J. Dentzer-Feydy, J.-M. Dentzer, P.-M. Blanc (eds), Hauran II : les installations de Sī‘ 8 du sanctuaire a l’établissement viticole, Beyrouth 2003.

Hauran III = P. Clauss-Balty (ed.), Hauran III : l'habitat dans les campagnes de Syrie du sud aux époques classique et médiévale, Beyrouth 2008.

Hauran IV.1 = J. Dentzer-Feydy, J.-M. Dentzer, F. Renel, A. Sartre-Fauriat (eds), Hauran IV.1 Sahr Al-Leja Volume I : le sanctuaire et l’agglomération à l’époque romaine. Recherches syro-franco-allemandes 1998-2008, Beyrouth 2017.

Hauran IV.2 = J-M. Dentzer, Jean-Marie, T. M. Weber, Hauran IV : Sahr Al-Ledja : recherches syro-européennes 1998-2008, 2. Die Skulpturen aus Sahr und die Statuendenkmäler der römischen Kaiserzeit in südsyrischen Heiligtümern, Beyrouth 2009.

Hauran V = M. Al-Maqdissi, F. Braemer, J.-M. Dentzer (eds.), Hauran V : la Syrie du Sud du néolithique à l’Antiquité tardive : recherches récentes. Actes du colloque de Damas 2007, Beyrouth 2010.

Hauran VI = J. Rohmer, Hauran VI : d’aram à Rome la Syrie du Sud de l’âge du Fer à l’annexion romaine (XIIe siècle av. J.-C. – Ier siècle apr. J.-C.), Beyrouth 2020.

IGLS XIII.1 = M. Sartre, Inscriptions Grecques et Latines de la Syrie, XIII, No.1, Bostra N° 9001 a 9472, Paris 1982.

IGLS XIII.2 = M. Sartre, Bostra (Supplément) et la plaine de la Nuqrah. Inscriptions recques et Latines de la Syrie, XIII, No. 2, Beyrouth 2011.

IGLS XIV.1 = A. Sartre-Fauriat, M. Sartre, La Batanée et le Jawlan Oriental. Inscriptions Grecques et Latines de la Syrie, XIV No.1, Beyrouth 2016.

IGLS XIV.2 = A. Sartre-Fauriat, M. Sartre, La Batanée et le Jawlan Oriental. Inscriptions Grecques et Latines de la Syrie, XIV No.2, Beyrouth 2016.

IGLS XV.1 = A. Sartre-Fauriat, M. Sartre, Le plateau du Trachon et ses bordures. Inscriptions Grecques et Latines de la Syrie, XV No.1, Beyrouth 2014.

IGLS XV.2 = A. Sartre-Fauriat, M. Sartre, Le plateau du Trachon et ses bordures. Inscriptions Grecques et Latines de la Syrie, XV No.2, Beyrouth 2014.

IGLS XVI.1 = A. Sartre-Fauriat, M. Sartre, L’Auranitide : Qanawāt et la bordure nord du Jebel al-‘ Arab. Inscriptions Grecques et Latines de la Syrie, XVI No.1, Beyrouth 2020.

IGLS XVI.2 = A. Sartre-Fauriat, M. Sartre, L’Auranitide : Suwaydā' et la bordure ouest du Jebel al-‘ Arab. Inscriptions Grecques et Latines de la Syrie, XVI No.2, Beyrouth 2020.

IGLS XVI.forth = A. Sartre-Fauriat, M. Sartre, L’Auranitide. Inscriptions Grecques et Latines de la Syrie, XVI, Beyrouth forthcoming.

IGR = Inscriptiones Graecae Ad Res Romanas Pertinentes, Paris 1901-1927.

PAES II = H.C. Butler, Syria. Publications of the Princeton University Archaeological expeditions to Syria in 1904-5 and 1909. Ancient Architecture in Syria. II. Ancient Architecture in Syria. Section A. Southern Syria Section, Leiden 1919-1920.

PAES III= E. Littmann, Publications of the Princeton University Archaeological Expeditions to Syria in 1904-1905 and 1909. vol. III Greek and Latin inscriptions, Leiden 1918.

PAES IV= E. Littmann, Publications of an American Archaeological Expedition to Syria in 1899-1900. IV. Semitic inscriptions, Leiden 1914.

RAO = C. Clermont-Ganneau, Recueil d’archéologie orientale, Paris 1888-1924.

RES = J.-B. Chabot, C. Clermont-Ganneau, Répertoire d’épigraphie sémitique : publié par la commission du Corpus Inscriptionum Semiticarum, Paris 1905.

Suw. = J.-M. Dentzer, J. Dentzer-Feydy. Le Djebel Al-Arab : histoire et patrimoine au musée de Suweida, Paris 1991.

Wadd. = W. H. Waddington, Voyage archéologique en Grèce et en Asie Mineure. Pt. 6. Syrie, Paris 1870.

Haut de page

Notes

1 Pollard 2000, 1.

2 Ibid.

3 These figures are the result of the work published by Sartre-Fauriat (2005, 118-119).

4 Sartre-Fauriat 2005, 130-131.

5 Sartre-Fauriat 2005, 125-127; 2015, Mazzilli 2018, 88-94.

6 In these instances, inscriptions were in Latin, and it has been already discussed that this language was used in inscriptions commissioned by soldiers or foreign functionaries in the Hauran (Sartre-Fauriat 2015, 307). For gods associated with the army see Mazzilli 2018, 95-98; 2021, 249-250, for identification of imported deities see Sartre-Fauriat 2015, 307 or deities with a Graeco-Roman name see Mazzilli 2018, 94-95; 2021, 248-249.

7 For example, Glomb et al. 2018; Glomb et al. 2020; Glomb 2021.

8 Mazzilli 2023. For example, see ERC Mapping Ancient Polytheisms Project (Bonnet 2022a; 2022b).

9 Mazzilli 2018.

10 IGLS XIII.1-2; XIV.1-2, XV.1-2, XVI.1-2, XVI.forth.

11 Hauran VI.

12 The preponderance of rural cult sites is due to the fact that the region seems to be based on a village administrative system (Mazzilli 2014, 234); where village had a certain level of autonomy (McLean Harper 1928; Sartre 1987, 251). Three sites (Suwaidā’, Shahbā and Shaqqā) became cities only in the second and third centuries (Sartre 1999, 213, 216).

13 For further information about the site typology see Methodology Section.

14 There are a couple of examples of inscriptions by soldiers at cult sites from the pre-provincial period, but they have not been included in this analysis.

15 Josephus, AJ 17.28.

16 Millar 1993, 95-96.

17 The dating probability of the dataset is not included in the analysis but may be found in the Appendix (Table 2).

18 AAES II, 405; IGLS XVI.1, 191.

19 IGLS XVI.forth, 751.

20 Sartre-Fauriat 2005, 130-131.

21 Sartre 1991, 328, 252.

22 Sartre 1987, 244.

23 Sartre 1993, 113-135.

24 I have elected to use modern names, as we do not have names of ancient villages and sites for all the sites under analysis.

25 IGLS XV.1, 90 for Zebīreh and IGLS XV.1, 172 for Ezra‘.

26 MacLean Harper 1928, 105-116; Sartre 1993, 120-121.

27 Sartre-Fauriat 1999, 223-240.

28 McLean Harper 1928,123-127.

29 Bonnet et al. 2018; Belayche 2017.

30 Network analysis with each part of the divine formula (e.g., theonym, epithets) seems a valuable method to fully understand the complexity of the naming deities in the Hauran, but it is not undertaken due to a different focus of the paper, which is on the Roman army. For this reason, a discussion of the epithet ‘ancestral’ is not pursued in this paper, which is used with a variety of theonyms. This epithet is recorded in the Appendix Table 2 and it is occasionally used for Fig. 6, 8.

31 Kubiak-Schneider 2021.

32 Sartre-Fauriat 2007, 18

33 IGLS XIII.2, 9571.

34 Sourdel 1952, 69-74.

35 IGLS XVI.2, 304, 307a.

36 Will 1985, 105-125; IGLS XVI.2, 304.

37 See further discussion on the meaning Mazzilli 2018, 66-67; Mazzilli 2021.

38 IGLS XVI.2, 307a.

39 This has been extensively discussed elsewhere; see, for instance, Sartre-Fauriat 2015; Mazzilli 2018; 2021. For further, broader, discussion of religious cultural identities in the Hauran see Hauran I, II (especially conclusive chapter by Dentzer 1986); Mazzilli 2014; 2018.

40 For the use of ‘gods of individuals’ who have alternatively also been called ‘personal gods’, see Parker 2017, 113-132.

41 IGLS XIV.1, 349; IGLS XVI.2, 382d.

42 Kaizer 2007, 449.

43 Brughman, Peeples, 2017, 2; Knappett 2011, 9; Fulminante 2014; Knox et al. 2006.

44 Da Vela 2015, 229, for examples of two-mode networks and their meaning.

45 Rivers 2016, 133.

46 Broodbank 1993; Davis 1982; Hunt 1988; Terrell 1977; Collar 2013.

47 Bauzou 1985; 2003.

48 Witcher 1997; Kolb 2015, 649. For display of power over landscape see Purcell 1990, 15, 23.

49 Speidel 1984, 691-692.

50 IGLS XV.1, 17, 18, 19; Sartre-Fauriat, Sartre 2014, 36.

51 Sartre 2017, 175.

52 For Canatha see Sartre 1981; Denzter 1986, 396-67; for Bosra see Bowersock 1983, 73 ff.; Sartre 1985; Dentzer 1986, 406-07; Dentzer et al. 2007, 31-38; Sartre 2007, 9-12.

53 Speidel 1984, 691-692.

54 For the Herodian garrison at Ṣūr al-Lejā’in the pre-provincial period see IGR III, 1144; PAES III, 797/1; Brünnow, von Domaszewski 1909, 309; IGLS XV.1, 103. For dedication by a soldier at the end of the first century AD at Ḥibrān see IGLS XVI.2, 431a

55 Segal 2008, 110; Sartre-Fauriat, Sartre 2014, 37.

56 IGLS XV.1, 5.

57 Sartre-Fauriat, Sartre 2014, 37.

58 IGLS XV.1, 11.

59 IGLS XV.1, 17, 18, 19.

60 Sartre-Fauriat, Sartre 2014, 36.

61 IGLS XV.1, 13.

62 Sartre-Fauriat, Sartre 2014, 34.

63 IGLS XV.1, 140a.

64 Wadd., 2367; AAES III, 428b; IGLS XVI.1, 263.

65 IGR III, 1144; PAES III, 797/1; Brünnow, von Domaszewski 1909, 309; IGLS XV.1, 103.

66 Hauran VI, 508-509.

67 Southern 2007, 103; Kolb 2015, 664.

68 Kaizer 2007, 449.

69 Sartre 2017, 175; IGLS XVI.2, 311a.

70 Seyrig 1941, 44, Pl.4 No. 2; Sourdel 1952, 89; Pollard, Berry 2012, 158.

71 Sartre-Fauriat 2015, 307.

72 Will 1952, 67-68 ; Turcan 1989, 238-239; IGLS XVI.1, 272a.

73 Beck 1996, 178; Gordon 2009, 379-450.

74 IGLS XVI.1, 156.

75 Sourdel 1952, 19-31; Mazzilli 2018, 61-64.

76 Sourdel 1952, 44 fn. 5.

77 Regarding discussion about clusters in network analysis, see Newman 2006.

78 IGR III, 1144; PAES III, 797/1; Brünnow, von Domaszewski 1909, 309; IGLS XV.1, 103.

79 IGLS XV.1, 7.

80 Examples of the study of onomastics or kinships through networks are in Da Vela 2019; Brughmans et al. 2021.

Haut de page

Table des illustrations

Titre Fig. 1: Map of cult sites from the pre-provincial and provincial period within the road network. Data from Appendix, Tables 1, 2, 3
Crédits Bing Virtual Earth Basemap adapted by Author 2022
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-1.jpg
Fichier image/jpeg, 286k
Titre Fig. 2: A weighted graph showing relationships between benefactors (squares) and cult sites (circles) according to inscriptions from the provincial period. The colour reflects the benefactor category, and when the circle, standing for the cult site, is wholly coloured according to one of the benefactor categories, it means that the benefactor commissioned the temple or a main component of the building. Data from Appendix, Table 2 (Author 2022)
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-2.jpg
Fichier image/jpeg, 148k
Titre Fig. 3: Map showing the cult sites with different benefactor categories in the road network. Data from Appendix, Table 2
Crédits Bing Virtual Earth Basemap, adapted by Author 2022
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-3.jpg
Fichier image/jpeg, 308k
Titre Fig. 4: Map showing PPA of cult sites where benefactors commissioned dedications. Data from Appendix, Table 2
Crédits Bing Virtual Earth Basemap adapted by Author 2022
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-4.jpg
Fichier image/jpeg, 297k
Titre Fig. 5: Map showing PPA of cult sites, using roads, where benefactors commissioned dedications. Data from Appendix, Table 2
Crédits Bing Virtual Earth Basemap adapted by Author 2022
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-5.jpg
Fichier image/jpeg, 308k
Titre Fig. 6: Graph showing the relationship between benefactors (colour of the line, and if it is grey it means that the inscription’s commissioner does not appear), their gods, and cult sites. (IOMH is the usual abbreviation for Jupiter Optimus Maximus Heliopolitanus, I have used JOM for Jupiter Optimus Maximus). Data from Appendix, Table 2 (Author 2022)
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-6.jpg
Fichier image/jpeg, 172k
Titre Fig. 7: A weighted graph showing the cult of Baalshamin at cult sites in relation to the cult of Zeus, Zeus with various epithets, and these epithets preceded by the generic god-term. Data from Appendix, Table 2 (Author 2022)
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-7.jpg
Fichier image/jpeg, 68k
Titre Fig. 8: A weighted graph showing the cult of deities worshipped at the pre-provincial cult sites. Data from Appendix, Table 2 (Author 2022)
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-8.jpg
Fichier image/jpeg, 44k
Titre Fig. 9: Map showing the cult sites where the main deities were worshipped in the road network. Data from Appendix, Table 2
Crédits Bing Virtual Earth Basemap adapted by Author 2021
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-9.jpg
Fichier image/jpeg, 309k
Titre Fig. 10: Map showing PPA of cult sites where the main deities were worshipped. Data from Appendix, Table 2
Crédits Bing Virtual Earth Basemap by Author 2022
URL http://journals.openedition.org/mythos/docannexe/image/5802/img-10.jpg
Fichier image/jpeg, 288k
Haut de page

Pour citer cet article

Référence électronique

Francesca Mazzilli, « Roman Soldiers in the Religious, Social, and Spatial Network of the Hauran »Mythos [En ligne], 16 | 2022, mis en ligne le 15 décembre 2022, consulté le 06 décembre 2023. URL : http://journals.openedition.org/mythos/5802 ; DOI : https://doi.org/10.4000/mythos.5802

Haut de page

Auteur

Francesca Mazzilli

Department of Archaeology, History, Cultural Studies and Religion, University of Bergen, Postboks 7805 5020 Bergen, Norway, Francesca.Mazzilli(at)uib.no

Royal Holloway, University of London, Egham, Surrey TW20 0EX, United Kingdom, Francesca.Mazzilli(at)rhul.ac.uk

Haut de page

Droits d’auteur

CC-BY-NC-ND-4.0

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.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search