David Noy
Open University
Migration was an integral part of many lives in the ancient world, whether it was voluntary or forced. In terms of activities which have left tangible evidence, it affected everything from the naming of a baby to the choice of a burial method. Its effects on household formation and family relationships have received little explicit attention in discussions of the Roman family, even though the population of the city of Rome always contained a substantial proportion of migrants.1 Most migrants who moved from one province or state to another formed small minorities within their new communities, but this chapter will focus on two areas where they may have been in the majority: Delos under Roman influence at the end of the second century BCE, and the military population of the city of Rome around the third century CE. These are two groups whose enthusiastic adoption of the epigraphic habit means that they are unusually well documented.2
The source material used here does not provide direct information about living arrangements: someone who commemorated an uncle did not necessarily live with him.3 It will therefore be necessary to think about social relationships rather than patterns of co-residency. The people discussed would in any case have had unusual households: traders with bases on Delos sometimes had business interests in other locations and perhaps residences in all of them; soldiers serving at Rome must have lived officially in their barracks even if they had family members elsewhere in Rome.
Migration disrupted whatever living arrangements were normal at home. For example, an individual who might at home have remained in his original household after marriage and eventually formed part of an extended family could, if he married after migration, form a simple family household purely because he had no other relatives with him. Alternatively, migration might throw together people who would not have co-resided at home, for example brothers or cousins who would normally have formed their own simple family households, but combined in an extended family at their destination. Modern evidence suggests that this is most likely to happen with poor migrants (Glick, Bean, and Van Hook 1997), unlike the groups studied in this chapter. The presence of relatives and others from the same home community at the destination may have been one of the factors which encouraged further migration (the process of “chain migration”), although it seems that nineteenth-century migration in England, the United States and northern Italy did not substantially increase the formation of extended family households other than by women arriving to marry into existing households (Ruggles 1987: 54, 220; Kertzer and Hogan 1990: 491–3; Boyd 1989: 639–41). These are issues which cannot be addressed directly, but some of the inscriptions may be suggestive.
The preponderance of migrants from the Eastern Mediterranean in the population of Delos in the late second and early first centuries BCE, as attested by epigraphic sources, is well known. The original inhabitants were removed after 167 BCE and a remarkably cosmopolitan society developed until the island was sacked in 88 BCE and the massacre and subsequent economic decline brought it to a rapid end. Relatively little attention has been paid to the circumstances of individual free migrants. Did they come alone or in family groups? If they formed families on Delos, did they do so with others of the same geographical origin? Did they intend to settle on Delos permanently or to return home at some point? Were extended family connections important in bringing new migrants to Delos? The focus here will be on people from Asia Minor and Syria,4 since the circumstances of those who came to Delos from mainland Greece, particularly Athens, were rather different, and it was much easier for them to move between the island and their home.
There are clearly no direct answers to these questions. All discussion has to be based on the inscriptions which have survived: primarily building inscriptions, honorific texts in the form of religious dedications, lists of ephebes, and epitaphs. Identifying family connections beyond the nuclear is a haphazard activity, since relationships are rarely spelled out. Epitaphs from the burial island of Rheneia do not normally name the commemorator, although the practice of erecting statues on Delos itself to relatives and friends provides some compensation for this. Most epitaphs, which were inscribed on relatively portable altars and stelae, have not been preserved in situ, so family connections which the original layout of the tomb would have clarified are now lost (Couilloud 1974: 258).
The use of the χρηστὲ χαῖρε formula and of local styles of stele or altar indicate that standard Delian practices were preferred by migrants to ancestral ones. Greek was the common language, widely used by Italians for public purposes (Adams 2003: 645–9). It is likely that many people from the East, in particular, acquired the epigraphic habit through living on Delos, and would not have left inscriptions if they had remained at home, although there is one example of an ex-voto by a Tyrian in Phoenician, with a Greek date, from the mid-second century BCE (Baslez and Briquel-Chatonnet 1990). It should be remembered that not everyone who died on Delos necessarily intended to move there permanently. Pharnakes and Myron from Amisos were killed when shipwrecked on Seriphos, and commemorated on Rheneia by their friend Protos (Couilloud 1974, no.475). These are the sort of people for whom Delos may only have been a temporary residence. However, Protos had a statue erected there by his four sons (ID 1984), which suggests a much stronger bond; he was probably a permanent resident if his friends were not.
Around 100 BCE, a statue was erected with this inscription (ID 1723):
[Φι]λ&c.dotbl;όστ&c.dotbl;[ρατ]oν Φιλo[στράτoυ] Ἀσκαλωνίτην τραπεζιτεύoντα ἐν [Δήλωι], Διόδoτoς Ἀντιπά[τ]ρ[oυ] Ἀσκαλωνίτης τὸν ἑαυτoῦ θεῖoν καὶ τρoϕέα [κ]αὶ σωτῆρα καὶ εὐεργέτην Διὶ Kυνθίωι καὶ Ἀθηνᾶι Kυνθίαι.
Philostratos son of Philostratos, Ascalonite, banker on Delos. Diodotos son of Antipatros, Ascalonite, (honored) his uncle and guardian and savior and benefactor. To Zeus Kythnios and Athena Kythnia.
Philostratos is a well-known figure who is mentioned in 18 inscriptions including a eulogy by the poet Antipater of Sidon; see Santamaria (1982) and Leiwo (1989) for full discussion. He was active on Delos c.140/130–90 BCE, and acquired Neapolitan citizenship towards the end of that period, although there is no reason to think he had any direct connection with Naples. He provided various benefactions, particularly a portico in the Agora of the Italians. He had a wife and children (ID 1719–22, 2628). His nephew Diodotos is known only from this inscription. The use of the term τρoϕεύς suggests that Philostratos had played a part in Diodotos’ upbringing, and there is no sign of the presence of Antipatros the father on Delos, so Diodotos had probably been with his uncle for most of his life. However, the language of “savior and benefactor” was normally reserved there for honoring rulers, or benefactors of the Athenian people. Diodotos was apparently part of an extended family group, but he was not quite in the position of an adopted son or he would not have used such fulsome language.
How exceptional was this family? The best evidence for family migration to Delos is the presence of numerous youths explicitly labeled as foreigners in lists of ephebes or dedicatory inscriptions from gymnasia, namely males in their mid/late teens taking part in educational and ritual activities. According to the calculations of Baslez (2002: 56), in the late second century BCE 14 percent of ephebes were Roman/Italian and 34 percent oriental. In one list of 91 names, dated 119/18, over half were from eastern cities (Rauh 1993, 46; ID 2598). Participation in such an institution suggests that they were there in family groups, since anyone of this age-group who had migrated independently would presumably need to spend his time working rather than living an ephebic lifestyle. It indicates that Delos was the family’s primary, if not sole, residence, and that they expected to remain there. Shared involvement in the ephebate would create links between people which went beyond the familial and communal, but it is significant that ephebes’ ethnics were normally recorded. It was an institution for promoting shared values, but evidently not for eradicating the past.
Diodotos may have been an ephebe himself: an Ascalonite whose name is lost but was the son of [Antipa]tros appears in a list from between 105 and 102 BCE (ID 2599). In the case of an ephebe such as this, there are several possible backgrounds:
The last possibility is the one of most relevance here, and is the most plausible for Diodotos. Evidence for the importance of migration of children at Rome has recently been produced from isotopic analysis of teeth from the Isola Sacra cemetery (Prowse et al. 2007). The authors suggest that single males in early adulthood may not have predominated as much in migration as is usually supposed (e.g., Noy 2000: 66), and that could perhaps be extrapolated to other places such as Delos. Migration of children would normally be expected to involve nuclear families (possibility 1 above), but the case of Diodotos suggests that other scenarios are possible.
Marriage patterns among migrants to Delos are important for the understanding of family relationships. It is unusual for a husband and wife recorded together to be given the same ethnic, for example Hermione and Demetrios, both from Antioch (SEG 31.731). This is probably because an ethnic applied to the one named first could be assumed to apply to the other, for example Euporia of Antioch, wife of Eunikos (Couilloud 1974: no.61). It is also possible, however, that the second person was simply regarded as having no ethnic affiliation. There are some cases where husband and wife have different ethnics, suggesting marriages which were made on Delos (possibility 4 above):
Marriages between free migrants of different places of origin are almost never attested at Rome (Noy 2000: 74), perhaps because the much greater size of different migrant groups made endogamy easier. The offspring of such marriages would have been in the very unusual position (for the ancient world) of having relatives in two other, completely different places. The effects of this cannot be traced, but it may have been of significance when the economy of Delos collapsed after the sack of 88 BCE and survivors of the massacre (which no doubt included oriental residents as well as Italians5) emigrated. If such marriages were made on Delos, there must have been marriageable females among the migrants there. The commercial opportunities which attracted people were very largely limited to men. There is unlikely to have been a demand for free female labor in domestic service or textile production when slave labor was so plentiful, so independent female migration would not have been common.
There are a number of epitaphs for female migrants identified by a patronymic rather than a husband’s name, for example Nikaia of Ptolemais (probably the one in Phoenicia), daughter of Nikanor (Couilloud 1974: no.32). The stele shows a seated woman and a young man, suggesting that she was commemorated by her son. Presumably they were on Delos as permanent migrants. In context the tombs may have shown other family relationships; joint epitaphs for husband and wife, or naming one in the epitaph of the other, were not normal Delian practices. Whether or not Nikaia was a wife and mother, the fact that she was commemorated as a daughter suggests that she was born on Delos or came there with her father.
In 115/14 BCE, statues were erected anonymously to the Sidonian sister and brother Isidotē and Dionysios, children of Dionysios (ID 2091), again suggesting a woman whose most important connection was with her natal family, whether or not she was married. A more perplexing case commemorates in verse Alinē of Ascalon, who was pregnant when she died (Couilloud 1974: no.468). She is described as μόνην πλανῆτιν δημότιν λελει < μ > μένη, which Couilloud translates as “une solitaire itinérante, une femme du peuple abandonnée” (“a solitary wanderer, a woman of the people abandoned”). She would hardly have received a stele and eight-line verse epitaph if she had really been abandoned. Could she have been left on Delos while her husband was engaged in business elsewhere, or had gone back to Ascalon? Or was she commemorated by other Ascalonites? The description suggests a new arrival, perhaps one who had come to Delos to join her existing husband (cf. possibility 2 above), or in order to get married.
There are a number of examples of large family groups, like Protos and his four sons discussed above. A tomb on Rheneia commemorates the five sons of Apollodoros of Anthedon, probably the city in Palestine rather than Boeotia (Couilloud 1974: no.160) – perhaps the members who died young of a nuclear family which migrated to Delos or was formed there. In 128/7, Achaios son of Apollonios of Hieropolis built a temple to his ancestral gods Adatos and Atargatis on behalf of himself, his wife Euboula, his children Apollonios, Dionysia, Protogeneia, Achaios and Lysimachos, and his brothers Apollonios, Lysimachos and Protogenos (ID 2226). The nuclear family may have migrated from Hieropolis or been formed on Delos, or they could have arrived as an extended family with Achaios’ brothers too. However, the brothers were not necessarily all present on Delos, and in families with extensive trading interests it would be natural for some members to settle on Delos while others remained in their home city or managed the business in other trading centres. Absence is suggested more strongly in a dedication to Heracles and Haurona, gods of Iamneia, by Zenorodoros, Patron and Diodotos, Iamnites, on behalf of themselves, unnamed brothers, relatives and co-citizens who participated in cult banquets (ID 2308; Baslez 1988: 140).
In a rare example of epitaphs being preserved in situ on Rheneia, an excavation of two linked funerary enclosures in 1975–7 shows the cooperation of an extended family of Tyrian origin (Le Dinahet-Couilloud 1997). Athenais, daughter of Dies of Tyre, built a family monument. Two members of the family, Dies and Heliodoros, sons of Dies, were ephebes in 104/3 BCE as Tyrians, and later acquired Athenian citizenship with their father, who was probably Athenais’ brother (Le Dinahet-Couilloud 1997: 636–7). One member of the family evidently lived in Athens, as Athenion, envoy of Mithridates, was welcomed there in 88 BCE in the house of Dies, a wealthy man with revenue from Delos (Athenaeus, V 49 Kaibel). A stele has an eight-line verse epitaph for a girl with Roman citizenship, the daughter of Artemisia (perhaps Athenais’ sister) and Q. Furius (Le Dinahet-Couilloud 1997: 648–51). Several other people in the complex had Roman or Italian names, including Ammia daughter of Sosis (Le Dinahet-Couilloud 1997: 652), probably to be identified with Sosis wife of Ariston son of Archelaos of Tyre (ID 2130). There was also one Antiochene. Le Dinahet-Couilloud believes there must be links between the families in the tomb, even if they cannot all be restored, and she writes (1997: 665–6, my translation): “These documents also illustrate the complexity of Delian society; business relationships, certainly, but also matrimonial alliances united the notables of the most entrepreneurial communities of Delos at the end of the Hellenistic era.”
A group of Delian stelae formerly at Shanganagh Castle in Ireland was acquired by Captain Rowan Hamilton in the 1820s (Nicolson 1943: 200–1). It seems likely that he took them from one place on Rheneia, and they were perhaps from an enclosure like that of Athenais. Five commemorate Ascalonites, including four members of the same family: Dionysios, his two sons and one grandson (Purser 1925: nos 6, 7, 10, 71, 72).
Artemisia and Philē may have belonged to the same family as the others, or the tomb may have been shared by unrelated Ascalonites.
An extended family of Velians emerges in a semi-circular exedra in the Portico of Antigonos (ID 1965). Agathokles son of Hermon honored his brother Thrasydeios, who was also honored by his own children Hermon and […]ana. Thrasydeios reciprocated to his brother, with Hermon who was probably Agathokles’ son. Elsewhere, Hermon son of Thrasydeios was honored by his wife Theodotē daughter of Dioskourides (who was not said to be a Velian) (ID 2368). There was clearly cooperation between uncles and nephews/nieces in this family; perhaps the inscription even represents a household arranged on the frérèche principle with married brothers living together.6
Tréheux (1992: 76) suggests a possible stemma for the family of Iason from Arados, based on an exedra from the South Portico with inscriptions for six statues (IG XI.4.1203). The people named are:
Eukleia of Arados and Iason appear to be the parents of Strato and […]eia who married Sillis of Arados: Sillis’s daughter Timokleia married her uncle Strato. A marriage between uncle and niece (sister’s daughter) would fit a scenario where one member of the family moved to Delos and then sent for another, but the evidence would only reveal such a marriage if the whole family was recorded on Delos, as in this case.
Family relationship is much less clear in a stele depicting two women, one with a young child and a baby, and this epitaph (Couilloud 1974: no.187):
Mυρσίνη Ἑτo[ρ]ηία Δέκμoυ Ῥωμαία ἀδελϕὴ δὲ Koΐντoυ Aὐϕιδίoυ Kασιoδώρoυ χρηστὴ χαῖρε.
Σαβεῖ Πύρρoυ Ἀπάμισσα χρηστὴ χαῖρε.
According to Delian naming conventions (Couilloud 1974: 332) the first woman was Myrsinē the slave of a Roman citizen D. Hetereius, and her brother Kasiodoros was the freedman of a Q. Aufidius. The second woman’s name is the vocative of Sabeis, daughter of Pyrros, of Apamea. Couilloud believes they are sisters-in-law, commemorated by their husband or brother. Did the whole family come from Apamea, or did Sabeis marry someone from a different background to her own? Slavery complicates the picture in this case but family ties could sometimes be preserved despite it.
Life on Delos clearly threw together people of different geographical backgrounds in ways other than marriage. One epitaph on a plain stele lists 22 individuals, including one mother and daughter, with a variety of ethnics from all around the Eastern Mediterranean, such as Joppa, Apamea, Cyrene, Nabatea (Couilloud 1974: no.418). All the deceased were slaves of Protarchos, and were probably killed together in an accident, such as a workshop fire. Another inscription commemorates two people who survived a fire where 30 were killed (an Antiochene woman and Theodoros of Laodicea in Phoenicia were commemorated by Meges the Phoenician), perhaps the same event, as suggested by Couilloud (1974: no.477). However, it is possible, although it was not normal Delian practice, that the 22 slaves pooled their resources to set up a common tomb as was often done by slave and ex-slave households and work-groups at Rome.7
There is one case where a house can be linked with occupation by a household not based on kinship (perhaps to be identified rather as a “houseful”). Rauh (1993: 222–31) discusses the living arrangements of the slaves and freedmen of the Paconii in the House of the Hermes, which “was not a ‘single family dwelling,’ therefore, but the communal residence or ‘club house’ of the Paconian slave sunetheia”; it was not the home of the Paconii patrons although they would presumably use it when visiting the island. The only common feature of the occupants was the fact of belonging to various Paconii. Slaves of Italians did not normally give any other indication of their own origins, so it is not clear (although likely) that they were from the Eastern Mediterranean. The households of Protarchos and the Paconii would both have been places where people of a variety of backgrounds were brought together, few people had close kin with them, and the household provided a family-substitute. The Latin curse tablet of T. Paconius, gives an alternative insight into such relationships (ID 2534; Adams 2003: 680–2). Twenty individuals are cursed, most of whom seem to be freeborn Roman citizens. There are three other Paconii, all with different praenomina, and some people with purely Greek names such as Serapion son of Serapion. The range of people hated by this Paconius was a wide one.
It is likely that shared nationality also offered a substitute for family. However, ethnic affiliation could change, as it did for Philostratos. Pouilloux (1973: 412) reconstructs the family of Timarchos of Salamis (Cyprus), who had one grandson with Tarentine citizenship and one with Athenian, as well as one Salaminian. Were these technicalities which did not interfere with ordinary social relationships, or would they be accompanied by transfer to a different social milieu? Timarchos’ son Simalos, described in different places as Salaminian and Tarentine, was compared to Alkinoos in an elaborate poem commissioned by his Athenian friend Stolos, and it is unclear whether the splendors of his hospitality were offered at Salamis or Delos (ID 1533; Pouilloux 1973: n.61). The Salaminians are also discussed by Baslez (1994). Her slightly different stemma illustrates the mobility of some families. Simalos son of Timarchos was gymnasiarch at Salamis c.107–105 BCE, and was on Delos with Tarentine citizenship by 100 when he was part of an association of Italians (ID 1755). His son Simalos did ephebic activities on Delos in 103/2 and at Athens in 101/0 BCE. Another Timarchos, a distant cousin according to Baslez’s reconstruction, had Athenian citizenship and was an ephebe at Athens in 106 BCE, but was buried on Rheneia (Couilloud 1974: no.210).
Baslez (1988: 146–7) discusses the associations (koina) of orientals formed at Delos, Athens, and Rhodes, and concludes that they were not corporations, professional associations, or exclusive ethnic groupings, but fundamentally different from typical Greek voluntary associations: based on worship of their own gods by people who felt uprooted, but tried to integrate into the Greek milieu through the use of Greek terminology and the Greek concept of euergetism. The worship of ancestral gods probably provided an alternative support network for migrants without family on Delos (Hasenohr 2007: 87). The Jewish and Samaritan synagogues are examples of this, and also offered an extension of the network, since two men honored for their benefactions by the Delian Samaritans came from Crete and were probably not resident on Delos (IJudO Ach66–7). Italians had their own collegia, nominally religious in function, and these generally used Latin in their own inscriptions as part of their “institutionalised collective activity,” while the Competaliasts consisted primarily of slaves and ex-slaves who were of Greek origin and therefore used Greek (Adams 2003: 667–8).
Dedications to ancestral gods were made by, for example, people from Berytos (ID 1783, 1785). The Poseidoniasts of Berytos were clearly more than a religious association, since their official title was “the koinon of the Berytian Poseidoniasts, merchants and shipmasters and receivers,” and the Heracliasts of Tyre were “merchants and shipmasters” (ID 1519). The gods represented Melkart (Heracles) and Baal Berith (Poseidon) (Hasenohr 2007: 84). Baslez and Briquel-Chatonnet (1990: 29) suggest that the sanctuary founded in 154/3 BCE by the Tyrians may have been for visitors to Delos rather than permanent residents at first, and that the first indication of definitive immigration is the ephebe lists from 119/18 BCE. There was an association of merchants and shipmasters from Laodicea in Phoenicia which honored a member of the Seleucid royal family ca.178 BCE (IG XI.4 1114; Baslez and Briquel-Chatonnet 1990: n.22). The Italians could act as a collective too, for example when they honored Philostratos the banker and his sons (ID 1722).
Philostratus was responsible for making dedications to numerous gods, including Astarte, Aphrodite Ourania and Poseidon of Ascalon on behalf of the city of Ascalon and his wife and children. These dedications come from the Syrian Sanctuary, where a variety of religious activities seem to have taken place. The Syrian divinities Hadad and Atargatis were worshipped there. Dedicators to them in the late second and early first centuries included Romans (ID 2255, 2266, 2269), Laodiceans (ID 2259, 2262, 2264, 2297), Antiochenes (ID 2256, 2285; SEG 31.731), and a Hieropolitan (ID 2261). The wording of ID 2255 is a reminder that people with Roman citizenship were not necessarily of Italian ancestry: P. Plotius son of Lucius, Roman, built the exedra for himself and his wife and children, for his grandson C. Plotius son of Gaius, for Heras son of Diokles from Seleucia and his mother, on behalf of the Athenian people and the Roman people, as an offering to Hagnē Aphrodite, and Hadad (90/89 BCE).
People who made such dedications often honored their relatives. Extended family connections can be traced for a number of migrants to Delos. They can be assumed to have existed for many more, since their discovery in the evidence is so fortuitous. The connections may have been instrumental in bringing people to Delos, or may have developed in importance among those who found themselves on Delos for other reasons. For the no doubt numerous migrants who did not have such family, membership of koina which officially worshipped national gods but in fact had much wider functions offered some compensation.
In the third century CE, there may have been 17,000 provincial soldiers stationed in Rome at any one time.8 The Praetorian Guard had a nominal strength of 15,000, recruited principally in the Danubian provinces by this time, and there were up to 2,000 equites singulares, mostly from the Balkans although sometimes referred to as Batavi, and an unknown number of sailors from the fleet (often recruited in Egypt or Thrace).9 There were also soldiers from the legions stationed in Rome in the Castra Peregrinorum for a variety of reasons. Troops served for 16 or 17 years (praetorians) or 25 years (equites). They lived at least officially in barracks away from the general population, and they often tended to return home after their military service although significant numbers remained in Rome or Italy, probably those who had formed families in particular (Durry 1938: 301–2; Coulston 2000: 82–5; Ricci 2005: 88). By the third century there was no longer any legal obstacle to soldiers marrying while still serving.10
Within their own world, these soldiers form a comparable population to the migrants on Delos: a society where migration was the norm and therefore something to be proud of rather than to disguise, where the pull factors attracted adult men directly but perhaps family members indirectly, and where the epigraphic habit was pronounced, in the form of epitaphs and dedications to gods, often ancestral ones.
Soldiers normally began their military service in their late teens, so it is unlikely that many would have been married before they enlisted, given that all studies of Roman age at marriage have accepted that men married much later than women, typically no earlier than their mid-twenties (Scheidel 2007: 421–3). They are unlikely to have come to Rome with parents (although it is possible that parents sometimes joined them later). However, the importance of brothers among soldiers, and among migrants at Rome generally, is clear from epitaphs (Noy 2000: 67–75). Thirteen percent of explicitly immigrant soldiers and 19 percent of male pagan civilian immigrants were commemorated by brothers (sometimes brothers who were soldiers).
T. Aurelius Primus, librarius (military clerk) commemorated his brother aged 16, M. Ulpius Ursinus son of Ulpius Respectus from Aelius Cetius in Noricum (CIL VI 33036). This is likely to be a case where the presence of an older brother at Rome encouraged a younger brother to join him there. The younger brother was probably intending to enlist when he was old enough. The different nomina could indicate that they were half-brothers or that the elder brother only got a Roman name on enlistment, like the Egyptian Apion who joined the fleet at Misenum and acquired the name Antonius Maximus (BGU II 423).
The uncle-nephew relationship whose importance can be glimpsed among migrants on Delos is of more explicit significance among military migrants at Rome. In some cases where an uncle commemorates a nephew or vice versa, both are soldiers or veterans. Some examples are:
None of these inscriptions refers to geographical origin, but all the people involved must have been migrants. M. Aurelius Sabinus evidently had his sister with him in Rome, and there are other examples of this among the praetorians, such as Q. Sittius Iulianus who commemorated his sister Iulia Donata (Noy 2000: 70–1; CIL VI 32730). It is possible that the sisters were there not primarily because of their brothers, but because of their unnamed husbands: if a soldier serving in Rome wanted to marry, a natural choice would be the sister of a fellow-soldier from the same homeland (perhaps the same village).
Other inscriptions suggest women coming to Rome to marry men from the same place already serving there. When the eques singularis Ulpius Ianuarius died at 29 and was commemorated by his fellow-eques Octavius Dignus and his wife Aurelia Novana, he was described as a native of Pannonia Superior from the territory of Savaria and the village of Voleucio (CIL VI 3300); this indicates pride in a very specific place which would have had significance to few people at Rome. It may also suggest that the place was of significance to the commemorators; perhaps Novana had come from there to Rome too. Valeria Palladia commemorated two brothers: her husband Valerius Proclianus aged 44, protector (an honorary military rank), and her cognatus Valerius Nepotianus aged 30, exarc(h)us promotus domnicus (a cavalry officer) (AE 1946, 127 = 1949, 190). Cognatus may be simply mean “brother-in-law” here, but it is also possible that Palladia and her husband’s family were cognati before the marriage.
There are also cases of cousins commemorating each other where both were soldiers:
Two cousins could be exactly the same age and therefore better able to enlist together than two brothers; that may be the case with the Aurelii above.
Sometimes one relative is not stated to have a military connection, which could be because he really did not or because it was not felt necessary to mention it. For example:
In a sample of 46 epitaphs of praetorians (CIL VI 32647-32714, using only those with an identifiable commemorator; each relationship counted once), the following commemorators are found:
| wife | 13 |
| child | 2 |
| parent | 4 (1 soldier) |
| brother | 7 (5 soldiers) |
| freedman | 5 |
| cognatus/consobrinus | 2 (2 soldiers) |
| fellow-soldier/heir/amicus | 13 |
| self/by will | 6 |
Commemorations by nuclear family only slightly outnumber those by extended family and non-kin. The nuclear family was less significant than among civilians at Rome (for whom 70 percent of commemorations were by nuclear family, 27 percent by extended family, 3 percent by non-kin, according to Saller and Shaw 1984: table 4), the extended family rather more important and non-kin vastly more important. Soldiers rarely had parents available to commemorate them; other relatives and more particularly non-relatives took their place.
Soldiers were bound together by the practice of making each other their heirs. No doubt in many cases the choice of heir was influenced by shared geographical or cultural background. An alternative support network was provided by people from the same locality (who may of course have been relatives too). The best evidence for this is provided by groups of Thracians serving in the Praetorian Guard in the third century, who joined together to make dedications, often to local gods, and recorded their home villages in their inscriptions (Noy 2000: 219–20; Ricci 2005: 83–8). For example, a dedication to Deus Sanctus Hero Briganitius by Flavius Proclianus of Cohort III, Claudius Mucianus of Cohort V, Aurelius Valerianus of Cohort V and Valerius Maximus of Cohort X (CIL VI 2807 = 32582). They describe themselves as cives Usdicensis vico Agatapara, that is from one village, perhaps in the territory of Hadrianopolis, which was called Uscudama in Thracian. Two praetorians from different cohorts (Iulius Iustus and Firmius Maternianus) from Veromandus in Belgica made two dedications (one dated 246 CE) to diis sanctis patriensibus (actually Jupiter, Apollo, Mercury, Diana, Hercules, Mars, Nemesis, Sol and Victoria) (CIL VI 32550–1). Coming together for religious activities may have been one aspect of a shared social life, or may have been the factor which kept fellow-villagers together. It transcended loyalty within the military unit, suggesting that the connections existed before coming to Rome.
Migration disrupted and altered families in many ways. Migrants themselves were likely to be predominantly men of marriageable age, and migration would at least have slowed down the formation of their own family. Soldiers moving to Rome would not usually be married before they went there. If they did decide to marry while they were serving there, their wives might well be women from their home areas and/or relatives of their fellow-soldiers. It was more likely that whole families would move to Delos together, but there are plenty of examples of people who arrived there unmarried, in which case they might marry someone from their homeland or someone whom they met (or whose family they met) on Delos, probably someone with whom they would never have come into contact in other circumstances.
Migrants could and did form nuclear families, but there were a number of reasons why other relationships were potentially more important to them than to people who stayed at home. Connections at the destination could encourage the migration in the first place: an elder brother serving in the Praetorian Guard at Rome; an uncle with a business on Delos. Relatives remained potentially useful afterwards: running other parts of an international business based on Delos, or helping a soldier to return home when he retired from the army. People whose relationships might not have been very significant at home were drawn together by the circumstances in which they found themselves: siblings, uncles, cousins and relatives by marriage.
On Delos it was normal for people to advertise their homeland in inscriptions and to pass their ethnic designation to their children; this tendency would no doubt have decreased over time, but the events of 88 BCE cut short any natural development. Soldiers at Rome also recorded where they came from, often down to a specific village. Many of them no doubt intended to return there, but for those who stayed at Rome it was apparently not usual to pass on their ethnic designation. For both groups, nominally religious associations of people from the same homeland played an important role, probably replacing family for those who did not have kin with them, but also surely including many relatives whose kinship is not apparent in the epigraphic evidence.