Dalmatians and Dacians—Forms of Belonging and Displacement in the Roman Empire

Abstract

Inspired by the catalyst papers, this essay traces the impact of displacement on existing and emerging identities of groups and individuals which were relocated to ‘frontier’ areas in the aftermath of conflict and conquest by Rome during the reign of emperor Trajan. The Dacian Wars, ending in 106 CE with the conquest of Dacia by Roman armies, not only resulted in the deliberate destruction of settlements and the society of the conquered, but also the removal of young Dacian men by forced recruitment into the Roman army, some serving the emperor in the Eastern Egyptian Desert. In turn, the wealth in gold and silver of the newly established Roman province of Dacia was exploited by mining communities arriving from Dalmatia. As a result of these ‘displacements’ caused by war and the shared experience of mining in the remote mountains of Dacia or guarding roads through the desert east of the Nile, we can trace the emergence of new senses of belonging alongside the retainment of fixed group identities.

The term ‘displacement’ initially evokes wholly negative connotations; of loss, alienation, precariousness; it evokes televised images of deadly conflict, social and ethnic tension, economic and ecological collapse, of people amassing on boats and borders; images of sprawling refugee camps, ramshackle sheds, of dirt, disease and hopelessness. The catalyst papers, however, provide an angle which focuses not on the dynamic moments of forced migration and the causes. Instead, their emphasis is on the comparatively static consequences for refugees or those people refered to as economic migrants establishing themselves in new environments, be that through the assignment of space in ordered camps in Jordan, or through the self-directed occupation of land on the peripheries of Brazilian cities.

The example of the Dandara community (Ribeiro et al. 2017) near Belo Horizonte is thought-provoking: it established itself on vacant property as a result of the costliness of available living space in the city and far larger socio-economic dynamics still at play in Brazil. Most striking is the self-organization of this community, not only in terms of resistance against authorities, but crucially so in the establishment of a communal sense of belonging. This is expressed through the choice of a place name, Dandara, a female protagonist in the historic slave revolts of Brazil. Issues of belonging and community formation abound even more so in the case of the Zaatari camp in northern Jordan near the Syrian border (Dalal 2017). There, refugees from Syria are confronted with the social practices and habits of compatriots from different parts of the country. Whilst the commonality of being ‘Syrian’ and a refugee brings them together in their peripheral camps, there is also a recognition of diversity in the practices—a recognition often obscured by the perception of those seeking refuge being a homogenous group. Both examples concern the creation of a place, either by the common will of the displaced, or the assignment by the governing authority of a place to people who are displaced. The ensuing (re-)negotiation of social positions amongst those party to these ‘new’ communities raise questions that make us sensitive to the predicament of people in new communities in earlier contexts, as those from the Roman world. The rich evidence particularly from Dacia (c. modern Romania) and Roman Egypt, allow for the further exploration of these issues, in relation to a different type of mobile community that often had to exist on the margins—that of the miners and soldiers.

The conquest of Dacia by the Roman army under emperor Trajan in 106 was the culmination of lengthy wars which began in the late 1st century CE. The integration of this region into the empire was accompanied by a significant movement of people in and out of the newly established Roman province of Dacia. In the case of certain cavalry and infantry soldiers from Dacia and miners from the province of Dalmatia, we can trace their relocation after 106 CE to the desert forts in the Eastern Egyptian Desert and to the gold mining district of Alburnus Maior in the Apuseni mountains of Romania, respectively, in the textual evidence (funerary epitaphs, votive altars, graffiti on rockfaces, letters or contracts written on pottery sherds, so-called ostraca, or wood/wax tablets). The perception that Dalmatians and Dacians were forcibly relocated or resettled does not arise necessarily from reading ancient texts in Greek and Latin, but derives from the scholarly interpretations of the written evidence: Stanislaw Mrozek, for instance, used the German verb ‘umsiedeln’, implying the resettlement or relocation of Dalmatians by the Roman state to a mining district far from home—and his view has proven rather influential.1 Whether the arrival of Dalmatians at a remote mining district was orchestrated by Roman authorities is in need of further investigation. As for the Dacians drafted into the Roman army during and in the immediate aftermath of the war, many of them did not have a say, not least about the location they would be seconded to.2 The term ‘displacement’ therefore certainly applies here, whereas the phenomenon of Dalmatians in Alburnus Maior might also be explored through the concept of labour mobility. The latter allows for individuals and communities from Dalmatia to relocate to the gold-mining district in Dacia out of their own will in response to economic push-and-pull factors.

What makes both phenomena comparable, and links them with the cases of the Zaatari camp or the Dandara community is the negotiation of identities in the formation of (new) communities in the aftermath of forced relocations to peripheral zones, be that the Apuseni mountains of Dacia, the Eastern Egyptian Desert, Jordan’s northern border, or the urban periphery of Belo Horizonte in Brazil. Rather than undertaking a direct comparison of ancient and modern phenomena, both catalyst papers inspire the re-evaluation of archaeological and written evidence related to ‘displacement’ in Roman Dacia in the 2nd c. CE in the light of the same themes—forced relocation to, and identity formation on, the periphery.3 As we shall see, the displacement, the confrontation with other people, and the shared experience in an unaccustomed, if not hostile, landscape is instrumental in the emergence of a sense of commonality.

Even though geographically distant, both Alburnus Maior and the forts along the roads to the Red Sea are comparable in that they are within a ‘frontier’, a concept originally devised by Frederick Jackson Turner. Since its inception in a paper in 1893, this concept has been decontextualised and developed further to encompass comparable processes throughout the world of land grabbing and exploitation of natural resources.4 Scholarly attempts at delineating the concept ‘frontier’ have not necessarily resulted in one handy definition: a ‘frontier’ is not simply something locatable on a map, it can also encompass dynamic social and economic processes. They unfold in a space of cultural interactions between two or more distinct societies in an asymmetric power relationship with one invading society coming from the outside often to the detriment of the invaded. These frontier processes are the result of private initiative rather than driven by the state.5

This definition is perhaps not adequate enough in framing the context in which the mining community at Alburnus Maior or the military society in the Eastern Egyptian Desert find themselves. The concept ‘frontier’, is not necessarily limited to a definable physical zone between two political or social entities but can entail an ecological dimension: Turner had already thought of ‘mining frontiers’ as a distinct category (Osterhammel suggests ‘resource-extraction frontiers’) and, more recently, ‘frontier of settlement’ has been used in terms of claiming agricultural land or the extraction of valuable resources from the wilderness on the periphery of settled and developed land.6 It is this context the Dalmatians at Alburnus Maior find themselves in and which provides a specific socio-economic and ecological backdrop for the expression of identities. In this respect the Dacians in the forts east of the Nile valley found themselves on a ‘frontier’ in the more common sense of the term: it can also describe a void or zone which is not controlled or cannot be controlled by the state beyond an area or line of military defence and from which threats to the security and order can emerge (Osterhammel 2014, p. 328). As we shall see, this concept of frontier applies more adequately to the Dacians relocated to these forts, from where they watched over a transient community of merchants, camel drivers, quarrymen, along with their families or prostitutes travelling on desert roads between the Nile and the Red Sea—a constant ebbing and flowing of goods and people punctuated, according to ancient authors, by sudden attacks of the Bedouin seemingly out of nowhere. It is this specificity of geographical place, of the real or perceived remoteness from settled, agriculturally developed, and secure areas, along with the anxieties and hopes affixed to these places on the fringe of tamed and wild nature, which seems best captured by the concept ‘frontier’.7 In applying this approach to the inscribed evidence for Dalmatian miners in Alburnus Maior and Dacian soldiers in the Eastern Egyptian Desert this paper seeks to re-examine and provide a new interpretation of the known epigraphic material in light of displacement.

We shall first turn our gaze to Dacia and explore why Dalmatians established themselves at a remote mining site in a distant province, examine remnants of ethnic ‘divisions’ amongst them, whilst probing the epigraphic evidence for signs of an emerging sense of community. The absence of Dacians from inscribed monuments at Alburnus Maior raises the question of what happened to the native population of the recently conquered province.8 The little evidence we have points towards the enrolment of parts of their young men into the Roman army and their secondment to the Eastern Egyptian Desert. The effects of displacement on these men form the second half of my investigation. As we shall see, both Dalmatians in Dacia and Dacians in Egypt retained aspects of their identities whilst developing shared senses of belonging shaped by their new existence, remote from home.

The focus of this study is on texts written on stone, on pottery sherds, or on wooden tablets, as these sources offer a more intimate and immediate reading of the sense of difference or belonging an individual or a community could experience, than the examination of material evidence might allow for. The close reading of these different genres of textual evidence, however, is limited by social, religious, or legal conventions. The full name of a person including patronym and ‘nickname’ or an ‘ethnic’ self-designation, for instance, might only be relevant in the context of signing a contract or when inscribed on an epitaph. In other words, the personal or communal identities expressed in these texts might not be of importance in everyday encounters with others, where a ‘nickname’ might suffice. With this caveat in mind, let us turn to the evidence from Alburnus Maior.

1. Alburnus Maior, Roman Dacia

The grab for natural resources, which drove the exploitation of gold lodes at Alburnus Maior (Roşia Montană) in the Apuseni mountains of Romania, followed upon the bloody conquest of the Dacians in 106 CE by Roman armies under the emperor Trajan. The Dacians, unified under king Decebalus, were seen as threatening neighbours operating for some three decades to the north of the Roman Danube provinces. The repeated Roman campaigns under Domitian and Trajan were mainly a response to this external threat posed by the Dacian tribal federation led by, first, Diurpaneus, and then Decebalus. These wars ended with the incorporation of Dacia into the Roman empire in the absence of any viable alternatives (Speidel 2009, pp. 140–50; Strobel 2010, p. 89ff.).

Apart from the deployment of a provincial garrison with the dual task of securing the border and pacifying the conquered population (Piso 2008, pp. 303–12), Rome began to establish a legal and fiscal framework in this new province—an administration headed by a (military) governor (Piso 2008, pp. 308–9). As in other provinces created by Rome post-conquest, the provincial administration oversaw the (re-)assignment of territory, the establishment of new communities, and instituted control over the mineral resources of the land. The latter is evident in post-conquest Northwestern Spain, the short-lived Augustan provincia Germania, or in Britain, where dated ingots and archaeological evidence appear to indicate the commencement, if not the intensification of metal resource exploitation. Whether this was initially driven by the state or by private individuals (i.e. Roman citizens and/or individual provincials) is hard to fathom.9

This grab for natural resources is also evidenced in the new province of Dacia. At Ampelum (modern Zlatna), some 21 km southeast of Alburnus Maior, the administrative headquarters of the imperial procurator in charge of goldmining in Dacia seems to have been established soon after 106 CE.10 More importantly, at Roşia Montană recent surveys and excavations have allowed us to trace Roman mining tunnels and shafts and the associated settlements. Finds of wooden beams, ladders, even segments of a waterwheel, provide dendrochronological dates for when the mines were in operation; it appears the mines were up and running soon after 106 CE (Damian 2003; Cauuet and Tămaş 2012, pp. 235–37). In the late 18th and early 19th century, miners found inscribed wooden tablets in some of these shafts. In and around Roşia Montană funerary epitaphs and votive altars were recovered (most of which date to the 2nd century CE).11

The vast majority of men and women mentioned in these inscriptions and documentary texts have non-Roman names. Their names in fact are ‘Illyrian’, that is, a fragmentary language (surviving in personal names and place names) assumed by modern linguists to have been spoken in the Western Balkans in the Roman province of Dalmatia and Pannonia, an area once identified as Illyria, that is, the ‘home’ of the Illyrians.12

1.1. Displacement or Labour Mobility?

The circumstances by which these ‘Illyrians’ from Dalmatia arrived in the new province are not quite clear. Eutropius, writing in the later 4th century CE, claims that “Dacia defeated, Trajan brought in countless masses of people from the whole Roman world to till the soil and inhabit cities.”13 He also suggests that “Dacia, in fact, had been exhausted of men.”14 The influx of people from other provinces to Alburnus Maior and other parts of the newly established provincia Dacia is explained as a settlement movement initiated by Trajan. In the case of the Dalmatian miners at Alburnus Maior, prevailing scholarly opinion presumes their presence to be the result of a forced relocation due to their mining expertise.15

The literary and epigraphic evidence certainly supports the notion of Dalmatians being directly involved in the exploitation of gold and other metals within their own province. Pliny the Elder notes that in Dalmatia during the reign of Nero gold was detected near the surface; in Statius’ ode to Vibius Maximus, Dalmatia is again identified as a source of gold, as it is in one of Martial’s epigrams.16 Florus claims that after the Pannonian Wars the provincial governor, on the orders of the emperor Augustus, forced the Dalmatians to work in the goldmines; the zeal and diligence with which they exploited the deposits, seemed as if they were extracting it for their own gain (Flor. 2.25). Although Roman mining sites per se have not been unmistakeably identified as of yet, the scholarly consensus is that mining for mineral resources was surely undertaken during the Roman period as well—this is strongly suggested by literary and epigraphic sources.17 We may therefore presume that, initially, some Dalmatians arrived with their own expertise in mining at Alburnus Maior.

As of yet, there is no clear literary or documentary evidence for a forced resettlement of Dalmatians. Although Florus tells of Asturians (in Northwestern Spain) and Dalmatians being forcefully relocated to exploit the gold deposits in their homeland, this pertains to the period in the immediate aftermath following conquest under Augustus (Flor. 2.25, 2.33.60). The relocation of Dalmatians to Alburnus Maior might have been prompted by the invitation of the emperor to come to Dacia, the promise of wealth, and the prospect of improving one’s social status.18 Whether or not this movement was facilitated by a decline of mining in Dalmatia or other push factors awaits a more profound study of mining sites there.

1.2. ‘Illyrians’ in Alburnus Maior

The wooden tablets found in the mining shafts—twenty-five inscribed with texts in Latin (one in Greek)—concern various legal arrangements and contracts. The names of those either party to the contract or acting as witnesses, reveals a diverse make-up in terms of legal status and origins. Besides men with the full tria nomina, that is, men who according to prevalent scholarly opinion are Roman citizens, the texts also render the names of peregrini, i.e., members of non-Roman communities within the Roman Empire. The chronological range of these texts stretches from 131 to 167 CE, with the bulk dating to the 160s. The texts refer mostly to economic activity in and around Alburnus Maior and further afield (Russu 1975, pp. 165–256; Noeske 1977, pp. 336 ff.). These wooden tablets allow us to capture a segment of local society which would not necessarily appear in the inscribed funeral and votive monuments discovered on site, either because those mentioned were unable to afford funerary epitaphs, their presence in the goldmining district was only temporary, or they only had loose connections to the district (Noeske 1977, p. 336).

One such contract in Latin documents the sale of an enslaved girl by Dasius Verzonis to a Maximus Batonis and was signed on 17 March 139 CE at Kartum; here the information relevant arises at the beginning and end of the document:

Maximus Batonis puellam nomine | Passiam, sive ea quo alio nomine est an|norum sex [above line: circiter p(lus) m(inus) empta sportellaria] emit mancipioque accepit | de Dasio Verzonis Pirusta ex Kaviereti[o] | * ducentis quinque. … dari fide rogavit | Maximus Batonis, fide promisit Dasius || Verzonis Pirusta ex Kaviereti[o]. Proque ea puella, quae s(upra) s(cripta) est, * ducen|tos quinque accepisse et habere | se dixit Dasius Verzonis a Maximo Batonis …

Maximus son of Bato has bought and accepted as a mancipium a girl by name Passia, or if she is (known) by any other name, m(ore or) l(ess) around six years old, having been bought as a foundling, for 205 (denarii), from Dasius son of Verzo, a Pirustian from Kavieretium. … Maximus the son of Bato asked to be given in faith, Dasius son of Verzo a Pirustian from Kavieretium promised in faith. Dasius son of Verzo said that he received and has for this girl, w(ho) i(s) w(ritten) a(bove), 250 denarii from Maximus son of Bato. Done at Kartum on the 16th day before the Kalends of April when (emperor) Titus Aelius Caesar Antoninus Pius and Bruttius Praesens were consuls (for the second time)… 19

Our focus here is not on the legal implications of this contract, nor on the slave herself (which raises a wholly different aspect of displacement), but on those party to the contract. Dasius Verzonis (read as Dasius, son of Verzo) is understood to be an ‘Illyrian’ name, i.e. the name ‘Dasius’ as well as the name of his father (the patronym) ‘Verzo’ are both attested in the Roman province of Dalmatia. More importantly, Dasius Verzonis is identified as Pirusta ex Kaviereti[o], a man of the Pirustae, a civitas in the southern parts of Dalmatia, from Kavieretium.20 Regarding Maximus son of Bato (Maximus Batonis), Maximus is a widely popular Latin cognomen and Bato an Illyrian name seems to be typical of Pannonian tribes in Dalmatia and in Pannonia.21

This contract over the sale of a slave girl also notes a host of witnesses to the contract.22 We may presume that the witnesses to this document were present at the place where the contract was concluded and written, i.e., Kartum, which we suspect was within or near the mining district of Alburnus Maior.23 Given the findspot of the tablet in a mining shaft, the involvement of the contractual parties and the witnesses—if not directly in the mining ‘industry’, then at least in auxiliary services (smelting, provisions, credit, etc.)—is highly likely. The witnesses listed are Maximus Veneti, a princeps; Masurius Messii, a decurio; Anneses Andunocnetis; Planius Verzonis from Sclaietae; Liccaius Epicadi from Marcinium; and Epicadus Plarentis qui et (also known as) Mico. Most have Illyrian names or filiations, indicating their non-Roman or ‘peregrine’ status and their origin, except for Masurius Messii.24 He is noted as a decurio, a member of a local council (ordo decurionum).25

More of a surprise, though, is the designation of Maximus Veneti as a princeps, a ‘first man’ or ‘chief’; this term resonates with the epigraphic evidence from communities in Dalmatia. Epitaphs note principes of the Delmatae and of numerous communities.26 These men have been addressed as the aristocracy of Dalmatian tribes.27 Whether Maximus Veneti had been princeps of a community back in Dalmatia, or whether he attained membership of the local elite and became princeps of a community in or near Alburnus Maior, is not easily decided. The backstory of Maximus Veneti might be very similar to another princeps called T(itus) Aurelius Aper who on a funerary monument from the late 2nd or early 3rd century CE at nearby Ampelum, is noted as a Delmata and princeps adsignatus from the town (municipium) of Splonum in Dalmatia.28 Some assumed that Aurelius Aper, as princeps, was a member of the elite at Ampelum, having arrived there from Dalmatia under Septimius Severus and bringing his estate, including slaves, with him.29

The other witnesses with ‘Illyrian’ names do not provide much further indication of their social status beyond being peregrini.30 To find Illyrian names in this contract is not a surprise at all: as already noted, the vast majority of epitaphs and altars discovered in and near Roşia Montană display ‘Illyrian’ appellations of men and women.31 The case of the princeps Maximus Veneti, however, illustrates that the community was not egalitarian, but stratified—an observation underpinned by the contracts which reveals further socioeconomic distinctions between employers and miners, landowners, bankers even—between have and have-nots.32

1.3. Internal Divisions?

A closer look at the place names, so-called toponyms, mentioned in the wooden tablets and the inscribed votive and funerary monuments reveal further fault lines within this mining community. The contract provides the place name Kartum. Toponyms also emerge in other writing tablets, adding to our knowledge of local geography of this Dacian mining district: one contract, for instance, for the sale of a house is signed at vicus Pirustarum.33

The place names emerging in inscribed funerary epitaphs and monuments devoted to deities deserve further scrutiny: at the cemetery of Țarina in the immediate vicinity of Roşia Montană, a recently published epitaph reveals the following text:34

D(is) M(anibus) | Dasas Liccai (filius) | Del(mata) k(astello) Starvae | vixit an(nis) XXXV |⁵ pos(uerunt) Beucus | Sarius et ḌẠ[…] | […]i heredes b(ene) m(erenti).

To the spirits of the dead. Dasas, son of Liccaius, Dalmatian, from the fortified settlement (kastellum) of the Starvae, lived 35 years. The heirs Beucus, Sarius, and Da[-] set up (this epitaph), well-deserving.

Apart from Dasas, son of Liccaius, being identified as Dalmatian, his home is given as kastellum Starva. Of interest is the toponym kastellum Starva, which appears in a further epitaph set up at Țarina.35 The place name refers to a (fortified) settlement and, what appears to be an ethnonym, i.e., the name of a people called the ‘Starvae’. That these people must be located in Dalmatia, is already indicated by the designation ‘Delmata’. Luckily, an inscribed epitaph for a councillor, a decurio, of the municipium Salvium /Vrba (BiH), who is natus Starve, allows us to locate the Starvae more precisely within the territory of this municipality.36

Similar toponyms consisting of the term kastellum and ethnonyms are mentioned on other funerary stones and on votive altars: on ‘Hăbad’ hill south of Roşia Montană three altars are dedicated either by a k(astellum) Ansi, or by k(astellani) Ansi or Ansi(enses), the inhabitants of the kastellum.37 Two further altars clearly denote a k(astellum) Ansis.38 These votives seem to derive from a sanctuary or sanctuaries on Hăbad, which were excavated together with settlement remains and a cemetery (Damian 2003, pp. 121–57). Moreover, the cemetery yielded a funerary epitaph set up for Platino Verzonis of the k(astellum) Anso by her husband (CIL 3: 1271; Ciongradi 2009, no. 124). The same sanctuary on Hăbad also received an altar commissioned by Dalmatians for the well-being of the Maniates. A k(astellum) Man(iatium?) might be noted on an altar to Diana found at the cemetery of Țarina.39 On Carpeni hill at Roşia Montană a genius of the collegium k(astelli) Baridust(arum) receives a votive altar (AE 1944: 24; Ciongradi 2009, no. 64). We are thus left with a whole series of toponyms consisting of ‘kastellum’+ethnonym (and in one instance ‘vicus’+ethnonym, as in vicus Pirustarum).

A similar phenomenon is also observable for names of associations (collegium, sg.) and deities: altars from Drumuş near Roşia Montană mention a Genius of the collegium Sardiatarum; a Genius Sardiatensium; and a collegium Sardiatensium (AE 2003: 1487, 1488, 1491; Ciongradi 2009, nos. 22, 58, 85). A fragmentary inscription on an altar from Valea Nanului may render a c(ollegium) Sar(diatenisum/diatarum).40 The ethnonyms ‘Sardias’, ‘Sardiata’, or ‘Sardiates’ are documented on inscribed altars and epitaphs, after personal names, such as Plator Sar(dias^?) or Bisius Scenob(arbi) Sard(iata^?) (Ciongradi 2009, 14 and no. 8). In two epitaphs the deceased, both with ‘Illyrian’ names, are noted as Delmatae.41 What is more, an altar is dedicated to the divinity Apto Delmatarum at Valea Nanului, perhaps a water deity of the Dalmatians.42

As we observed already with the Starvae, these ethnonyms are not unknown to us: the term ‘Pirustae’—which describes Dasius, son of Verzo, noted in the contract earlier—also appears in the place name vicus Pirustarum at Alburnus Maior. The tribe is noted as the name of a people by Julius Caesar in his ‘Gallic War’ and, in 54 BCE, is reported to have raided parts of the Roman province Illyricum nearest to them. Livy makes note of the Pirustae as part of the Dassaretii who receive immunity from tax liabilities and political independence in 167 BCE during the 3rd Macedonian War. The ‘Peiroustai’, confusingly, are counted by Strabo amongst the people of the Pannonians. Velleius Paterculus, who served under Tiberius during the Pannonian uprising, marks the Pirustae and Desidiatae as Dalmatian tribes, who, located in remote strongholds on mountains, are finally pacified. No mention of Pirustae is made by Pliny the Elder in his description of civitates in Dalmatia, whereas Ptolemy perceives them to reside in southern Dalmatia together with the Sikoulotai, the Dokleatai, and the Skirtones. Appianus, who wrote his Roman History (Ῥωμαικά) under Trajan and Hadrian, details the Roman wars in Macedonia and Illyria, but makes only brief mention of the ‘Pyrissaioi’ (Pirustae) in the context of a campaign against Dalmatian tribes in 33 BCE.43 As for the region that the Pirustae inhabited, Geza Alföldy suggested northern Albania and the area north of the Albanian Alps.44

The Sardi/eates, who at Alburnus Maior seem to form an association of sorts and whose genius is evoked, are a people also located in Dalmatia: Pliny notes the Sardeates as one of the communities within the conventus of Salona.45 The Sardeates are not mentioned prior to Pliny and Ptolemy in literary sources, which has been taken to suggest the formation of this civitas only after the Roman conquest.46 They have been located near Sarnade/Sarute (in or near Pecka) or west of Jajce in the Pliva valley near Šipovo (BiH).47

The home of the Maniates may be sought near Salona, based on a mention of Μανιοί in the Periplous of Ps-Skylax from the late 4th c. BCE. Although they do not occur in literary or other epigraphic sources of the province, the ethnonym was still of relevance—provided the Μανιοί can be equated with the Maniates.48 The Baridustae likely hail from Dalmatia and inhabited the area around Bariduum/Livno (BiH).49

The Delmatae, noted in two epitaphs at Alburnus Maior, could refer to the province of Dalmatia in general, or perhaps more likely, to the large civitas of the Delmatae within the conventus, the assize, of Salona (Plin. n.h. 3.142). Though not noted explicitly in our written sources, the municipium Salvium (and thus the Starvae) also falls within the civitas of the Delmatae.50 The Ansi appear to be a people from provincia Dalmatia; their territory might be sought around the town of Ansium, located somewhere to the northeast of Corinium / Karin Gornij (CRO) within the civitas of the Liburni (Plin. n.h. 3.139 f.).51

What emerges from this brief survey of the relevant inscribed epitaphs and altars is the provincial landscape of Dalmatia being replicated here en miniature in the kastella and vici in the gold mining district of Alburnus Maior or in its vicinity. We know of the kastella Ansis, Starvae, Baridustarum, and (Maniatium^?); of a vicus Pirustarum; and a collegium Sardiatensium/Sardiatarum, which might hint at the existence of a vicus or kastellum of the same name, as there is also a collegium k(astelli) Baridustarum (see above). Scholarly opinion overwhelmingly locates these kastella and vici in or near Alburnus Maior.52 This opinion is certainly justified for the vicus Pirustarum, which is given as a place where a contract was concluded, and for the kastella Ansis and Baridustarum. Kastellum Ansis not only follows upon a personal name in a funerary inscription and thus renders the origin of an individual buried at Alburnus Maior; it is also named as a community sponsoring votive altars for a variety of deities at Alburnus Maior. And the existence of associations (collegia) of the kastellum Baridustarum and of the Sardiates certainly puts those groups in or near the mining district. In the case of the toponyms kastellum Starvae or Kavieretium/k(astellum) Aviereti(um) it remains uncertain whether they are to be sought near Alburnus Maior or in Dalmatia.53

The latter two toponyms aside, the mention of these settlements illustrates the fragmentation of these people arriving from Dalmatia into different ‘tribal’ communities: the existence of an association (collegium) of Sardiates and of Baridustae, the evocation of a genius of these groupings, and the indication of the kastellum in votive inscriptions indicates that divisions along these lines may have endured. This reinforcement of distinctions between these Illyrian groups could have been driven by their respective assignment to different parts of the mines, but there is no evidential basis for this.

1.4. Shared Experiences

How persistent these initial differences were is difficult to discern. After all, the Illyrian men and women, were either directly or indirectly connected with the unifying purpose of the district seemingly centred at Alburnus Maior—the mining of silver and gold lodes, its processing, the logistics and management of these work procedures, financial services, and other related activities. This and the fact that these people were inhabiting a place remote from the population centres of this new-ish province, a ‘frontier’, must have helped shape a sense of community beyond these internal sub-divisions.

Echoes of this might be identifiable in the divine entities addressed in the votive altars throughout the mining district. These fairly elaborate, if somewhat crudely executed, votive altars were not just symbols of a cultic rite, but were also an act of economic choice.54 In light of the costs of an altar, the choice of divinity to which it would be dedicated was deliberate and freighted with meanings specific to each group or individual commissioning this work. Given the location of the altars in sacred areas within a mining district and the likely occupational background of the devotees, the reading of these votive monuments is informed in part by the local context and, in part, by the scholarly aggregation and ordering of ‘function(s)’ of a specific god, evoked by ancient literature and iconography. Given the wide range of facets ascribed to some ancient deities, the teleological reading of their veneration at Alburnus Maior as closely linked with mining has to be taken with a grain of salt. Other interpretations, of course, remain possible.55

With this in mind, the extraction of subterranean resources in a remote part of a new province may be mirrored in the veneration of chthonic deities by those involved in mining at Alburnus Maior. Terra Mater, for instance, the goddess of the fertile Earth, often referred to in mining districts, is dedicated altars by men with Illyrian names.56 Aeracura and Soranus, both documented only once so far on altars set up at Alburnus Maior, also pertain to the underworld.57 Surio Sumeletis, who had an altar set up to Terra Mater, also commissioned one for Neptune (AE 1990: 845). Neptune might have been called upon due to the regular occurrence of ground water in underground mining; two further altars to Neptune were commissioned by a Roman citizen and by two peregrini, respectively.58 Perhaps Maelantonius and Naos/n both could be addressed as ‘aquatic’ deities as well.59 Given the dangers of underground mining and the processing of ore, dedications to Asclepius must be expected as well.60

Illyrians also set up altars to Diana and Apollo; as astral deities, Diana/Luna and Apollo/Sol represented silver and gold, respectively.61 This particular reading of the Apollo- and Diana-altars is warranted by the metalla-coinage minted under Trajan and his successors, where Apollo and Diana, together with Mars, depict the metals of gold, silver, and iron.62 Dedications to Liber Pater, a god associated with nature, fertility, and wine, may also pertain to the richness of the earth in metals, as may ‘Sidus’, but this is not a given.63 The altars set up to Mercurius, perhaps as god of commerce or as saviour and guarantor of new beginnings^? (under Augustus) could reflect the flourishing commercial activities documented in part in the contracts on writing tablets at Alburnus Maior.64 For the Dalmatians and others coming to Alburnus Maior the aspect of Mercurius as guarantor of new beginnings may have been attractive as well: the attraction of goldmines lie not only with the economic returns mining offers but perhaps also the hope of social mobility.65

The deity Diana also represented a ‘liminal’ quality as demarcating the line between ‘inside’ and ‘outside’, between the wilderness of nature and the order of the civitas; she also invokes connotations with the hunt and animals and can be venerated by soldiers on the frontier.66 As god of forests, agriculture, and guardian of the border between nature and culture Silvanus appears as the dedicatee on some seven altars at Alburnus Maior; few are commissioned by Dalmatians, although the kastellum Ansi appears to set up an altar to Silvanus collectively.67 The two altars for the Nymphs, female daimones of nature, of water, trees, mountains etc., allow for a variety of interpretations.68

The invocation of deities of border and nature may well express an inherent unease about the ‘frontier’ position of this unique community, distant from the settlements of legionary camps or coloniae and municipia created elsewhere in Dacia. Perhaps the high number of altars set up by soldiers and Dalmatians to Iupiter Optimus Maximus, the god of the state to whom the emperor was likened, and the altar to Iupiter Depulsor, the god who kept the barbarians at bay, might have occurred in response to these anxieties and expressed the dependency on the emperor and the imperial administration—not only for the provision of security, but for the legal framework, and, more importantly, the right to mine.69 The propensity for what appears to be ‘Roman’ gods by the Dalmatians has also been read as a sign for their self-imagination as ‘Roman’ in a wild and unknown environment (Dészpa 2012, p. 32).

1.5. Polyonymy

The shared experience in a frontier setting and in a unique socioeconomic environment fostered a sense of communality further affirmed by the regular and lasting contacts between members of initially distinct groups. The wax tablets are not the only indication of an intricate web of business and social interactions at Alburnus Maior. The phenomenon of polyonymy may also provide a refraction of a community emerging from diverse groups. A number of individuals with Illyrian names are also known under a different name, and thus have multiple names: Dasas Loni qui et [-]; Dasius Sta(–) (qui et ?) Durius; Epicadus Plarentis qui et Mico; Panes Epicadi qui et Suttius; Planius Baezi qui et Magister; Titus Beusantis qui et Bradua; and Verso Dasantis qui (et) Veidavius.70 In most cases all names, that is, name, patronym, and agnomen are Illyrian, except with Titus Beusantis qui et Bradua, whose name Titus is Latin (Piso 2004, p. 280, ns. 59, 60.). In the case of Planius Baezi qui et Magister, the byname ‘Magister’ seems to refer to a profession or office he (or an ancestor) held within the community.71

The use of agnomina, bynames, by peregrini has been the subject of considerable research, but only recently more has been done in trying to explain why these were in use. In Ptolemaic and Roman Egypt the use of two names, i.e. a Greek and an Egyptian name, suggests the confrontation of distinct ethnolinguistic groups, a dichotomy the navigation of which made having a name for each language advantageous. On rare occasions, that is, when unambiguous identification was desired or required (for legal reasons or in a funerary context), we learn both the Greek and the Egyptian name of one individual (Broux 2015, pp. 291–93; Coussement 2016, pp. 209–10). Similar phenomena of polyonymy can also be detected in communities in Asia Minor, where the use of a Greek and Latin name is indicated in the epigraphic sources (Van Nijf 2010).

The use of two names also appears in official documents from Alburnus Maior. The names, though, do not indicate the movement of the individuals between two linguistically different groups—both names connected with the phrase ‘qui et’ are Illyrian. Thus, some Illyrian inhabitants of Alburnus Maior seem not to have chosen a second name as a result of their contact with people speaking a distinct language (although they inscribe texts in Latin), but may have received this ‘nickname’ due to cohabitation with men and women of a similar linguistic background. One might postulate dialectal differences prompting this, but it is equally possible that, in the confrontation with groups from different parts of Dalmatia settling in this newly established community at Alburnus Maior, individuals with the same name required an alternative name for purposes of distinction. These names derive from the same linguistic stock or express a current or former profession or function within the community (e.g., magister). The need to note both names in legal texts and funerary inscriptions, as with the development of family names in medieval Europe or Greek and Egyptian names in Ptolemaic and Roman Egypt, is likely driven by legal requirements of legal contracts and other documents prevalent in this mining district. These practices are also reflected in inscriptions on funerary monuments where apparently the identification of the deceased must be clear.72

What did these men, women, and children of different socio-economic backgrounds and origins experience when confronted with distinct social practices and habits? Perhaps they recognized commonalities in their origin—most hailed from the same province, Dalmatia—and in their shared fate in a remote part of the Roman empire, facing the hardships and promises of underground mining in an environment perceived as hostile, but also full of promise. Would they have recognised in their displacement any similarities to the people who inhabit the camp of Zaatari in Northern Jordan?

2. Krokodilo, Eastern Egyptian Desert

What is striking about the written evidence from Alburnus Maior is the absence of any Dacians, which raises the question Eutropius’ report suggests, that is, “Dacia, in fact, had been drained of men” (8.6.2). Emperor Julian, too, in his satire The Caesars has Trajan boast that he “removed” or, more likely, “destroyed” the Dacians, named Getae (327 D).73 Similarly, in the scholia to Icaromenippos of Lucian of Samosata, the tale of the complete destruction of the Getae bar forty men, based on the Getica by Statilius Crito, is presented as fact.74 The two Dacian wars of Trajan certainly took a heavy toll on the people of Dacia, but perhaps not to the extent later sources appear to suggest: Lactantius claims a census took place on conclusion of conquest, which would suggest the survival of a significant remainder of the population.75

Men and women with Dacian names are almost absent from the epigraphic evidence, that is, they are not noted on inscribed funerary epitaphs, votive altars, and honorary statue bases found throughout the new province of Dacia Traiana. Only on some six inscribed monuments from Dacia and in military diplomas pertaining to auxiliary units garrisoned there have Dacian names been discovered.76 Moreover, soldiers with the indication of origin (natione Dacus) are enrolled with the equites singulares Augusti, the ‘horse guard’ of the emperor at Rome, arriving from units based in Dacia (Dana and Zăgreanu 2013, p. 156). This scarcity of Dacians attested in the province post-conquest (mostly in military contexts) cannot be taken as corroborative evidence for the total annihilation of the Dacians. What is more, the archaeological evidence appears to suggest the deliberate destruction or clearance of central settlements, hill-forts, and so-called ‘tower houses’ in AD 106, leaving behind a landscape of villages.77 It seems that the local elites were forcefully removed, disrupting the continuation of established social hierarchies; even so, Dacians of lower status appear to have remained (Strobel 2010, pp. 292–93).

Alongside these destructive measures, there is good evidence for the (forced) recruitment of Dacians into Roman auxiliary units, as is attested in the military diploma issued on the honorary dismissal from service. Typical Roman practice in the immediate aftermath of the conquest was to remove young men from subjected communities by drafting them into auxiliary units, as was the case in Asturia and Cantabria after the end of hostilities around 15 BCE.78 Similarly, Cassius Dio relates that after the subjection of Raetia in 15 BCE the potential of revolt by the large population of males was reduced by deporting the strongest men of military age, seemingly to be drafted into auxiliary units.79 The impact of (forced) relocation on those recruited, of leaving a family and community ravaged by war behind can only be imagined. Our sources remain silent on the issue.

This certainly applied to Dacia as well where male youths of communities which had no choice but to submit to the Roman conquerors found themselves recruited into the Roman army.80 We know of an ala I Ulpia Dacorum, a cohors I Ulpia Dacorum as well as cohors II and a cohors III Dacorum which may initially have been formed out of Dacian tribesmen who had sided with Rome already during the Dacian Wars.81 After 106 CE Dacians were recruited into existing auxiliary units distant from their place of origin and sent to their garrison, on occasion together with their family members. One military diploma from 31 July 131 CE, handed to an auxiliary soldier in the province Mauretania Caesariensis on his honorary dismissal (honesta missio) from twenty-five years of military service, confers Roman citizenship to a Diurdanus, son of Damanaeus, and his children and conubium (the right to marry) to his wife, all with Dacian names. One of his sons even carried the rather conspicuous name Decebalus, the name of the Dacian king who fought against Rome (Eck and Pangerl 2005; Strobel 2010, p. 294 with n. 7).

2.1. Dacians in the Desert

A significant number of Dacian names appear on ostraca, i.e. official and private letters, accounts, and receipts written with ink on pottery sherds, and on inscribed monuments found in the Eastern Egyptian Desert. This extraordinary assemblage of written evidence allows for unique insights not only into the life of these frontier garrisons but, more importantly, the life of Dacian soldiers truly displaced on the order of the emperor. They were stationed temporarily at the forts of Mons Claudianus and Kaine Latomia, guarding the imperial quarries there; along the desert roads out to the quarries; and the road to the harbours of Myos Hormos and Berenike on the Red Sea at forts like Krokodilo, Maximianon, or Didymoi, for instance.82

The Dacians here seem to have belonged mostly to cavalry alae or infantry cohorts with a cavalry wing (cohortes equitatae), and thus were experienced horsemen. Given their recruitment into equestrian units, Karl Strobel assumed that these young men had belonged to the aristocratic elite of Dacian tribes, siding with Rome or subjecting themselves to Roman rule in time in 106 CE (Strobel 2010, p. 294). According to the lists of military personnel in which our Dacians are mentioned, they did not serve in ethnically homogenous units but together with men bearing Thracian, Roman, or Greek names. Within these lists, though, the Dacians are mostly grouped together.83 This reflects established Roman administrative practice; the Roman military administration listed recruits from distant provinces and their ethnic backgrounds with the aim of preventing the creation of auxiliary units or legions based in Egypt consisting solely of men recruited locally.84

Ostraca are not the only genre of texts in which Dacians emerge. A rockface near the military fort of Krokodilo / el-Muwayh was inscribed with the following text (in Figure 1):

¹Dida Damanai filius nationis Daqus |² eques alae Uocontiorum turma Maturii |³ armatum(!) feci stationi (!) me(n)ses quinque |⁴ pro salute{m} imperatore (!) feliciter85

I, Dida, son of Damanaus, born in Dacia, cavalryman in the wing of the Vocontii, squadron of Maturus, stood five months under arms on post. Long live the emperor! Good luck!

The fact that Dida, son of Damanaus, eques of the ala Vocontiorum, calls himself Daqus or ‘Dacian’ in the inscription is remarkable. The commonality of being identified as ‘Dacian’ by the Roman army may have fostered a stronger sense of shared origin and fate, which might have overlaid previous, more divergent self-definitions based on specific tribal groups or communities within the tribal confederacy the Romans identified as ‘Dacian’.86 ‘Dacus’ or ‘Dax’ was used by others in the Eastern Desert as a distinguishing term as well: a prostitute girl (κοράσιον), on rejecting the advances of another man, declares on an ostracon (pot sherd) that she is in love with ‘the Dacian’ (φιλεῖ τὸν Δάκα), even imploring her ‘pimp’ to give her to ‘the Dacian’.87 Two further unpublished ostraca in Greek, one from Krokodilo, the other from Kaine Latomia / Umm Balad document the use of ‘Dacus’ as a descriptive term for individual soldiers.88

In this respect, one ostracon stands out: in early 109 CE, a private letter from Dekinais at fort Persou reaches Kaigiza/Kaikeisa at the fort Krokodilo:

Δεκιναις Καικεισα τῷ ἀδελφῷ χ(αίρειν). | ἀσπάζου Ζουτουλα καὶ Πουριδουρ. | ἐρωτῶ σε, Καικισα, σκύλητει | πρὸς ἐμὲ ἐπὶ χρίαν σου ἔχω· |⁵ ἐρωτῶ σε, ἔρχου ὡς πρὸς ἐμέ· | ἐγὼ ἤκουσα ὅτι πάντες οἱ Δάκες | ὑπάγουσιν μετὰ τοῦ ἡγεμόνος | ἰς Ἀλεξάνδ(ρειαν)· ἐὰν εἰδῇς ὅτι ὑπάγου|σιν ἰς Ἀλεξάνδ(ρειαν), γράψον ἰς Κόπτον |¹⁰ ἵνα ταχὺ ἀναβῇ. | ἔρρωσο.

“Dekinais to Kaikisa, his brother, greetings. Greet Zoutoula and Pouridour. I beg you, Kaikisa, move yourself and come because I have need of you. I beg you come and join me. I heard saying that all the Dacians are going with the prefect of Egypt to Alexandria. If you learned that they are going to Alexandria (with certainty), write to Koptos so that he(?) may hurry to go up(?). Farewell.”89

Both men have Dacian names, as do two further soldiers at Krokodilo, Zoutoula and Poridour, whom Dekinais extends his regards to. Dekinais asks Kaigiza to join him and writes that he has heard that all Dacians are required to join the prefect (governor) of Egypt in Alexandria.90 The letter throws light not only on the network of Dacian soldiers in the Eastern Desert of Egypt, but also the fact that they communicated with each other in written Greek.91 Given the date of this and other letters, some of the Dacian soldiers and cavalry men must have acquired Latin and Greek language and writing skills soon after they were enrolled in the auxiliary units in Egypt.92 The letter also reveals the relevance of ethnic categories to the prefect of Egypt in this instance, and in the administration of the army in general: the rumoured call for Dacians to assemble in Alexandria on the orders of the governor might be connected with formation of an ala Ulpia Dacorum, which was to serve in Trajan’s Parthian War (Strobel 2010, p. 294).

Being registered as ‘Daci’ and grouped together in lists by the authorities; identified as ‘Dacian’ by fellow soldiers and civilians; describing oneself as ‘Dacus’ and maintaining contact with other Dacians serving in the Eastern Desert—all this could be read as indicative of a close-knit, perhaps even isolated, group within this fluid ‘desert society’, made up of merchants and caravans, soldiers and officers, imperial slaves and officials, quarrymen and prostitutes travelling through or residing temporarily at forts or quarries in the desert. This notion is somewhat misleading as the ostraca certainly attest men with Dacian names in correspondence with soldiers or civilians who do not share their place of origin: a Diurpanus (a typical Dacian name), for instance, is greeted in a letter written in Latin by Numosis sent to a Claudius (Dana 2003, p. 183). A Dida is noted in letters (concerning debt) sent from the fort at Persou by an Ischyras to Zosimos and Parabolos at Krokodilo.93

2.2. Shared Experience

As the lists of soldiers indicate, the Dacians served together with soldiers of other ethnic backgrounds and performed the same duties as everyone else, for instance, as couriers for official messages between forts, such as Kaigiza, Dida, or Auizina.94 One long text from Krokodilo, the ‘amphorae of barbarians’, attests the death of a Damanias, horseman of the cohors II Ituraeorum, based at the desert fort of Patkoua, who lost his life during an attack on the fort by sixty barbarians in 118 CE.95 It is perhaps this shared risk and fate which underlies the identification of soldiers with their respective unit, apart perhaps from the daily military routine. The graffito left by Dida, son of Damanaus, at Krokodilo not only sees him referring to his military function as cavalryman (eques) and his unit, the ala Vocontiorum, but also to the squadron, the turma, and its respective commander, a certain Maturus. First and foremost, the description as eques or hippeus is repeated throughout in the journals and letters found at Krokodilo and it is noted in reports referring to the deaths of soldiers, their secondment, or assignment to a task.96 Both the cavalry ala and, even more so, its sub-division, the turma, were highly important in identifying soldiers—so much so that they become a quintessential part in self-identification, as expressed in votive inscriptions written on rock faces or on funerary epitaphs.97

The power of daily military routine, of common culinary, hygienic, and dress habits, of shared combat and religious experience, moulded those of divergent backgrounds into something resembling a unit—and distinguished them from the civilian travellers, merchants, camel drivers, prostitutes, quarrymen, and Bedouin around them (Haynes 2013, pp. 165–88). The recovery of faunal remains, for instance, at military sites such as Mons Claudianus, Krokodilo, Maximianon, and other places in the Eastern Desert do reveal a typically military diet.98 A further feature is the existence of baths at remote outposts such as Maximianon or Mons Claudianus, highlighting a practice not exclusive to the Roman army but predominantly pursued by military personnel who probably also shaved on a regular basis—something not necessarily done in the Eastern Desert.99 The Latin inscription left by the cavalry soldier Dida, son of Damanaus, at Krokodilo is just one expression of integration into the army and his unit. Equally the employment of Greek in correspondence even by Dacian horsemen writing amongst themselves is indicative of a swift integration in the army and the embrace of their local practices.

There is no epigraphic evidence that Dacian soldiers kept their distance or were excluded from cult rituals and religious practices of their respective units; if so, the participation in rituals strengthened identification with the unit they served in and reaffirmed their loyalty towards the emperor. Dida’s graffito at Krokodilo in celebration of being relieved from his station at the desert fort, is pro salutem imperatoris, for the well-being of the emperor (Haynes 2013, p. 216).

2.3. Decebalus!

Despite their full integration into the auxiliary units during and after the conquest of Dacia, the retention and popularity of the name Decebalus is striking; it is the name of the last Dacian king who had led the Dacian tribal confederation in the wars against Rome and committed suicide to escape Roman captivity (see Dio 68.14.3).100 Numerous soldiers attested in the second century CE ostraca from the Eastern Egyptian Desert carry the name Decebalus as their name or patronym, and the name remains popular with soldiers based in other provinces throughout the second and third centuries CE (Dana 2014, p. 117).

The name of Decebalus appears to be a deliberate choice not without significance: the military diploma noted earlier (31 July 131 CE), issued to an auxiliary soldier serving in Mauretania Caesariensis, notes the name Decebalus for one of his sons.101 Evidently, after being recruited into the Roman army in 106 CE the soldier awarded one of his sons the name of the unfortunate king. Whereas Dacian slaves (presumably) could receive the ‘historical’ name Decebalus from their Roman masters in commemoration and, perhaps, perpetuation of Roman victory over a hostile enemy,102 the deliberate choice of the name by Dacian soldiers or their parents must call for a different explanation. The allusion to the Dacian king was surely intended, but whether it was just another way of evoking ‘Dacian’ commonality, or pride in once being formidable ‘barbarian’ opponents of Rome and now highly skilled horsemen or foot soldiers in the service of the emperor, or perhaps even a form of ‘resistance’ against Roman subjugation (perhaps like the choice of Dandara for the new community at Belo Horizonte)—the range of possible explanations remains wide open.103

3. Conclusions

The evidence from what I have termed ‘frontier’ societies highlights the hardships not only of the process of displacement itself, but also in the formation of new communities, and the discord that may accompany it: some Dacians might have found it hard to integrate into their unit and deserted, or internal strife may have erupted amongst Dalmatian miners at Alburnus Maior; one can imagine miners absconding from their work, returning home, or coming to blows with Roman officials or with the Dacian natives remaining in the province. Our sources, however, are silent on these issues.

What our evidence on the Dalmatian miners at Alburnus Maior and the Dacian cavalrymen in the Eastern Egyptian Desert does illustrate is the mid- to long-term impact of forced relocation on individuals and communities as a consequence of war and conquest in Dacia. In the case of Alburnus Maior, the evidence suggests that the shared experience in mining gold and its auxiliary ‘industries’ impacted on the formation of a sense of community. Such group formation processes are observable in early modern and modern mining frontiers: in the mid-1500s CE, indigenous people from various parts of Mexico were drawn to Zacatecas to work in the silver mines owned by the Spanish, forming new communities across ethnic divisions; and, during the gold rush of 1849 CE, miners in California, despite different backgrounds, found conformity in dress and expressed a ‘powerful sense of identification as a group’, further shaped through the toils of mining.104 These and other examples are highly suggestive that similar processes were unfolding at Alburnus Maior. What is more, the remoteness of this mining community in the Apuseni mountains bred anxieties about the wilderness and the real or perceived threat of attacks from beyond the reach of Roman control, both of which helped overcome initial ‘tribal’ distinctions (still present in the toponymy of the district in later decades).

A similar process may apply to the Dacians displaced to the Eastern Desert, although the Roman authorities were inclined to emphasize ethnic difference for military purposes. The practice of naming their offspring Decebalus might just be an expression of the warrior pride the Roman authorities did little to suppress. Sharing in the fate of their fellow soldiers fostered the formation of a further public persona of these Dacians—that of the Roman cavalry soldier. The stark confrontation with a new and alien environment further reinforced the sense of a shared experience and outlook. It also exposed the necessity of dependence, by Dacians and other soldiers, on the Roman state and the emperor. They depended on them for almost everything, from clothing, kit, horses, and food, to pay, promotions, and legal privileges. It is this same dependency on the Roman state for protection, security, and for the legal framework, which at Alburnus Maior may have been expressed in the altars set up to ‘state gods’ like Iupiter Optimus Maximus or Iupiter Depulsor. Revisiting the epigraphic evidence in light of the themes and approaches raised by the catalyst papers offers enticing prospects. Although the available data only offers us snapshots of the well-established ‘frontier’ communities at Alburnus Maior and in the desert between the Nile and the Red Sea, it does provide a useful basis to explore the dynamics of how people who have been displaced retained ‘old’ identities, whilst responding to contexts that demand a reconsideration of belonging.