Hostname: page-component-74656f9884-skzmx Total loading time: 0 Render date: 2026-11-06T01:35:57.053Z Has data issue: false hasContentIssue false
  • English
  • Français

Heu male tum Libyæ solis erratur in agris: the Punic tribunate at the Bir ed-Dreder necropolis

Published online by Cambridge University Press:  12 November 2025

Robert Martin Kerr
Robert Martin Kerr*
Affiliation:
Inârah Insititut, Universität des Saarlandes, Germany
*
Email: kerrrm@gmx.com
Rights & Permissions [Opens in a new window]

Abstract

In this paper the Latinate gentilicia Flavius and Iulius, as well as the rank tribunus with its Punic equivalent, found in the Latino-Punic sub-corpus from the necropolis at Bir ed-Dreder are discussed. The texts date roughly to the mid fourth century AD, and attest to the continued survival of Punic in the Tripolitanian pre-desert, also in an official Roman context. While the inscriptions are difficult to understand, direct Latin influence is limited to these three nouns related to their service in the Roman army. The Roman military rank tribunus could, however, also be rendered in Punic. By all accounts, knowledge of Latin was still at best limited in this region during early Late Antiquity.

يا للأسف، يتجول المرء بائساً في أرض ليبيا القفر التريبون البونيقي في بئر دريدر

تناقش هذه الورقة أسماء فلافيوس ويوليوس اللاتينيين، بالإضافة إلى مرتبة التربيونس مع نظيراتها البونيقية الموجودة في المجموعة الفرعية اللاتينية-البونيقية من مقبرة بئر دريدر. يعود تاريخ النصوص تقريباً إلى منتصف القرن الرابع الميلادي، وتشهد على استمرار وجود البونيقية في منطقة ما قبل الصحراء بإقليم طرابلس، و في سياق روماني رسمي. في حين يصعب فهم النقوش، فإن التأثير اللاتيني المباشر يقتصر على هذه الأسماء الثلاثة و المتعلق بخدمتهم في الجيش الروماني. ومع ذلك، يمكن أيضاً ترجمة رتبة التربيونس العسكري الروماني إلى البونيقية. على كل حال، كانت معرفة اللاتينية لا تزال محدودة في أحسن الأحوال في هذه المنطقة خلال أوائل العصور القديمة المتأخرة.

Information

Type
Articles
Information
Libyan Studies , Volume 56 , November 2025 , pp. 44 - 48
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of The British Institute for Libyan & Northern African Studies.

Among the corpus of Latino-Punic inscriptions,Footnote 1 the largest sub-corpus is constituted by the epitaphs from the Bir-ed-Dreder (bi’r duraydir) necropolis.Footnote 2 Here, some 50 stelae were found by Goodchild in Reference Goodchild1950,Footnote 3 of which some 21 are inscribed, though only 11 are legible enough to be more or less understood.Footnote 4 Bir ed-Dreder itself is located in the Sofeggin Basin (wādī sawfağğīn), at the edge of the pre-desert and al-Ḥammāda al-Ḥamrāʾ (‘the Red Rock Desert’), some 60 miles South of Tripoli. The inscriptions were written in Latino-Punic, i.e. Punic using Latin graphemes,Footnote 5 and hence like the rest of the corpus, with full contemporary vocalization, something atypical for northwest Semitic texts.

The stelae, as Goodchild noted (op. cit.), were made of local limestone. The texts are written in, as wonted in pre-desert Latino-Punic, crude lapidary capitals in scriptio continua, incised between guidelines; palaeographical dating is consequently unfeasible. Partly because the texts are rather damaged, and owing to the fact that our understanding of Latino-Punic is still vestigial, they are not yet understood in their entirety. They are interesting because they evidently shed light on Punic speakers, by all accounts local levies, in Roman military service during late Antiquity/early Late Antiquity, as evidenced by the frequent attestation of the Roman gentilicia Flavius, Iulius and the military rank tribunus. The dating of these texts has hitherto been quite approximate: IRT ad loc. gives fourth–early fifth centuries AD,Footnote 6 based on the content, i.e. largely the aforementioned nouns of Latin origin. There is no evidence for Christianity at the site: the burials are not inhumations but cremations. In the following we hope to shed some additional light on the date of these texts, some of the Punic vocabulary, the role of those commemorated and the use of Punic in (late) antique Tripolitania.

Generally speaking, by the end of the second century AD, the current scientific communis opinio is that the wadis to the east of the Gebel were no longer protected by regular Roman troops, and that the security of the territory was predominantly contingent on local resources. Goodchild identified those buried at Bir ed-Dreder as having been leaders of fœderati, entrusted by Rome with the defence of the region, and considered that the title tribunus in casu was chiefly honorary.Footnote 7 This understanding of their mission and of the term tribunus nonetheless remains debated. R. Rebuffat, for example, believed that, while these tribuni belonged to the Roman army proper (i.e. the comitatenses), they as traditional leaders commanded their own men.Footnote 8 A similar view is shared by Mattingly, who argued that the cemetery at Bir ed-Dreder was that of a local tribal group, some of whom as gentiles, were associated by treaty with Rome to defend the frontier.Footnote 9

While the exact details still largely escape us, primarily due to the paucity of particulars, several points should be noted. Of the 50 stelae discovered, only 21 are inscribed; this might be construed to imply that there was a social hierarchy, with only the elites being commemorated by epitaphs. There might also be some discrepancy in the length of texts, though this is difficult to determine due to their poor state of preservation. Based on what still can be read, the following pertinent data can be gleaned from the texts with regard to one or more of the three terms already mentioned.

Bir ed-Dreder: analysis of the pertinent passages

Here and in the following small sans serif capitals are used to render Latino-Punic in philological contexts, otherwise they are rendered in cursive. Words and letters between [square brackets] refer here to graphemic renderings, those between /strokes/ refer to reconstructed phonetic transcriptions.

LP 1: 1flavi saich2 am bn ma3 carcvm 4σonmo5 n tribynvs …

LP 2: 1 ivlivs ma2 sthalvl 3 byn chyr4 did ryb m5 sir …

LP 3: 1[fla]bivs mac[a]rcv2[m .. σ]obm[o]n [.] trib3[vnus b]yn [.]r[.]nia … 8macarcvm tribv9nvs [σ]ob[mon] …

LP 4: 1maσicama 2byn isachv 3tribvnvs str

LP 5: 1ivlivs [..] 2ibitva- [..] … 5[..b]yn chy6[rdid] ryb 7[msir] …

LP 6: 1ivlivs nasif 2tribvnvs by(n) 3siraban by(n) 4iyσihan rira5chan …

LP 10: … 3?tri[bv]nvs …

LP 13: 1flabiv2s masin3[th]an tr[i]4[b]vnv[s] …

LP 14: 1flabivs isi2gvari trybv3nvs byn iarnv4han …

LP 17: 1[..fl]ạb2i byn va3sira …

Philological remarks

The orthography is not standardized, as is customary in Latino-Punic. Flavius is rendered either as in LP 3, 13 and 14 in the Latin nominative case, or as in LP 1 and 17 in the vocative, as is customary for Latin loans in Punic;Footnote 10 the spelling with [b] instead of [v] is a phonetic spelling.Footnote 11 Iulius and tribunus are always rendered in this sub-corpus as per the Latin nominative. The variant spellings of the former – LP 1 tribynus, LP 14 trybvnvs – show vowel reduction of Punic unaccented syllables using [y]=/ə/ or /∅/.Footnote 12 This vowel reduction is also found in the regentes of status constructi LP 2, 3, 4, 5, 6, 14 and 17 with byn ‘son of’ (i.e. *bin > /ṷən-/) and LP 2 and 5 ryb ‘commander of’ (i.e. *rabb > /rəṷ-/). LP 1, bn is analogous through syncopation (/ṷᵊn-/). The ligature s (i.e. [t] + [s]) represents /ṣ/,Footnote 13 the grapheme here rendered with [σ] renders /z/.Footnote 14 The use of the vocative, noting vowel reduction by [y], and the use of [s] and [σ] shows this sub-corpus adheres to the linguistic and orthographic conventions of Tripolitanian Latino-Punic.

Syntactically speaking the inscriptions are rather stereotypical. In LP 6, 13 and 14 we find the Roman gentilicium followed by the diacritic name,Footnote 15 invariably Libyco-Berber in this sub-corpus,Footnote 16 the Roman rank tribunus, followed (LP 6 and 14) by the patronymic affiliations (‘son of …’); in LP 6 the grandfather is also mentioned. In LP 1, tribunus followed by σobmon precedes the patronym, but would seem to refer to the son (Flavius Saicham), which would also seem to be the case in LP 3.Footnote 17 In LP 3 and 5 with the gentilicium Iulius, following the diacritic name, we find a Punic designation of military rank ryb msir (see below).

The Punic tribunal Flaviate

With regard to the gentilicium Flavius, we see that it becomes increasingly common in the fourth century, after Constantine the Great’s victory over Licinius in 324, in particular among barbarians to whom citizenship rights had been granted upon entering the army in some capacity.Footnote 18 This practice continued the former practice of using the gentilicium Valerius during the preceding Tetrarchy.Footnote 19 Even after the end of the Constantinian dynasty with Julian’s death in 361, the practice continued until at least the ninth century, cf. e.g. Leo VI (reg. 886–912: Cæsar Flavius Leo Augustus), that is by subsequent emperors who could claim no descent from Constantine, it having become (like Cæsar after the end of the Julian dynasty in AD 68) an element of imperial nomenclature. In his study of the bestowal of the name Flavius in Egypt, Keenan noted that there, the families of most who took on the gentilicium Flavius had become citizens when Caracalla (i.e. Marcus Aurelius Antoninus) issued the Constitutio Antoniniana, usually dated to 212.Footnote 20 At that time, the Egyptians who had become Roman citizens generally took on the imperial gentilic Aurelius, though indigenous examples are still attested until the second half of the fourth century, after which time they become increasingly uncommon. KeenanFootnote 21 notes a distinction in social class between the Egyptian Flavii and Aurelii, the former being largely military and civil servants of the government, the latter being civilians sensu stricto: craftsmen, merchants, labourers, farmers, etc.Footnote 22 There was also a socio-economic distinction (loc. cit.): landowners, lessors and creditors were invariably Flavii, whilst tenants, lessees and debtors were Aurelii.Footnote 23

That we are dealing with Roman military personnel, in all likelihood local levies charged with frontier defence, is apparent from the frequent attestation of the Latin loan tribunus.Footnote 24 As Keenan’s analysis of the Egyptian evidence shows: ‘by far the greatest number of Flavii are found among the soldiers of later Roman Egypt’Footnote 25 – ‘both limitanei and those belonging to “the static units of comitatenses”, were levied locally and spent their service in or near their home towns.’Footnote 26 After 328,Footnote 27 attestations become increasingly copious, from simple recruits to officers; in some cases where prosopographical information is available, it is clear that Egyptian Aurelii upon joining the Roman Army nearly always replaced this gentilic with Flavius, the latter term thence serving as a ‘status designation, setting these soldiers and functionaries apart from the masses of the population who continued to retain the name Aurelius’.Footnote 28 Of final relevance here is Keenan’s observation that while among elite administrative and landowning Flavii, the gentilic was seemingly hereditable, this was not the case ‘among low-grade military families’.Footnote 29

While the ample Egyptian evidence clearly shows the military proclivity of the gentilicium Flavius, the name being given e.g. when enlisting, we are clearly dealing with men whose families were already Roman citizens. It is unclear whether the Bir ed-Dreder border troops were already Romans or rather became novi cives when they were enlisted as part of the local Roman land forces. Judging by the fact that the forefathers mentioned in the Dreder texts do not have Roman gentilicia, the former scenario would seem more probable. A more exact parallel, though less well-attested than Egypt, might be that of Germanic tribesmen who during the fourth and fifth centuries were Romanized on being recruited, the gentilicium Flavius being an asseveration of Roman citizenship and military status (somewhat similar to ‘legionnaire’).Footnote 30 One might compare the Dreder necropolis with the Germanic funerary inscriptions from Concordia Sagittaria (Iulia Concordia, in the Regio X Venetia et Histria) where we also find the gentilicium Flavius followed by Germanic diacritic namesFootnote 31 and a military rank, e.g. Fl(avius) Marcaridus/tribunus mil/itum (CIL 5, 8753).Footnote 32

Our preliminary findings would seem to indicate a terminus post quem with Constantine the Great’s victory over Licinius in 324 when the gentilicium Flavius came into vogue, especially after the mid-fourth century when barbarians were increasingly integrated into the Roman army, being especially implicated in frontier defence. A more precise indication might be the use of another Latin name as a gentilicium in these texts, namely Iulius, which in Tripolitanian epigraphy is only attested here preceding an indigenous African diacritic name.Footnote 33 As Borhy has posited, the use of both Flavius and Iulius side by side in this Latino-Punic sub-corpus, can be connected to the joint reign of Flavius Julius Constantius (II) and Flavius Julius Constans (I) from 337 to 350.Footnote 34 This finding comports well with the use of tribunus in these texts, since in the late fourth and fifth centuries, the title tribune was the most common senior officer rank and was used variably for any commanding officer below the highest level of command, sometimes in addition to the title præpositus (though probably not in TripolitaniaFootnote 35), and was also awarded to gentiles, local non-Roman nobles or notables who protected the imperial borders as commanders of allied ethnic groups (fœderati) on behalf of the Romans.Footnote 36

Whatever the precise nature of the indigenous military unit in question, judging by the site of Bir ed-Dreder on the southernmost edge of the Tripolitanian pre-desert, they seem to have been a unit of limitanei charged with frontier defence. They were given Roman imperial gentilicia, probably upon enlistment in the Roman land army, to indicate their new civic and military status, the native commanders also being given the title (or were later promoted to) tribunus. The fact that these gentilicia were mostly rendered in Punic as Latin nominatives (and not in the vocative as is customary in Punic) might be an indication that these were seen as titles (cf. tribunus) and not as names. If Flavius Macarcum in LP 3 is the father of Flavius Saicham in LP 1, and Chyrdid is the father of both Iulius Masthalul in LP 2 and Iulius … in LP 5 then we would seem to be dealing with a clan-like structure that had served in this capacity for at least two generations. The Dreder cemetery, as well as the Sitz im Leben of pre-desert Latino-Punic epigraphy in general, comports well with the local frontier situation from the mid-fourth into the fifth centuries. As Nicasie has pointed out, the Libyan limes remained largely stable and defensive in nature from the first to sixth centuries, marked by scattered forts, towers and fortified farms that protected farmland and caravan routes. Serious threats were rare, so troop numbers stayed small. In Tripolitania, the fourth century saw decline: garrisons shrank, frontier defence shifted to fortified farmsteads and the army mainly secured roads and caravan routes; archaeology and the Notitia Dignitatum suggest few regular troops but numerous units of limites supervised by praepositi.Footnote 37

Punic elements

a) ryb msir

Having discussed the Latin military loans, and having mentioned the Libyco-Berber onomasticon of the Dreder Latino-Punic texts, let us now turn to some Punic particulars. In LP 1 and 5 (supra) we find the imperial gentilic Iulius, but without the addition of the term tribunus. This is mentioned with Iulius Nasif in LP 6. The former two, though, both have the genitive noun phrase ryb msir /rəṷ mᵊṣīr/ (<*raḇ maṣṣīr <*rabb mansīr).

For the regens or ‘head’ ryb from the root √rbb, cf. Hebrew ‐רב /raḇ-/ ‘chief of’ or ‘head of something or someone'. This usage is frequent in biblical Hebrew, especially in military contexts, e.g. רב־טבחים /raḇ-ṭabbāḥīm/ ‘chief of the guard' (AV),Footnote 38 רב החבל /raḇ-haḥōḇēl/ ‘shipmaster’ (AV; lit. ‘chief of the rope-men’);Footnote 39 and is similarly used in Modern Hebrew רב־אלוף /raḇ-’alūf/ ‘lieutenant general’ (lit. ‘chief of the major generals’), רב-טוראי /raḇ-ṭūr’ay/ ‘corporal’ (lit. ‘chief of the privates’), רב־סרן /raḇ-séren/ ‘major’ (lit. ‘chief of the captains’).Footnote 40

As for the second element, the rectum of the bound form, the root is √nṣr,Footnote 41 cf. biblical Hebrew נצרים /nōṣrīm/ ‘watchmen’ (AV).Footnote 42 Especially relevant here is Akkadian rabi maṣṣārti ‘chief of the guards’; ša maṣṣārti lit. ‘the one of the guard’, i.e. ‘guard, watchman’; maṣṣaru ‘guardian, watchman, watch, garrison’; e.g. in the letter of Rib-Addi of Byblos EA 122:13 ina ūmī abbūttija lú.meš ma-ṣa-ar šarri ittišunu ‘in the days of my forefathers they had a royal garrison’.Footnote 43

The phrase ryb msir might then be best rendered in English as ‘commander of the guard’. The Punic would then be a dynamic equivalent rendition of the Latin noun tribunus.Footnote 44 It is worth noting that in two (LP 2 and 5) of the three (with LP 6) attestations of the gentilicium Iulius in this sub-corpus the Punic term is employed whilst with Flavius the Latin loan tribunus is always used. The use of the Punic term here instead of the Latin might have been a personal preference since, in both attestations, sons of (probably the same) Chyrdid are being commemorated.

b) σonmon

In LP 1 we have a term σonmon (/zonmōn/) following the patronym Macarcum and preceding tribunus which cannot easily be explained as a personal name. In LP 3 we find once σobmon apparently following Macarcum and preceding tribunus, and once coming after Macarcum tribunus. As was noted above, Macarcum the father of Flavius Saicham in LP 1 would seem to be the Flavius Macarcum also attested in LP 3. The lexeme σon/bmon has hitherto had no convincing interpretation. The orthography with [b] would seem to be due to de-nasalization, which would suggest a root √znm with a nominal suffix /-ōn/ (< /-ān/). Such a root is rather rare in Semitic, but is nonetheless attested in Arabic. The verbal root is only attested in the second form, possibly indicating that it is denominal, for example Lane (1867, 1260): zannama: [1] ‘the cutting a small portion of the ear of a camel or other animal, and leaving it hanging thereto […]’; [2] ‘The act of marking with a sign or token […]’; [3] ‘And the conjoining any one with a people, or party, to which he does not belong’. From [1] we find a noun zanamaḧ indicating the tragus or a wattle (i.e. something joined to something else, but considered redundant). We find a nominal formation which, as is often the case, has multifarious meanings, namely zanīm: [1] ‘One adopted among a people to whom he does not belong […] not being needed […] likewise one who is adopted among a people or by a person’;Footnote 45 [2] ‘conjoined’; [3] ‘Base, ignoble, or mean; known by his baseness, ignobleness, or meanness, […] or his evil character’; [4] ‘The son of an adulteress or a fornicatress’; and [5] ‘A commissioned agent, a factor, or a deputy’.Footnote 46 Further information is provided by Ibn Fāris (1972, 1432), who gives the base meaning of the root as ‘indicating the attachment of two things to one another. From this is derived zanīm which is a “supplicant”’ (aṣlun yaduluw ʿalā taʿlīqi šay’in bi-šay’in. min ḏ;alika l-zawnīmu, wa-huwa d-dawʿiyuw); cf. Lane’s verbal meaning [3].

Lane’s Verbal meaning [3], together with the basic meaning given by Ibn Fāris, i.e. ‘being joined together’ (possibly with Lane’s nominal meaning [5]) bears some resemblance to the etymology of Latin tribunus: tribus ‘tribe’ (people joined together) + the suffix -nus ‘chieftain, commander, tribune’. Although a historical dictionary of Arabic is still a desideratum, and our knowledge of Phoenico-Punic vocabulary is but embryonic, with all due caution we proffer this interpretation: a nominal derivation of the root √znm, /*zVnVm/ + /-ōn/ > /zVnVmōn/ >/zŏnmon/Footnote 47 – σonmon, probably a Punic neologism. Since this word occurs together with tribunus in all three attestations, it would seem to be a Punic designation for the (hereditary?) autochthonous office held by Flavius Macarcum and his (eldest?) son Flavius Saicham, either ‘chieftain’, or ‘commander’.

Final thoughts

In the preceding we have discussed the Latino-Punic sub-corpus from the Bir ed-Dreder necropolis at the Southern edge of the Tripolitanian pre-desert. The comprehensible inscriptions refer to the deceased in general by their imperial gentilicium, their diacritic name and in varying sequence their patronymic and military title, either using the Latin loan tribunus or Punic ryb mṣir. In the case of one seeming lineage, a further Punic title σonmon (‘chieftain, commander’?) is also added. Those commemorated by these stelae by all accounts served as limitanei on the limes Tripolitanus. Based on the gentilicia Flavius and Iulius, these native recruits would seem to have (originally) enlisted during the co-reign of Flavius Julius Constantius (II) and Flavius Julius Constans (I) from 337 to 350.

While these texts are clearly written in Punic, they employ the Latin writing system and Latin technical terms (gentilicia and ranks) and the onomasticon is Libyco-Berber. While the concept of ‘identity politics’ has become increasingly common in studies of the ancient world, the available evidence makes it difficult to ascribe a distinct ‘Punic identity’ to Punic speakers in Late Antique Roman North Africa. It would seem that even in the fifth century in the interior regions there were many coloni who spoke only Punic and no, or hardly any, Latin.Footnote 48 Even in the Roman army, as the Bu Njem ostraca attest, seemingly written largely by local recruits, the knowledge of Latin in the mid-third century was rudimentary at best (as was the case elsewhere in the empire).Footnote 49 Punic survived because there was no overt exigency for the indigenous population not aspiring to a career in the Roman administration to learn Latin, especially at the lower echelons of society and in the interior regions where Romanization was often quite limited. This says nothing about (ethnic) ‘self-identification’ – as noted, the presence of Libyco-Berber onomasticon (or any other) is not an indicator of ethnicity. Even if the Bir ed-Dreder tribuni could be identified as Libyco-Berber, Punic, the colonial language prior to Latin, seems to have continued to be the primary regional lingua franca.Footnote 50 The use of the Latin alphabet to write Punic seems to be explained by the pre-desert frontier defence context of much of the Latino-Punic corpus – the Bir ed-Dreder limitanei were recruited Punic speakers, who must necessarily have had a basic knowledge of Latin, including bare-bones literacy (cf. again Bu Njem). The use of the Latin writing system had probably two primary reasons: first, the itch to write influenced by the Roman ‘epigraphic habit’, i.e. with the Roman presence, public epigraphy diffused to non-Latinate cultures such as Punic (and Libyco-Berber); second the absence of other writing systems at this time. The Roman-era neo- and Latino-Punic corpora attest to the first phenomenon: the use of Punic for genres and of calqued expressions attests both to the influence of Latinate writing culture and the survival of Punic. As for the second, the latest datable neo-Punic texts, i.e. written with the Punic writing system (literacy in which seems even in Carthaginian times to have been limited), are from the end of the first century AD. While it is clear that Punic did not die then,Footnote 51 it was no longer used in public epigraphy in the Romanizing city of Lepcis Magna, remaining then a largely spoken language. That is, the Latin alphabet was employed since this was probably the only writing system which the author(s) of the Latino-Punic corpus had any active proficiency in, hence a case of diachronic digraphia (and not synchronic bigraphism).

The Bir ed-Dreder Latino-Punic sub-corpus dating to approximately the mid-fourth century AD, attests to the survival of Punic in the Tripolitanian pre-desert amongst limitanei in Roman service. Influenced by Latinate ‘epigraphomania’, some of those buried at this necropolis were commemorated with Punic epitaphs. The Latin elements are limited to the gentilicia Flavius and Iulius, which were probably viewed as titles or status indicators, and precede the diacritic name as well as the Roman rank tribunus. The former could be rendered by a Punic term, apparently a neologism, indicating that the language was still very much vital. That the inscriptions were written in Punic, either the mother tongue of this group or their primary contact language with their Roman superiors, was probably due to the fact that this group’s ability to write and read (intended audience) Latin was exiguous. Being implicated in some fashion in the Roman army, they unavoidably had at least limited Latinate literacy which in Tripolitania was more or less systematically adapted to write Punic, Punic script having heretofore fallen into disuse.

Abbreviations

CAD

= Oppenheim, A. L. and Reiner, E. (eds) 1977. The Chicago Assyrian Dictionary, Volume 10 M part 1. The Oriental Institute, Chicago.

ECDS

= Epigraphik-Datenbank Clauss / Slaby. Available at: http://www.manfredclauss.de/ (accessed on 25 September 2024).

IRT

= Reynolds, J. M. and Ward Perkins, J. B. et al. 1952. The Inscriptions of Roman Tripolitania. British School at Rome, Rome and London. Updated version available at: https://irt2021.inslib.kcl.ac.uk/ (accessed on 25 September 2024).

Footnotes

1. See Kerr Reference Kerr2010, for Bir ed-Dreder, 169–83, also for complete bibliographical information. Since this publication, one further such text has been published: see Elmayer Reference Elmayer2019, 147–49.

2. See Barker et al. Reference Barker, Gilbertson, Jones and Mattingly1996, 70f., site Dd1, for further information, including a map of the spot, and a reconstruction of one of the mausolea (fig. 5.36).

3. Goodchild Reference Goodchild1950, Reference Goodchild1954.

4. Kerr Reference Kerr2010, Bir ed-Dreder 1–21=IRT 886a–x.

5. It must be noted that Latino-Punic is not a linguistic designation. The language employed is Punic, the script employed is Latinate.

6. So too the ECDS (13900455): ‘Datierung 301 bis 430’.

7. Goodchild Reference Goodchild1976, 59–71. See also the assessment of the evidence in Roques Reference Roques1987, 265–68.

8. Rebuffat 1976, 412–14.

9. Mattingly Reference Mattingly1995, 195–97. See in general now Campbell Reference Campbell, Cancick and Schneider2005, 601–602, and Rocco Reference Rocco2024, 153–57. On a by all accounts similar phenomenon in Arzugitana (Southern Tunisia) at the end of the fourth century AD, see St Augustine, Epistola 46 by Publicola: ‘[…]decurioni qui limiti praeest vel tribuno solent iurare barbari, iurantes per daemones suos, qui ad deducendas bastagas pacti fuerint […]’.

10. Kerr Reference Kerr2010, 44f.

11. Op. cit., 108–10.

12. Op. cit., 86–100. Presumably in Punic, the accent was on the final syllable, i.e. /ṭrəṷənū́s/.

13. Op. cit., 126–30.

14. Op. cit., 130–37.

15. I.e. a person’s individuating name versus adopted, foreign gentilicium. A distinctive feature of the binominal system was that, unlike the single-name system, a person’s diacritic (the name marking individual identity) was not necessarily their most important name. While the prænomen could serve as a household identifier, its limited diversity made it insufficient for distinguishing individuals in broader contexts; see in detail Salway Reference Salway1994, 126.

16. Based on the late Punic corpus (neo-Punic and Latino-Punic) in its entirety, it is impossible to derive ethnicity based on the origins of names, as Punic, Libyco-Berber and Latin names are used discretionarily in what by all accounts was a multi-ethnic melting pot.

17. We exclude here LP 17 as not enough remains legible, though one would expect a rank to follow. Similarly omitted is LP 10 in which only the rank tribunus is legible.

18. See Mommsen Reference Mommsen1910, 476ff., also now Le Bohec Reference Le Bohec2007. While limitanei played an important role in frontier defence during the reign of Diocletian, it was probably not until his successor Constantine I that the distinction between them and the comitantenses was formalized (see Nicasie Reference Nicasie1998, 18).

19. Which in turn is similar, though not identical, to the practice of peregrini, or other groups of persons, on being granted citizenship by an emperor adopting the imperial gentilicium, e.g. Flavius Josephus/Ἰώσηπος Ματθίου παῖς, Bel. Iud. A3. This of course continued the practice of slaves at their manumission adopting their former masters’ gentilicium upon becoming clientes.

20. Keenan Reference Keenan1973, 39–42.

21. Art. cit., 1973, 34, 52.

22. This is clear in Keenan, art. cit. 55, where those ναῦται at Syene who also served as soldiers were designated as Flavii ‘in virtue of their military rank’, while those boatmen who did not remained Aurelii.

23. Keenan Reference Keenan1974, 286.

24. Bir ed-Dreder LP 1, 3, 4, 6, 10, 13, 14.

25. Keenan Reference Keenan1973, 61

26. Op. cit., 34, citing Jones Reference Jones1964, 668f.

27. Art. cit., 49.

28. Art. cit., 49.

29. Keenan Reference Keenan1974, 296.

30. Cf. Harmoy-Durofil Reference Harmoy-Durofil2016, 101f.: ‘On peut constater un usage plus fréquent du gentilice Flavius, pour la période comprise entre le milieu du ive siècle et le milieu du ve siècle, période pendant laquelle, les barbares entrent en grand nombre dans la militia armata, parce que celle-ci a besoin de reconstituer des unités mais aussi des états-majors décimés par les incursions barbares, notamment après le désastre d’Andrinople en 378. Cette période correspond aussi à la politique menée par les empereurs Valentiniens puis Théodosiens qui ont confiance en leurs officiers barbares, qui accèdent […] davantage et en plus grand nombre aux postes les plus prestigieux et au consulat.’

31. See the lists provided by Harmoy-Durofil Reference Harmoy-Durofil2016, 394f.; Bratož Reference Bratož, Dörfler, Gleirscher, Ladstätter and Pucker2017, 237(–40); and Hoffmann Reference Hoffmann1963. For a similar phenomenon in Late Antique Asia Minor, see Destephen Reference Destephen2019.

32. Dated by Harmoy-Durofil (Reference Harmoy-Durofil2016, 460) tentatively to the reign of Constantius II (337–361).

33. One might mention here the Latino-Punic text from Wadi Uaeni (Kerr Reference Kerr2010, 220f.): 1 iv flav 2 sanu av lxvi ‘Iu(lius) Flav(ius) years liv(ed) 66’. Note the Punic loan translation from Latin sanu<th> av(o) < vixit annos.

34. Borhy Reference Borhy1989, 154.

35. Cf. the Notitia dignitarum xxi ‘Dux provinciæ Tripolitanæ’, which in its listing ‘sub dispositione viri spectabilis ducis provinciæ Tripolitanæ’ then lists twelve præpositi limitis.

36. Franke Reference Franke, Cancick and Schneider2009, 904.

37. Nicasie Reference Nicasie1998,140–41.

38. IReg 25:8, 10, 11, 12, 15, 18, 20; Jer 39:9; Aramaic Dan 2:14.

39. Jon 1:6.

40. We give here only examples from Hebrew, Punic’s closest cognate, but such usage is well attested in the Semitic languages, e.g. Arabic رب العالمين /rabb al-‘alamīn/, traditionally rendered ‘Lord of the worlds’ (Qur’ān 1:2), رب الدار /rabb ad-dār/ ‘chief of the house’, رب بحري /rabb baḥrī/ ‘seaman’ (naval rank, lit. ‘chief of the sea’). In Aramaic, e.g. רב חילא /raḇ ḥaylā/ ‘general’ (lit. ‘chief of the armed force’); רב אספקלטוריא /raḇ ʾespuqlāṭriyyā/ (<speculator via Greek) ‘chief executioner’.

41. Note the cognate roots √nṭr in Aramaic and √nẓr in Arabic.

42. Jer 31:6; 2 Reg 17:9, 18:8 מגדל נוצרים /miḡdal nōṣrīm/ ‘tower of the watchmen’ (AV).

43. CAD M/1: 333–44.

44. In Roman-era Late Punic, we find both formal and dynamic equivalent renderings of Latin mots et choses; see in general Kerr Reference Kerr2017.

45. Lane mentions here the lexeme Qur’ān 68:13, though some modern translations render it with meaning [4]. The latter meaning, someone ‘cut off from their lineage’, is possibly influenced by zanā ‘to fornicate’ (zāniyaḧ ‘adulteress; fornicatress’; zīna’ ‘adultery, fornication’).

46. Lane refers here to az-Zabīdī’s Tāğ al-ʿArūs, which I am currently unable to consult.

47. V indicates a vowel of unknown quality and quantity. The first [o], by all accounts short and unaccented, might in all likelihood be due to vowel harmony (Kerr Reference Kerr2010, 76–82).

48. Cf. the discussion of the evidence in Kerr Reference Kerr2010, 13–24. Note the testimony of St Augustine in the early fifth century, especially relating to the incident at the castellum Fussalensis (op. cit., 18–20).

49. Cf. Kerr Reference Kerr2010, 123–25; Adams Reference Adams1994; Reference Adams1999; Reference Adams2003, 236. Here the distinction must clearly be made between indigenous limitanei and regular levies.

50. See in detail Kerr Reference Kerr2010: 13–24.

51. Cf., for example, St Augustine (354–430: Punic was certainly his mother’s tongue and possibly his mother tongue) who noted in 391 that Punic was not dead and that Punic books existed. (Ep. 17.2): ‘Quæ lingua si improbatur abs te, nega Punicis libris, ut a viris doctissimis proditur, multa sapienter esse mandata memoriæ. Poeniteat te certe ibi natum, ubi huius linguæ cunabula recalent.’ We do not know how literate Punic speakers were, though this was presumably quite low, or how readily available ‘Punic books’ were.

References

Adams, J.N. 1994. Latin and Punic in contact? The case of the Bu Njem ostraca. Journal of Roman Studies 84: 87–112.CrossRefGoogle Scholar
Adams, J.N. 1999. The poets of Bu Njem. Language, culture and the centurionate. Journal of Roman Studies 89: 109–34.CrossRefGoogle Scholar
Adams, J.N. 2003. Bilingualism and the Latin Language. Cambridge University Press, Cambridge.CrossRefGoogle Scholar
Barker, G., Gilbertson, D. D., Jones, B. and Mattingly, D. 1996. Farming the Desert: The UNESCO Libyan Valleys Archaeological Survey, 2. Society for Libyan Studies, London.Google Scholar
Borhy, L. 1989. The rank indicator role of the names Flavius and Iulius for prefects on the Tripolitanus Limes (remarks on Late Roman frontier defence In Tripolitania). Acta Archaeologica Academiae Scientiarum Hungarica 41: 151–57.Google Scholar
Bratož, R. 2017. Zur Präsenz und Mobilität ethnischer Kleingruppen im Alpen-Adria-Raum während der Ostgotenherrschaft. In: Dörfler, I., Gleirscher, P., Ladstätter, S. and Pucker, I. (eds), Ad Amussim: Festschrift zum 65. Geburtstag von Franz Glaser. Klagenfurt am Wörthersee, Landesmuseum Kärnten; Österreichisches Archäologisches Institut, Vienna: 215–48.Google Scholar
Campbell, J.B. 2005. Limitanei. In: Cancick, H. and Schneider, H. (eds), Brill’s New Pauly. Encyclopaedia of the Ancient World, 7 K-Lyc. Brill, Leyden: 601–602.Google Scholar
Destephen, S. 2019. Christianisation and local names in Asia Minor: fall and rise in Late Antiquity. Proceedings of the British Academy 222: 258–76.CrossRefGoogle Scholar
Elmayer, A.F. 2019. Two Punic inscriptions from Roman Tripolitania. Libyan Studies 50: 147–52.CrossRefGoogle Scholar
Ibn Fāris = Aḥmad Ibn Fāris al-Qazwīnī, ʻAbd as-Salām Muḥammad Hārūn (ed.), 1972. Mu‘ğam maqāyīs al-luġa. Muṣtafá al-Bābī al-Ḥalabī, [Cairo].Google Scholar
Franke, Th. 2009. Tribunus cohorhtis. In: Cancick, H. and Schneider, H. (eds), Brill’s New Pauly. Encyclopaedia of the Ancient World, 14 Sym-Tub. Brill, Leyden: 903–904.Google Scholar
Goodchild, R.G. 1950. The limes Tripolitanus II. Journal of Roman Studies 40: 30–38.CrossRefGoogle Scholar
Goodchild, R.G. 1954. La necropoli Romano-Libica di Bir ed-Dreder. Quaderni di archeologia della Libya 3: 91–107.Google Scholar
Goodchild, R.G. 1976. The Romano-Libyan cemetery at Bir ed Dreder. In: idem, Libyan Studies. Society for Libyan Studies, London: 59–71.Google Scholar
Harmoy-Durofil, H. 2016. Chefs et officiers barbares dans la militia armata (IVE–VIe siècle). PhD dissertation, Université François-Rabelais, Tours.Google Scholar
Hoffmann, D. 1963. Die spätrömischen Soldatengrabschriften von Concordia. Museum Helveticum 20: 22–57.Google Scholar
Jones, A. H. M. 1964. The Later Roman Empire. A social, economic, and administrative survey, 2, 284–602. Basil Blackwell, Oxford.Google Scholar
Keenan, J.G. 1973. The names Flavius and Aurelius as status designations in Later Roman Egypt. Zeitschrift für Papyrologie und Epigraphik 11: 33–63.Google Scholar
Keenan, J.G. 1974. The names Flavius and Aurelius as status designations in Later Roman Egypt. Zeitschrift für Papyrologie und Epigraphik 13: 283–304.Google Scholar
Kerr, R.M. 2010. Latino-Punic Epigraphy. A Descriptive Study of the Inscriptions. Mohr Siebeck, Tübingen.CrossRefGoogle Scholar
Kerr, R.M. 2017. Præsenti tibi maturos largimur honores. Zum punischen Prinzip des kaiserlichen Opferkultes zu Lepcis Magna. Kleine Untersuchungen zur Sprache des Alten Testaments und seiner Umwelt 22: 53–74.Google Scholar
Lane, E.W. 1967. An Arabic–English Lexicon Book 1, Part 3. Williams and Norgate, London.Google Scholar
Le Bohec, Y. 2007. ‘Limitanei’ et ‘comitatenses’. Critique de la thèse attribuée à Theodor Mommsen. Latomus 66: 659–72.Google Scholar
Mattingly, D.J. 1995. Tripolitania. Routledge, London.Google Scholar
Mommsen, Th. 1910. Ostgothische Studien. In: idem: Gesammelte Schriften Vol. 6, Historische Schirften. 3. Weidmann’sche Buchhandlung, Berlin: 476–82.Google Scholar
Nicasie, M. 1998. Twilight of Empire. The Roman Army from the Reign of Diocletian until the Battle of Adrianople. Brill, Leyden.CrossRefGoogle Scholar
Rebuffat, R. 1977. Une zone militaire et sa vie économique : le limes de Tripolitaine. In: Chastagnol, A., Nicolet, C. and van Effenterre, H. (eds), Armées et fiscalité dans le monde antique. Éditions du CNRS, Paris: 412–14.Google Scholar
Rocco, M. 2024. Limes. Vivere e combattere ai confini di Roma. Salerno Editrice, Rome.Google Scholar
Roques, D. 1987. Synésios de Cyrène et la Cyrénaïque du Bas-Empire. Éditions du CNRS, Paris.Google Scholar
Salway, B. 1994. What’s in a name? A survey of Roman onomastic practice from c. 700 B.C. to A.D. 700. The Journal of Roman Studies 84: 124–45.CrossRefGoogle Scholar