Onomastics and linguistic contact in Aegean Thrace
p. 39-58
Résumés
The aim of this paper is to analyse language contact phenomena in the stock of indigenous personal names from Aegean Thrace, a region which, according to Dana’s classification of Thracian personal names (OnomThrac, p. LXXVII), belongs to a transitional onomastic area. The material coming from the Greek coastal colonies will be examined from the point of view of linguistic contact between the Greek and Thracian languages. The focus will be on the definition of the mixed character of anthroponyms from a twofold perspective: the description of the Thracian evidence in light of Dana’s categorization and its adaptation to the Greek system of personal names. Particular attention will be paid to the emergence and diffusion of morphological processes of Hellenization of Thracian names.
L’objectif de cet article est d’analyser les phénomènes de contact linguistique dans le stock de l’anthroponymie indigène de la Thrace égéenne, une région qui, selon la classification de noms de personnes thraces établie par Dan Dana (OnomThrac, p. LXXVII), appartient à une aire onomastique de transition. Le matériel provenant des colonies côtières grecques sera examiné du point de vue du contact linguistique entre les langues grecque et thrace. L’accent sera mis sur la définition du caractère mixte des anthroponymes dans une double perspective : la description des témoignages thraces à la lumière de la catégorisation de Dana et leur adaptation au système anthroponymique grec. Une attention particulière sera accordée à l’émergence et à la diffusion des processus morphologiques d’hellénisation des noms thraces.
Texte intégral
1. Linguistic contact and anthroponymy in Aegean Thrace
1In ancient times, the peripheral region known as Aegean Thrace was a zone of intense linguistic and cultural contact, first between Greek colonists and the non‑Greek populations – designated as “Thracians” by the Greeks – and later with Latin speakers.1 Traditionally, the Rhodope Mountains acted as a physical barrier between this area colonized by Greek settlers from the 7th c. BC and Inner Thrace, the main dwelling place of Thracian tribes. According to Bouzek, Graninger 2015, p. 15, “Aegean Thrace was, due to its Greek neighbors, the most civilized part of the country and it became also soon Hellenized”.2
2Linguistic contact must have been a reality in this crossroads region of the Aegean Sea. In the broadest sense, we expect to find evidence not only of language contact between Greeks and the indigenous populations, but also among the different Greek linguistic varieties that arose in the area. As far as the Greek language is concerned, various epichoric Greek dialects must have come into contact from the 7th c. BC as a consequence of the establishment of Greek colonies in this coastal region, most of which were founded by Ionians. From this period, different Ionic varieties emerged in the broader area alongside Attic, which appeared from the 5th c. BC, because of the Athenian political activity in the Northern Aegean. In later times, the Attic-Ionic Koine gradually supplanted epichoric dialects in the region. In addition to the rich Graeco-Thracian cultural exchanges, linguistic contact between Greek and the indigenous languages must have taken place from the Archaic period to Late Antiquity. Various Thracian tribes (Sapaioi, Bistones, Kikones, Korpiloi) settled in Aegean Thrace prior to the Greek colonization, although they flourished between the 4th and the 2nd c. BC, especially when the Odrysian tribes grew in strength north of the Rhodope Mountains.3
3The analysis of Thracian onomastics, and more specifically of Aegean-Thrace names, constitutes one of the most useful devices for the study of cross-linguistic and cultural interactions in the area. Personal names (PNs) are a growing source of knowledge, as they provide the largest amount of Thracian linguistic evidence. We will concentrate on the area between the rivers Nestos and Hebros, including the Samothracian peraea, as it is the region covered by I.Thrake Aeg. and because the adjacent Ionic areas, such as Thasos and its peraea (between the rivers Strymon and Nestos), belong to a different Thracian onomastic unit (see infra § 2.1). The analysis of Greek PNs begins in an early period (7th‑6th c. BC), while the most recent attestation of Thracian names pushes the date forward to the 4th c. AD.
4This research focuses on the classification of the extant Thracian onomastic material from this region based on its Graeco-Thracian hybrid character. On the one hand, we will analyse the corpus of Thracian PNs in accordance with the classification of Thracian anthroponymy into four onomastic areas (see infra § 2.1) and, on the other, we will examine the mechanisms of adaptation of these groups of foreign names into the Greek onomastic system (see infra § 2.3). We will also discuss possible cases of the reverse process, the “Thracianization” of Greek PNs (see infra § 2.4) and list some problematic names in the last section (see infra § 2.5). We seek to determine the type of linguistic contact and accommodation processes which can be detected on this micro-regional level and describe the transitional character of this Thracian onomastic region. However, our approach to language contact is conditioned, on the one hand, by the scarcity of the written evidence and, on the other, by our lack of information on the Thracian language. Consequently, the relevant phenomena of linguistic contact will be examined from the perspective of onomastics with all the necessary methodological precautions. To this effect, we will differentiate between Greek and non‑Greek onomastic bases, suffixes and formations, considering them from both a diachronic and synchronic point of view.
2. Thracian anthroponymy in Aegean Thrace
5Thracian onomastic material is now available thanks to the publication of new corpora of inscriptions from Aegean Thrace: I.Thrake Aeg. (2005), the reference corpus, and Αρχαία Ζώνη I (2015), devoted to the inscriptions found in the sanctuary of Apollo at Zone, a colony founded by Samothracians on the coastline.4 We also have at our disposal up‑to‑date linguistic descriptions of the Greek and Thracian languages and recent research studies on onomastics.5 One of them is the inventory of Thracian anthroponyms in Dana’s Onomasticon Thracicum (OnomThrac, 2014), which has proved crucial to the present survey of North-Aegean Thracian names. Regularly updated online (OnomThracSuppl),6 this volume is essential for supplementing previous catalogues of PNs, such as the LGPN IV (2005). Frequency analysis of Greek and Thracian names has been facilitated by Parissaki’s work (2007), whose approach is more prosopographical and historical than linguistic. Del Barrio Vega 2018, pp. 503‑507, offers a concise summary of the principal linguistic features in this Greek-speaking region, which covers the geographical area between the rivers Strymon and Nestos, including Thasos. Her description on anthroponymy focuses on the most frequent onomastic formations of Greek and Thracian names. Finally, as far as the Greek PNs are concerned, Masson’s studies have also proved decisive, especially his 1984 survey on the Greek names of some magistrates from Abdera and Maroneia (see Masson 1984).
6Turning to the statistical distribution of Aegean-Thrace anthroponyms, Parissaki 2007, p. 267, provides 714 different PNs, distributed in around 1,250 occurrences. Greek names represent the highest proportion (78,5%) in comparison with Pre-Greek/Thracian (7,6%) and Latin names (12,3%).7 The larger amount of Greek evidence matches with the Greek-speaking character of the area, well established even during Roman times, and with the existence of epigraphic documents mainly in Greek. Concerning non‑Greek anthroponymy, Parissaki 2007, p. 289, argues that 54 PNs must be Thracian or – more generally – Pre‑Greek names.8 This terminology leaves the precise origin of this group of names unspecified. In this respect, most of the local anthroponyms are generally labelled as “Thracian” by default, despite the likely existence of previous non-Indo-European languages in the area and of different Thracian linguistic (sub)varieties over the centuries.9 In terms of chronological distribution, the ca 80 occurrences of Thracian PNs in the imperial period contrast with an average of ca 11‑12 PNs from the Pre-Hellenistic and Hellenistic periods. This is in part due to the establishment of the Roman province of Thrace in 46 AD and the later bestowing of the Roman citizenship thanks to the Constitutio Antoniniana (212 AD). Indigenous elites started to be more present in local politics and, by extension, in local epigraphy. For earlier periods, the presence of non‑Greek anthroponyms is hard to assess due to the conciseness of the inscriptions (reduced to PNs in most cases) and their combination with Greek(‑sounding) names (see infra § 2.5). It is thus difficult to determine the actual phonetic and morphological interpretation of these foreign names and, ultimately, reconstruct Thracian linguistic features. The reason is not only their Greek rendering with all its diachronic peculiarities (confusion of vowel quantity, iotacistic spellings, etc.), but also the approximative spellings of non‑Greek sounds and the Hellenization of indigenous words.
7From a methodological point of view, the linguistic analysis of Thracian onomastics must address different issues which make linguistic reconstruction susceptible to bias and over-interpretation: our inability to understand Thracian, the scantiness of (early) Greek inscriptions of considerable length and of bilingual texts, the lack of previous works on regional anthroponymy and the risk of approaching the names from a single Greek or Thracian perspective. To begin with, it is generally acknowledged that Thracian belongs to the Indo-European family.10 It is considered a Restssprache which has survived through Greek and Latin sources, mainly in some glosses and onomastics. Proper Thracian texts are limited to some undeciphered inscriptions from the 6th‑4th c. BC written in the Greek alphabet.11 Thracian anthroponymy – and the Thracian language – have also been generally examined from an exclusively Thracian perspective,12 which ranges from a unified view of the material (and the language) to a classification into different onomastic areas, as will be seen below (see infra § 2.1). However, a comparative approach and frequency analysis can allow the detection of Thracian PNs and their formants. The rich amount of Thracian onomastics shows further typological differentiation into simple names, suffixed names and compound names according to Indo-European typology.
8Recent studies on specific Thracian onomastic subregions are few (Dana 2016a; Proeva 2017), and those that deal with linguistic phenomena only focus on some specific features (Slavova 2006; Slavova 2007; Slavova 2015; Dana 2017; Dana 2019). Therefore, we still lack a systematic and comprehensive linguistic analysis of the onomastic material from this region describing the interaction between Greek and Thracian names diachronically and geographically. In this regard, would it be possible to spot the specific Greek variety – or varieties – with which Thracian came into contact over the course of centuries? In order to define the actual influence of the extant Greek varieties on Thracian onomastics, we must take into consideration that the earliest Ionic varieties that arose in the Aegean-Thrace area were largely linked to Cycladic Ionic (in colonies founded by Parians and Thasians) and Eastern Ionic (in Maroneia, Abdera, etc.). As mentioned above, Attic appeared in the region during the 5th c. BC, succeeded by the Attic-Ionic Koine.13
2.1. Aegean Thrace as a Thracian transitional onomastic area
9Unlike the obsolete unified and homogeneous analysis of Thracian anthroponymy, Dana (2012a and 2014) claims that Thracian PNs should be examined from a diachronic and geographical perspective in order to be able to draw a further internal division within the onomastic stocks. According to the author (OnomThrac, pp. LXIII‑LXXXIV), four different onomastic areas can be distinguished: (i) Pan-Thracian (or more generally Thracian), corresponding to Thrace proper and to the future homonymous Roman province; (ii) Dacian (or Daco-Moesian), in the North-Danubian territories; (iii) Western Thracian, which covers Eastern Macedonia, Thasos (and its peraea), the southern part of Moesia Superior and the western part of the Roman province of Thrace;14 and (iv) Bithynian, in Northwestern Asia Minor.15 This classification is based on the distribution and concentration of the PNs in specific Thracian territories which permit the establishment of different onomastic areas and, in some cases, of different micro-regions (as it occurs in Western Thrace, where we can identify the zones of Philippi, Sirrha, the valley of Nestos, etc.).16 The theoretical approach to this onomastic analysis is based on similar studies applied to other peripheral regions such as Macedonia, Illyria and some western provinces.17 Therefore, following Dana’s classification (OnomThrac, p. LXXVII), Aegean Thrace belongs to a transitional area between Thrace proper and Western Thrace and is defined as a mixed onomastic area (fig. 1). This classification primarily applies to the more abundant onomastic material of the imperial period.18
Fig. 1 – The four Thracian onomastic areas. T: Thracian, TO: Western Thracian, DM: Daco-Moesian, B: Bithynian (D. Dana, OnomThrac, p. LXXVII).

10In the case of Aegean Thrace, few Thracian PNs can be isolated as epichoric or regional due to the scarcity of evidence. In this regard, Dana (OnomThrac, p. LXXXI) asserts: “Il est probable que d’autres microrégions seront individualisées par les nouvelles découvertes, particulièrement dans les zones de contact entre les quatre territoires : ainsi, le nom Εζβενις est épichorique autour de la vallée du Nestos et en Thrace Égéenne”. Nevertheless, this salient anthroponym is attested only twice in this region between the 1st and the 3rd c. AD with different spellings (Εσβενο, Εζβενο), while it is widespread in Western Thrace.19
11In terms of occurrences, we find at least 15 of the 29 specific (Pan‑)Thracian features identified in OnomThrac, pp. LXV‑LVII (tab. 1; occurrences in parentheses). The four most common simple names are well documented: Κοτυς, Τηρης, Σευθης and Βειθυς. Some of these are well known, as they were also carried by Thracian kings. The latter also appears in dithematic formations such as Βειθυκενθου (E, no. 84, Topeiros, 46‑54 AD).20 The most common dithematic formations comprise Αουλου‑ as a first element and ‑ζενις/‑σενις as a second. If not mere spelling variants, the names Δολης/Δωλης/Doles attested in Aegean Thrace with a root in Δολo or Δωλo and an ‑ης ending would be Pan-Thracian (cf. Δολης [E, no. 84.22, Topeiros, 46‑54 AD]; Δολη, Δοληου [E, no. 212, Maroneia, 2nd‑1st c. BC]; Δωλεου̣ [E, no. 249, Maroneia, 2nd c. AD]; Δολεοῦ [E, no. 386, Asomatoi, 3rd c. AD?]). Alternatively, according to OnomThrac (pp. LXXIX, 155‑159 and 166‑169), the Western Thracian rendering of this name would correspond to Δουλο and Δολας/Dola, although there are no examples of them.
Tab. 1 – Thracian and Pan-Thracian PNs.
| Dithematic PNs | Simple PNs | |
| Αουλου‑ (8) | rescu‑ | Βειθυς (3) |
| Βειθυ‑ (1) | Ροιμη‑ (4) | Κοτυς (4) |
| brinca‑ | ‑κενθος (4) | Τηρης (3) |
| Δαλ‑ (1) | ‑πορις (5) | Σευθης (4) |
| derzi‑ | fem. ‑puius | |
| Δια‑ (3) | ‑tocus | |
| Δινι‑ (1) | ‑tralis | |
| ditu‑ | fem. τουρμη (1) | |
| diza‑ | fem. ‑sucus | |
| dul‑/δουλ‑ | ‑zelmis | |
| dutu‑ (f.) | ‑ζενις/‑σενις (8) | |
| Επται‑ (1) | ‑zeris | |
| μουκα‑ | ||
12Western Thracian features (iii) can also be observed, specifically in 6 of the 15 occurrences gathered in OnomThrac, pp. LXXVIII‑LXXIX. They comprise first elements in sudi‑ or μοκα‑, second formants in ‑bithu (perhaps in ‑πης) and the simple Tραλ(λ)ις:
- sudi‑ in Σουδεικένθου̣ (E, no. 21, Abdera, 2nd c. AD);
- Tραλ(λ)ις in Τραλλεις (SEG XL, no. 596, Kalyva, imperial period), which is a simple PN vs. compound names with a second element in ‑tralis;21
- ‑bithu in Κιληβύζō (E, no. 415, Zone, ca 400‑350 BC), vs. (Pan‑)Thracian bithu‑, Β(ε)ιθυ‑;
- μοκα‑ in Μοκαπορις (E, no. 84, Topeiros, 46‑54 AD; E, no. 387, Dymi, 1st c. AD)22 vs. Pan-Thracian μουκα‑;
- perhaps the suffix ‑πης in Σκωμπης (E, no. 105, Kalyva, 2nd c. BC).23
13Other forms could also be Western Thracian, such as Διουκίλου (E, no. 279, Maroneia, 2nd c. AD) provided that its first formant is associated with the names [Diula], Diulas and Διουλας: according to OnomThrac, p. 143, Diul‑ and Διουλ‑ seem to be characteristic of the Western Thracian onomastic area. There may be a further connection between Καρτους (E, no. 295, Maroneia, 2nd‑3rd c. AD) and fem. Καρτουζα (E, no. 328, Maroneia, 3rd c. AD; E, no. 450, Traianoupolis, imperial period) with Καρδους, considered Western Thracian (for a similar change ‑ρτ‑ > ‑ρδ‑, see Σπαρτοκίων > Σπαρδοκίων, OnomThrac, p. XCV).24
14The influence of the Dacian and Bithynian onomastic stock (iv) is not evident. In the latter case, the name Σουσᾶς (cf. Σ̣ουσᾶ Βείθυος, E, no. 482, unknown provenance, Roman period) could be characteristically Bithynian in light of the distribution of this name (although it has also been assigned to the Asia Minor onomasticon).25 Similarly, it has been proposed that Σποκης, found in the genitive in a late inscription (Κοισορμας Σποκου, E, no. 84, Topeiros, 46‑54 AD), and in mid-4th c. BC coins (Bα(σιλεὺς) Σποκης, Peter, Münzen, pp. 146‑147, territory of Abdera, ca 360 BC) could belong to a common Bithynian and South Thracian stock due to the geographical distribution of the examples (from Southern Thrace, Bithynia, Athens and Egypt). However, the evidence must be revised in the light of data from Egypt.26 In any event, the absence of Dacian and/or Bithynian PNs in our region perfectly matches the minor impact of these Thracian populations in Aegean Thrace, whose Greek character is more evident in the local epigraphy even during Roman and imperial periods.
2.2. Graeco-Thracian linguistic contact
15In the context of linguistic contact, there are different possibilities for introducing foreign names into a new (Greek) linguistic environment. They range from the direct preservation of the indigenous name to its full replacement. Therefore, non‑Greek anthroponyms could be (a) maintained in the original language, retaining their phonology and morphological features; (b) adapted into Greek; or (c) replaced by Greek names.27 The adaptation of Thracian PNs to Greek is the most common acculturation pattern in the accommodation process. The complete substitution of Thracian names could have taken place in certain contexts, but, to judge from the extant epigraphic evidence, they remained in use until a late date. Nevertheless, they coexisted with other Greek and Latin names within the same onomastic formulae; as a matter of fact, we do not find onomastic sequences exclusively composed of Thracian names (with nomen, cognomen, praenomen and other developments). The unaltered preservation of Thracian PNs cannot be ruled out, especially in some Thracian inscriptions written in the Greek alphabet in Inner Thrace between the 6th and the 4th c. BC.28 In Aegean Thrace, several simple names (Bendis’ names?) might exhibit Thracian features, but we cannot be certain about their original form. In what follows, we will concentrate on the mechanisms of adaptation of Thracian PNs.
2.3. Adaptation of Thracian names
16Different linguistic strategies can be adopted in the process of adaptation (i.e. translation, Hellenization, hybridation or cover names) which divide the PNs into two different onomastic systems.29 In Aegean Thrace, the addition of a Greek suffix to a Thracian root prevails.
2.3.1. Übersetzungsnamen
17There are no certain examples of Übersetzungsnamen (noms de traduction, “loan translation names”) in Thracian: it is a complex task when one of the translation languages remains undeciphered. From a typological perspective, some correspondences between Greek and Thracian dithematic formations have been proposed on the basis of a shared Indo-European root (cf. Thrac. Δια-ζενις and Gr. Διο-γένης; Thrac. Διζα-ζενις and Gr. Θεο-γένης) and in the light of similar typological formations (cf. Thrac. Βρια-ζενις with Ἀστυ-γένης; Δεσα-κενθος/Διζα-κενθος with Gr. Θεό-τεκνος). Thus, the first elements Δια‑, Διζα‑/Δεσα‑ and Βρια‑ would correspond to Greek Δι(ϝ)ο‑, Θεο‑ and Αστυ‑ respectively. The same would apply to the Thracian second formant ‑ζενις and the Greek ‑γένης (<*ĝenh1‑es).30
18We may find occurrences in Aegean Thrace in the simple name Διας (Διας Μουπορεως, E, no. 273, Maroneia, 2nd c. AD) and in the compound Διασενις/Διαζενις (cf. Αυλουζενις Διασενεος, E, no. 104, Abdera, 3rd c. AD; Ῥοιμηταλκου Διασενεως, E, no. 387, Dyme, 1st c. AD). Final ‑ζενις, which is “in majority in the Thracian region” (Dana 2019, p. 170) along with ‑tralis and the one with the most convincing etymology,31 is also present in the name Αυλουζενις (cf. Αυλουζενις Διασενεος, Τε̣ι̣ο̣υτ̣α σύμβιος Αυλουζενεος [E, no. 104, Abdera, 3rd c. AD]32; [Αὐλου]ζ̣ενις [E, no. 476, Plotinopolis, 2nd-3rd c. AD]; Αυλουζενις Αυλ̣ο̣υ̣[ζ]ε̣ν̣εος [E, no. 478, territory of Hadrianopolis, 2nd-3rd c. AD]). Furthermore, a correlation between the Greek Εὐρυγένης and Eβρυζενις has been proposed, based on the derivation of the first element from *h1u̯erH‑ and its connection to the name of the river Hebros.33 In Aegean Thrace, Εβρενις (E, no. 84, l. 25 and 27, Topeiros, 46‑54 AD, 2 ex.), another possible variant of the Εβρ‑ names, is attested in the cognomen of two strategoi.
2.3.2. Hellenised and hybrid names
19In areas experiencing intense contact, it is common to find hybrid PNs in which a non‑Greek base is combined with a Greek base and a Greek suffix, as in Βενδίδωρος, but there are no such examples from Aegean Thrace nor cases of a Greek base + Thracian suffix. Hellenised names (Thracian base + Greek suffix) represent the most significant group in the four Thracian onomastic regions. They reflect the indigenous adaptation to Greek inflection and, by extension, the acculturation processes in progress. According to Dana 2017, p. 202 (see also Dana 2019, p. 172), a PN can be Hellenised through the transformation/substitution of an onomastic element and through derivation; in these cases, hypocoristic suffixation and suffix variation play an important role. There are examples of both linguistic devices in Aegean Thrace. They can be better analysed in a broader category as morphological phenomena of accommodation (see infra §§ 2.3.3‑2.3.5).
2.3.3. Transformation/substitution of an onomastic element
20As an Indo-European language, Thracian probably distinguished between masculine and feminine, but the morphological differentiation of masculine ‑ος and feminine ‑α coincides with that of the Greek masculine and feminine suffixes, so that we cannot isolate proper Thracian inflectional morphemes:34 e.g. masc. Καπρουβηβου (E, no. 387, Dyme, 1st c. AD) and fem. Καπρουβηα (E, no. 290, Maroneia, imperial period);35 masc. ‑ορμας (Κοισορμας Σποκου, E, no. 84, Topeiros, 46‑54 AD; see infra § 2.5) and fem. ‑(τ)ουρμη (cf. gen. Δαρουτουρμης, E, no. 387, Dyme, 1st c. AD); and probably the hypocoristic masc. Καρτους (cf. Καρτους Γαΐου, ἥρως, Καρτο[υ…], E, no. 295, Maroneia, 2nd‑3rd c. AD, fun.)36 and fem. Καρτουζα (E, no. 450, Traianoupolis, imperial period; E, no. 328, Maroneia, 3rd c. AD). Further Greek-looking suffixes are masc. ‑ουλας37 and fem. ‑ουλα,38 attested in other Thracian regions (cf. Μηφαζουλα, IGBulg III, 1, no. 1349, Burdapa, 2nd‑3rd c. AD). They would presumably sound Greek, although ‑ουλα could be extracted from feminine roots like oβουλα or oδουλα, etc. The feminine ending ‑σαλα of the hapax Μακεσαλα Βειθυος (E, no. 482, unknown provenance, Roman period) may be related to other μακo PNs.39 These last cases point to a convergence with the Greek masculine forms in ‑ας and the feminine forms in ‑α/η.
21Anthroponyms in ‑πορις were adapted to the Greek declension through the transformation of the final element ‑πορις > ‑πολις, as already known in the Thracian space:40
- (1) βασιλεὺς Θρᾳκῶν Κ[οτυς Ῥασκου]πορεως (E, no. 458, Plotinopolis, 42‑31 BC, ded.);
- (2) Βεσουλας Μοκαπορεως (E, no. 387, Dyme, 1st c. AD, fun.);
- (3) Διας Μουπορεως (E, no. 273, Maroneia, 2nd c. AD, fun.);
- (4) Καρώσης Αὐλουπόρεος (E, no. 84.30, Topeiros, 46‑54 AD, hon.);
- (5) εὐεργέτην Κ̣[οτυν Ῥασκου]π̣ολεως βασιλέ[α Θρᾳκῶν] (E, no. 207, Maroneia, 1st c. BC-1st c. AD, hon.).
22Examples (1‑4) present ‑πορεως/πορεος, while (5) exhibits ‑πολις in Rhascuporis (ca 48‑42 BC), Cotys’ father and king of the Sapaioi, the same person as in (1). The context probably exerted some influence: (1) was a dedication to Heracles Soter by Cotys himself and (2‑3) come from the private sphere. Conversely, (5) is a public honorary inscription devoted to the king by a Greek community which called him euergetes. If the phonetic adaptation was spontaneous, it could be explained through an intermediate pronunciation between /l/ and /r/ in Thracian, but we do not have direct evidence of this phoneme. It seems more reasonable to see here a deliberate change which would respond to an attempt of acculturation. In terms of language contact, it can be analysed as a process of accommodation, although the final outcome (Thacian base + Greek ‑πολις) can also be viewed as a hybrid form. Nevertheless, this mechanism of substituting ‑πορις names with ‑πολις was far from systematic.
23The evidence must be studied regionally and diachronically in order to detect contemporary phenomena of linguistic levelling applied to Hellenised names. We would expect oαλκo names to have been transformed/accommodated into the s‑stem declension oαλκης, ‑ους. However, the nominatives ‑αλκας (cf. Σιταλκας [E, no. 212, Maroneia, 2nd‑1st c. BC]; Ῥοιμηταλκας [E, no. 84, Topeiros, 46‑54 AD, 2 ex.]) and the Attic-Koine genitive Ῥοιμηταλκου (E, no. 84, Topeiros, 46‑54 AD; E, no. 387, Dyme, 1st c. AD) point to an analogy with the masculine a‑stems, as happens with other Greek PNs in the region. Similarly, the nominative Ταρσας (Αὐρ(ήλιος) Ταρσας Μύρωνος, E, no. 194, Maroneia, 3rd c. AD) presents a genitive Ταρσου (Ἡδεῖα Τάρσου, E, no. 298, Maroneia, 2nd c. AD). Τhe expected forms are with η as in Ionic or, more probably, the Attic-Ionic Koine, unless ᾱ‑forms were related to the neighbouring Macedonian dialect. However, what it is striking is the arbitrariness of the ᾱ/η distribution across the Thracian territory, as well as in Aegean Thrace, a more Hellenised region with Ionic speakers where the Attic-Ionic Koine was rapidly imposed. As has been suggested, these forms might reveal an original Thracian vocalism /æː/, which is sometimes completely Hellenised as ‑αλκης, probably by Greek speakers in the earlier forms.41 Here the question also arises as to the identity of the authors of the epigraphic texts, which could be Greek speakers or Hellenised Thracians, gradually more present in the cultural and political life of these Greek colonies: oαλκας names are naturally expected from Thracian speakers, at least at an early stage, since they do not immediately associate this homophone onomastic element to Greek ‑αλκης. Thus, the graphic rendering of this /æː/ was approximatively associated to an /aː/ rather than an /εː/, morphologically motivated by the Greek declension. Be that as it may, the convergence of the genitives in ‑ου and ‑ους is common in s‑stems in Greek forms from the mid-4th c. BC,42 and the ᾱ/η variation concerns Thracian forms in general beyond our borders.
24The Koine declension in ‑ης, ‑ου of the masculine a‑stems can occasionally be observed in Aegean Thrace from the 2nd c. BC onwards: Δορζινθης (E, no. 365, Maroneia, ca 200‑150 BC; E, no. 84.37, Topeiros, 46‑54 AD), Δορζινθου (E, no. 457, territory of Traianoupolis, 3rd c. AD; E, no. 8.294, Topeiros, 46‑54 AD), Σευθης (E, no. 255, Maroneia, 1st c. BC), Σευθου (E, no. 271, Maroneia, 1st c. BC; Osborne, Byrne 1996, no. 3549, Athens, Hellenistic and imperial periods), Διουζιης (E, no. 84.34, Topeiros, 46‑54 AD), Δι[ο]υζ[ί]ου (E, no. 87, Topeiros, 1st‑2nd c. AD)43, and Τηρης (E, no. 389, Maximianoupolis, 3rd c. AD), Τηρου (E, no. 384, Agiasma, 3rd c. AD; E, no. 396, Maroneia [Dione], 235‑238 AD). The nominative Δολης (E, no. 84, l. 22, Topeiros, 46‑54 AD; E, no. 212, Maroneia, 2nd‑1st c. BC) exhibits different spellings of the genitive forms: Δοληου (E, no. 212, Maroneia, 2nd‑1st c. BC), Δωλεου̣ (E, no. 249, Maroneia, 2nd c. AD), Δολεου (E, no. 386, Asomatoi, 3rd c. AD?) [see supra § 2.1]. In other cases, the nominative in ‑ης is the only documented form: thus, Καρδενθης (E, no. 84.16, Topeiros, 46‑54 AD), Καρωσης (E, no. 84.16, Topeiros, 46‑54 AD) and the hapax Σκωμπης (E, no. 105, Kalyva, 2nd c. AD) present the Attic-Ionic form and not the aforementioned ‑ᾱς. Outside this region, Καρδενθης and Καρωσης only have genitives in ‑ου.44
25The reconstruction of the paradigm exclusively based on the genitive forms is misleading and analogies between s‑ and ā‑stems often take place, as discussed earlier (a standardized nominative is given, for instance, in the LGPN). Some PNs such as Καπρουβηβου (E, no. 387, Dyme, 1st c. AD) and Κιληβύζō (E, no. 415, Zone, ca 400‑350 BC) are attested only once in these genitive forms. For the latter, Parissaki 2007, p. 150 proposes a connection between ‑βύζō, a variant of Βυζος (Βυζας/Βυζης), and Βιζου (E, no. 386, Asomatoi, 3rd c. AD?), which she reconstructs as Βίζος.45
26The restoration of the nominatives corresponding to gen. Ταρουλου/Ταλουλου, Σποκoυ, Τυρελσoυ, Διουκιλου and Δαδου is problematic. Ταρουλας must be the common nominative of the genitive Ταρουλου/Ταλουλου in the Thracian regions due to the absence of nominatives in ‑ουλος/‑ουλης (Ἀλέξανδρος Ταρούλου ὁ καὶ Βέβιος, <Ζ>ώσιμος Ταρούλο<υ> [E, no. 379, Maroneia, 3rd c. AD]; Γλαυκίας Ταλουλου Μαρωνίτης [IG II2, no. 9286, Athens, imperial period]). Similarly, the frequency of the nominative Σποκης outside the Αegean coast justifies this reconstruction from Σποκoυ (E, no. 84.38, Topeiros, 46‑54 AD), which shows numerous genitive variants (Σποκη, Σποκητος, Σποκεους, Σποκειους, Σποκεως, Σποκηους). Τυρελσoυ, only documented in the genitive, is attested three times in Topeiros (E, no. 84, l. 33, 46‑54 AD; E, no. 86, ca 50‑100 AD; E, no. 87, 1st‑2nd c. AD) and in Moesia Inferior (IGBulg II, no. 737, 200‑250 AD). According to Dana (OnomThrac, p. XCVIII), it has an enlarged ‑σης suffix with an epethentic ‑λ‑ (cf. Didalsa/Διδαλσα, Δοιδαλσης). As regards Διουκιλου (E, no. 279, Maroneia, 2nd c. AD), a nominative ‑κε(ι)λας has been proposed in view of other cognates (Διουκιλας, Abydos, Hellenistic period), suggesting a Western Thracian origin (OnomThrac, p. 143). Finally, the genitive Δαδου (E, no. 153, Molyvoti, mid‑4th c. BC), a Thracian Lallname (see infra § 2.5), probably has a well-attested nominative Δαδας (Latin Dada, ae).46 Δαδου is the usual genitive form (7 ex.) from the 1st c. BC-1st c. AD; it appears twice in Macedonia, a Greek area where the oldest example dates from the 3rd c. BC.47
27Regarding ‑κενθου names (Τιβέριος Κλαύδιος Ῥοιμηταλκας Βειθυκενθου, E, no. 84, Topeiros, 46‑54 AD), only nominatives in ‑κενθος are attested in OnomThrac (instead of ‑κενθης). In Aegean Thrace, we have Δινικενθος, Επταικενθος (E, no. 84, l. 8 and 20, Topeiros, 46‑54 AD). This ending ‑κενθου resembles the one of the genitives in Καρδενθου (5 examples in OnomThrac, 1st‑3rd c. AD), whose nominative ‑ενθης is documented in Aegean Thrace in Καρδενθης (E, no. 84.16, Topeiros, 46‑54 AD), if the segmentation is correct.48
28The corresponding nominative of the dative Δύδῃ̣ (Δύδῃ̣ [τ]ῷ καὶ Εὐαγρίῳ, E, no. 88, Topeiros, 3rd c. AD) is reconstructed as Δυδης/Δουδης by Parissaki 2007, p. 166 but as Δυδης in OnomThrac, p. 172. The identification with Asia Minor Δ(ο)υδης (also Δουδας, Δουδουπης, fem. Δουδουσα) leads the former, following Zgusta Personennamen, § 306, to suggest a Thracian origin parallel to the homonym Lallname from Asia Minor. However, this further connection applied to this Lallname is far from certain. Tellingly, dud‑/δ(ο)υδ‑ names are separated from dyd‑, δυδ‑, δοιδ‑ formations in OnomThrac (pp. 166, 172 respectively), where only Dydies/Δυδιης forms and gen. Δυδιγγου could belong to the Western Thracian stock. In any event, it shows a clearly Attic(‑Ionic) ending.
29The Greek third declension could be reflected in isolated nominatives in ‑ις, such as Δινις (E, no. 84, l. 8 and 7; E, no. 85, Topeiros, 46‑54 AD, 3 ex.),49 Εβρενις (E, no. 84, l. 25 and 27, Topeiros, 46‑54 AD, 2 ex.) and Τραλλεις (SEG XL, no. 596, Kalyva, imperial period; see supra § 2.1). Iotacism could have affected these names, but nominatives in ‑ις (or ‑εις), not in ‑ης, are exclusively attested in OnomThrac. Other forms also seem to belong to this paradigm: nom. Δινις, gen. Δινεως, Δινεος, Δινιος, Δινιδος (also present in the toponym Δινισκορτα).50 The genitive Δαλιος (E, no. 394, Maroneia, 3rd c. AD) seems certain.51 However, the genitives in ‑εος/‑εως are ambiguous because of the loss of vowel quantity. They could be Ionic s‑stem genitives in ‑εος (vs. Koine ‑ους), but these are absent from our corpus of Greek PNs after the mid‑4th c. BC.52 Alternatively, they could belong to the i‑stems following the Koine declension ‑ις, ‑εως, alongside ‑ιος, as seen in Δαλιος.53 This seems to be the best explanation for the Thracian names in ‑πορεως, clearly assimilated into Greek ‑πόλις, ‑πόλεως. Consequently, the nominative of Bωσεος (E, no. 84, Topeiros, 46‑54 AD) would be better understood as Βωσις (LGPN IV; OnomThrac, p. 64) and not as Βωσης (Parissaki 2007, p. 36).54 This would also apply to the pair Διασενεος (E, no. 104, Abdera, 3rd c. AD) and Διασενεως (E, no. 387, Dyme, 1st c. AD). Regarding the genitives of Αυλοζενις, the genitive form Αυλουζενεος appears twice (E, no. 104, Abdera, 3rd c. AD; E, no. 478, Komara, 2nd‑3rd c. AD), whereas Aυλουζειδος (E, no. 212, Maroneia, 2nd‑1st c. BC) indicates a nominative Αυλοζεις, documented in Inner Thrace, and points to a different morphological adaptation into ‑ις, ‑ιδος. The suffix ‑ζις and its graphic variants (‑ζεις/‑ζι/‑σις/‑σι) are particularly productive in Thracian onomastics.55
30Βειθυος, the genitive of Βειθυς (Τιβέριος Κλαύδιος Βειθυς, E, no. 84, Topeiros, 46‑54 AD), is the most frequent form compared to Βειθυδος, which presents secondary dental forms. The compound Βειθυκενθου (Κλαύδιος Ῥοιμηταλκας Βειθυκενθου, E, no. 84, Topeiros, 46‑54 AD) preserves ‑υ‑ as the final part of the first element.56
2.3.4. Hypocoristic suffixation
31Dana (2017, pp. 203‑208 and 2019, pp. 171‑176) has already collected the most common Greek hypocoristic suffixes used to adapt Thracian names, i.e., masc. ‑ίων, ‑αρίων, ‑ων, ‑ᾶς, ‑ίσκος; fem. ‑ιον, ‑άριον, ‑οῦς, ‑αροῦς, ‑ώ, ‑αρώ. In the region, we find Thracian PNs in ‑ᾶς in the Roman period (ὑπὲρ τέκνου Σ̣ουσᾶ Βειθυος, E482, unknown provenance, Roman period), but also in an earlier Βοστᾶς, if it is a Thracian specimen (see infra § 2.3.5). There also exists Βενδιον (Βενδιον Ῥούφο<υ>, E, no. 379, Maroneia, 3rd c. AD), whose base form Βενδις is attested three times as a PN (E, no. 268, Maroneia, 2nd‑1st c. BC; E, no. 279, Maroneia, 2nd c. AD; E, no. 386, Asomatoi, 3rd c. AD?). If Thracian, Παπυλος (Ἡροφῶν Παπυλου, E, no. 212, Maroneia, 2nd‑1st c. BC) would be derived from Παπας (cf. παπ(π)ας “father”), a common Lallname in Asia Minor and Greece (see infra § 2.5).57 On Καρτους, see supra §§ 2.1, 2.3.3.
2.3.5. Ionic declension of Thracian names?
32Due to the overall late chronology of Thracian anthroponyms, the search for examples of Ionization in the indigenous names remains a difficult task. Apart from the cases already mentioned, one must explore the possibility of finding Ionic-defining morphological features in specific anthroponyms. In Π̣όρκης Πόρκεω (E, no. 184, Maroneia, 2nd c. BC), Πόρκεω presents the characteristic Ionic inflection (vs. Attic and Koine ‑ου), but there is no agreement on its Thracian origin, as seen in Detschew 1957, p. 375 and OnomThrac, p. 275. Brixhe 2006, p. 139 and Parissaki 2007, p. 233 relate it to Greek πόρκης “hoop” in the absence of more parallels from other Thracian regions.58 If Ionic, it would be a late instance of this Ionic genitive, as it also occurs in another rare Greek contemporary example Ἀθήνεω Μητροφάνου (E, no. 168, Maroneia, 167 BC), where ‑εω is combined with the Koine genitive Μητροφάνου (vs. regular Ionic Μητροφάνεος). Otherwise, if Thracian, it would presuppose a later phenomenon of morphological transfer from Ionic crystallized through onomastics in a moment where Ionic was not the main linguistic variant used in the area.
33As far as Thracian names in ‑ᾶς, ‑ᾶδος and ‑ῦδος are concerned, the genitive ‑ᾶ is attested in some mid‑4th century coins from Maroneia, in Βουτᾶ (Schönert-Geiss 1987, VIII, no. 502‑503, 365‑336 BC), and much later in Σουσᾶ (E, no. 482, unknown provenance, Roman period). Ionic ‑ᾶδος appears earlier in Zone in the patronymic Βοστᾶδος. In examples (1)‑(3), this genitive accompanies a Greek idionym, even though the origin of Αδα in (3) is uncertain (see infra § 2.5). In (4), we can read Βοστᾶδος or the abbreviated Βοστᾶ (found on a pottery graffito):
- (1) Ἀδάμας Βοστᾶδος (E, no. 404, Zone, mid‑5th c. BC);
- (2) Ἀπο(λ)λόδωρο̣ς̣ [Β]οστᾶδος (E, no. 412, Zone, ca 400‑350 BC);
- (3) Αδα Βοστᾶδος (E, no. 411, Zone, ca 400‑350 BC);
- (4) Bοστα(δος?) (Πρακτικά 1984‑1987, IA, no. 29, 4th BC).
34There are also other parallels from Thasos and other Thracian areas: Βοστᾶδος appears in a contemporary Thasian inscription ([Δ]όρι[λ]λος Βοσ̣τᾶδος, Athens, prior to 385 BC) and a later Bοστᾶ in Mysia ([Μη]νόφιλος Βοστᾶ, Lehmann 1917, pp. 185‑189, no. 1, col. A, l. 126, Kyzikos, 1st c. BC).59 The distribution of these examples supports the Thracian nature of the name, but Dana (OnomThrac, p. 64; OnomThracSuppl, p. 19) and Parissaki 2007, p. 152 raise doubts. Admittedly, Βοστᾶς has no convincing etymology and its similarity to other Ionic simple names (Βουτᾶς, Βοτᾶς or Βαστᾶς) and co‑occurrence with Greek names can be arguments for its Greek origin.60 If Thracian, its Hellenised form would be due to the influence of Ionic, as seen in contemporary Greek PNs (Mολπᾶδος, Abdera, 450‑425 BC, coin).
35The Ionic rendering of the genitive Ανδραβυδος in Θρᾶι[ττ]α Ανδραβυδος Μ<α>ρωνῖτις (IG II2, no. 9288, Athens, 1st c. BC-1st c. AD) is stressed by its ending in ‑υδος. Again its Thracian character,61 associated to Αδραβυς and to the Mysian city of Αδραμυττειον, can be questioned. It is included dubitanter in OnomThrac (p. 6), probably due to the presence of three late examples of the simple Andra from Syria (3rd c. AD), and in military diplomata found in Raetia (2nd c. AD). Furthermore, Russu 1958 and Lhôte 2007, p. 280 accept it as a Greek, more precisely as a compound name containing ἀνδραο + oβιος.62
2.4. “Thracianization” of Greek PNs?
36It should be noted that the inscriptions and graffiti unearthed in Zone (6th‑5th c. BC) could include examples of adaptation of Greek borrowings into Thracian. According to Brixhe 2006 and 2015, pp. 290‑292, the theonyms <ΑΒΟΛΟ> for the Greek Apollo and certain PNs are “Thracianised” forms, in the same way as the name of the Thracian goddess Bendis (Βενδις) also appears in a local version of the Greek alphabet as Bενζι. The anthroponyms <ΠΙΛΑYΕ> (Αρχαία Ζώνη I, no. 5, l. 18)63, <ΑΠΟΛΟΔΟΡΕ> (Αρχαία Ζώνη I, no. 281) and <ΝΕΟΚΛΕ> (Αρχαία Ζώνη I, no. 32) could be nominatives equivalent to Greek Φίλαιος, Ἀπολλόδωρος and Νεοκλῆς:64 according to Brixhe 2015, pp. 294‑297 and Brixhe 2006, § 6.5, the loss of ‑ς and the change of unstressed <O> to <E> in the nominatives in ‑ος could be explained as a case of vowel reduction of /o/ > [ə], a common isogloss shared with Macedonia, Thessaly (only in the region of Histiaeotis) and Thasos. Alternatively, Slavova 2017, pp. 123‑124 sees vocatives pro nominatives and links them to the undeciphered <ΡΟΛΙΣΤΕΝΕ(?)> (Plovdiv [Bulgaria], 5th c. BC) and <ΔΑΔΑΛΕΜΕ(?)> (Duvanlij [Bulgaria], 5th c. BC), but these sequences, if correctly distinguished, have been read as genitives as well.65 Slavova’s suggestion must be dismissed since we cannot isolate sequences of words in these inscriptions nor suffixes or word endings. Furthermore, Brixhe’s interpretation has been contested by Méndez Dosuna 2007: this change /o/ > /e/, also attested in modern Tsaconian in unstressed and stressed positions, is conditioned by the context. Besides, it does not imply any vocalic reduction.
37Even so, there is no convincing explanation as to whether this change is better explained as Greek or Thracian, since there is no later evidence of this process (Yanakieva 2018, pp. 50‑51). Admittedly, it only occurs after dental or palatal consonants in Φίλαιος and Ἀπολλόδωρος, as J. Méndez Dosuna has proved with respect to other Greek examples. The loss of final ‑ς in <ΝΕΟΚΛΕ> reflects a different situation, in which the name is found in its Ionic form Νεοo without an intervocalic *w (*newo‑).66 What seems certain is that this change cannot be interpreted as a Thracian phonetic feature, as it only seems to affect Greek thematic names in ‑ος. For this reason, these ‑E forms are better explained as vocatives for nominatives, as I have argued elsewhere (Guijarro Ruano forthcoming, § 3.2). This phenomenon is quite common in situations of linguistic contact in which Greek speakers were involved; for non‑Greek audiences, the use of vocatives of PNs was highly frequent and normal, so that they rendered these Greek vocatives Φίλαιε, Ἀπολλόδωρε and Nέοκλε (from Νέοκλος, not from Νεοκλῆς) in the indigenous texts we found in the Thracian sanctuary of Apollo, located in the Greek settlement of Zone. Indeed, cultu(r)al and linguistic contact in this setting was granted, as this epigraphic corpus attests. Therefore, the spelling of the theonyms and of these particular ‑E forms invites us to interpret them more likely as Greek borrowings.
2.5. Problematic and ambiguous names
38The assessment of the earliest attestations of Thracian PNs in the extant epigraphic corpus from Aegean Thrace is particularly challenging, since the majority of the well-established anthroponyms date from the imperial period. The Thracian character of non‑Greek sounding names before this period is controversial, especially as regards simple names, which makes their classification doubtful. Along with other instances already examined (Πορκης, Βοστᾶδος, etc.), the mid‑4th c. PNs Αδα (E, no. 411, Zone), Δαδας (Δαδου, E, no. 153, Molyvoti) and Νανη (E, no. 43, Abdera)67 seem to be epichoric Lallnamen. They are comparable to the other well-attested homonymous names from Asia Minor and have therefore often been considered Microasiatic. However, Dana (OnomThrac, pp. XCVIII‑XCIX) claims a Thracian origin, regardless of the (non‑Thracian) Microasiatic occurrences, because of a common Lallnamen CVC(V) structure and their frequency in Thracian onomastic areas, including Bithynia and Egypt. They are “exemples de convergence fortuite avec l’espace iranophone du Pont septentrional (Δαδας), avec la Grèce [Νανα (f.)], avec l’espace microasiatique et les langues asianiques [Δαδας, Διδας, Μαμα (f.), Λαλα (f.)]”. By way of illustration, Δαδας is particularly present in Moesia Inferior and testimonies also exist from Egypt, such as Δαδας Ζιπυρου Θρᾶιξ (Ι.Syringes, no. 693 and 698, Thebai, Hellenistic period). If this were so, our Δαδου would be one of the oldest Thracian PNs attested in Aegean Thrace, also present in Thasos in the same period (cf. Ἀμύντωρ Θάσιον Δαδας, ca 400‑350 BC).68
39While Δαδου is Phrygian for Parissaki 2007, p. 155 (with previous bibliography),69 Νανη and the later Παπυλος (Ἡροφῶν Παπυλου, E, no. 212, Maroneia, 2nd‑1st c. BC, see supra § 2.3.4) belong to “the world of the Ionian colonists” (Parissaki 2007, p. 219). Dana is less certain about Αδα, Νανη and Παπυλος, which are introduced with a question mark in OnomThrac. The interpretation of Αδα is indeed more complex: its low occurrence in Greece compared to its higher frequency in Caria casts doubt on its Thracian provenance (Detschew 1957, p. 6). According to Robert 1973, p. 441 n. 33, the PN spread from Caria to Miletus. From there, it arrived at other Ionian parts, as Apollonia Pontica, its colony, where another occurrence is attested (IGBulg I2, no. 415, Apollonia-Sozopolis, 5th‑4th c. BC). Another example from Thrace (Αδα̣ Ἱκεσίου ἐλευθέρα, IGBulg I2, no. 334(9)a, Mesambria, 4th c. BC) must be added to those from Ionia (see LGPN V.A and V.B). Furthermore, its early date allows a Greek interpretation, favoured by the idea that the name, Αδα Boστᾶδος, is (morphologically?) related to that of Ἀδάμας Βοστᾶδος (see supra § 2.3.5) since the two individuals could be siblings.70 The Greek or Carian origin of the name is thus not well established.
40Another problematic category is that of the dithematic names attested only once, such as Ανδραβυδος (see supra § 2.3.5) and Κοισορμας (Κοισορμας Σποκου, E, no. 84, Topeiros, 46‑54 AD).71 Despite the Thracian interpretation of Κοισορμας,72 Parissaki 2007, p. 201 does not exclude a Greek origin: Κοισ‑ and ‑όρμας/ορμος/ορμίδας are documented in other Greek PNs (see Κοισέας, Κοίσων, Κοισύρα and Ἐγχόρμας, Λυκόρμας, Πυθόρμας, respectively). It may be, then, a Greek idionym adopted by a Thracian or a Greek-sounding name.
41Cover names (noms d’assonance or Decknamen) are absent from our corpus. They began to appear quite late in Thracian, as those pointed out by Dana 2020 Mucianus/Μουκιανος, Pistus/Πιστος Mestrius/Mestrianus, Torquatus, etc. However, certain PNs could be felt as ambiguous at an earlier date in this bicultural region, “grec ou thrace selon l’appartenance culturelle de l’individu”, as suggested by Brixhe’s (2015, p. 295) interpretation of some names from the sanctuary of Apollo in Zone (see supra § 2.4).73 Along with other PNs already examined (Κοισορμας, Δαλιος, etc.), this could also be the case of simple names such as Δινις (see supra § 2.3.3), Διας (Διας Μουπoρεως, E, no. 273, Maroneia, 2nd c. AD) or Διειος (Π̣α̣υσίμαχον Διειου Χ̣[]ον, Ε, no. 177, Maroneia, 2nd c. BC). Moreover, the phonetic similarity of Τηρης, a clear Thracian dynastic name attested three times in our region (E, no. 389, Paisoulai-Maximianoupolis, 3rd c. AD; E, no. 384, Agiasma, 3rd c. AD; E, no. 396, Dione, 235‑238 AD), to other Greek PNs (Tηρεύς)74 has been underlined by Hatzopoulos in I.Chalcidique actes, p. 53 and Hatzopoulos, Loukopoulou 1996, p. 315; they suggest that this Odrysian PN probably became an epichoric name in Macedonia.
3. Conclusions
42Onomastics constitutes a significant tool for describing the linguistic situation in Aegean Thrace and for a meaningful approach to specific and diachronic linguistic contacts, especially when the evidence is limited. The hybrid onomastic character of this region, at least on a micro-regional level, is twofold. On the one hand, onomastic features of two different Thracian stocks converge in our corpus: over half of the specific anthroponymic characteristics of (Pan‑)Thracian names are present in the studied examples (15/29), and almost the same proportion applies to Western Thracian names (6/15). This mixed situtation was already sketched by Dana in OnomThrac (see supra § 2.1), but now we have at our disposal the actual data thanks to the thorough reappraisal of the corpus. On the other hand, Thracian PNs present clear hybrid or mixed formations, composed of a Thracian base and a Greek suffix (but not the opposite, a Greek base + a Thracian suffix, due to the Greek nature of the epigraphic evidence). To the best of our knowledge, it seems that Thracian names did not preserve their original form unaltered, but were adapted to the declension system of Greek, the dominant language in the region and the only one epigraphically attested (see supra §§ 2.3.2‑2.3.5).
43Linguistic contact between Greek and Thracian speakers becomes evident in Hellenised names, reintroduced into the Greek morphological system by means of phonetic/morphological accommodation, and bears testimony to the ongoing acculturation processes in this transitional area, also in later times. The final part of simple or compound PNs can be assimilated into quasi-homophonous Greek forms (‑πορις > ‑πολις; οαλκης/αλκας) or inflected through the introduction of Greek suffixes, as well as hypocoristics. It is the most frequent strategy for the formation of names common to two onomastic systems. The Hellenization of non‑Greek names – ultimately a case of onomastic hybridation – was affected by the specific linguistic variety that arose in this Greek area. Koine features are the best represented in our corpus, since most of the items date from the imperial period. The majority of the PNs follow the pattern of the Greek masculine ā‑stems in ‑ης, ‑ου and the same levelling characteristics of the third declension, although these are often obscured by the spelling. Nevertheless, we could find some Ionic traces in the genitive Πορκεω (2nd c. BC), if we assume a Thracian rather than a Greek origin for this name. Furthermore, in earlier periods, the existence of Βοστᾶς-names from the 5th c. BC probably points to an Ionic formation, if we accept the indigenous nature of this simple name, attested in other Thracian regions apart from Zone in Aegean Thrace. Whatever the case, due to the difficulty of isolating Thracian features and determining the original inflectional system of the Thracian language, simple Thracian anthroponyms are difficult to identify. In this regard, some Thracian names show homophony with Greek PNs at an early date, mainly Lallnamen and simple names, which prevents their classification. Finally, the spelling of some recognizable Greek names in some graffiti from Zone which contain non‑Greek words reveals the existence of real linguistic contact through the transcription of vocative forms of thematic PNs in local Thracian texts dated in the 6th‑5th c. (cf. Απολοδορε, Νεοκλε or Πιλαyε). These ‑E vocatives discard the possibility of identifying these particular linguistic features of Thracian, as it was proposed.
44In essence, this survey hopes to contribute to a more nuanced understanding of contact onomastics in outlying Greek areas. It can be further enriched by future research lines exploring cultural contacts as reflected in onomastic formulae (with the addition of Latin names) and within the broader field of onomastic studies focused on Thracian toponymy.
Bibliographie
Abbreviations
Abbreviations of editions and of works of reference for alphabetic Greek epigraphy are those of the list of the AIEGL published online: GrEpiAbbr, version January 2022, https://www.aiegl.org/grepiabbr.html (accessed 06/07/2023).
Αρχαία Ζώνη I: P. Tsatsopoulou-Kaloudi, C. Brixhe, C. Pardalidou, Αρχαία Ζώνη. Το Ιερό του Απόλλωνα, vol. I, Komotini, Υπουργείο Πολιτισμού και Αθλητισμού/Εφορεία Αρχαιοτήτων Έβρου, 2015.
OnomThrac: see Dana 2014.
OnomThracSuppl: D. Dana, OnomThracSuppl. Supplément actualisé de l’OnomThrac, Lyon, HiSoMA, 2020, https://www.hisoma.mom.fr/sites/hisoma.mom.fr/files/docs/Formations/ressources-numeriques/OnomThracSuppl version 7.pdf (accessed 21/07/2022).
Peter, Münzen: U. Peter, Die Münzen der thrakischen Dynasten (5.‑3. Jahrhundert v. Chr.). Hintergründe ihrer Prägung, Berlin, Akademie Verlag, 1997.
Πρακτικά 1984‑1987: A.G. Kalogeropoulou (dir.), Πρακτικά τοῦ Η’ Διεθνοῦς Συνεδρίου ἑλληνικῆς καὶ λατινικῆς ἐπιγραφικῆς, Ἀθήνα 3‑9 Οκτωβρίου 1982, vol. I‑II, Athens, Υπουργείο Πολιτισμού και Επιστημών, 1984‑1987.
References
del Barrio Vega 2015: M.L. del Barrio Vega, “Η ιωνική του βόρειου Αιγαίου (από τον Στρυμόνα μέχρι τον Έβρο”, in G.K. Giannakis (dir.), Ο γλωσσικός χάρτης της κεντρικής και βόρειας Ελλάδας κατά την αρχαιότητα, Thessaloniki, Centre for the Greek Language, 2015, pp. 153‑182.
del Barrio Vega 2018: M.L. del Barrio Vega, “The Ionic of North Aegean (from the River Strymon to the River Hebrus)”, in G.K. Giannakis, E. Crespo, P. Filos (dir.), Studies in Ancient Greek dialects. From Central Greece to the Black Sea, Berlin/Boston, De Gruyter, 2018, pp. 473‑510.
Beševliev 1959: V. Beševliev, “Nachträge zu den thrakischen Sprachresten”, Linguistique Balkanique 1, 1959, pp. 65‑68.
Beševliev 1970: V. Beševliev, Untersuchungen über die Personennamen bei den Thrakern, Amsterdam, Hakkert, 1970.
Bouzek, Graninger 2015: J. Bouzek, D. Graninger, “Geography”, in J. Valeva, E. Nankov, D. Graninger (dir.), A companion to Ancient Thrace, Malden/Oxford, Wiley-Blackwell, 2015, pp. 243‑264.
Brixhe 2006: C. Brixhe, “Zôné et Samothrace: lueurs sur la langue thrace et nouveau chapitre de la grammaire comparée?”, CRAI 150, 2006, pp. 121‑146.
Brixhe 2015: see Αρχαία Ζώνη I.
Brixhe 2018: C. Brixhe, “Thracian”, in J. Klein, B. Joseph, M. Fritz (dir.), Handbook of comparative and historical Indo-European linguistics, vol. III, Berlin/Boston, De Gruyter, 2018, pp. 1850‑1854.
Brixhe, Panayotou 1997: C. Brixhe, A. Panayotou, “Le thrace”, in F. Bader (dir.), Langues indo-européennes, Paris, CNRS, 1997, pp. 179‑203.
Curbera 2013: J. Curbera, “Simple names in Ionia”, in R. Parker (dir.), Personal names in Ancient Anatolia, Oxford, Oxford University Press, 2013, pp. 107‑144.
Dana 2006: D. Dana, “Les noms de facture thrace dans LGPN IV: les noms fantômes et d’autres corrections”, ZPE 157, 2006, pp. 127‑142.
Dana 2011: D. Dana, “L’impact de l’onomastique latine sur les onomastiques indigènes dans l’espace thrace”, in M. Dondin-Payre (dir.), Les noms de personnes dans l’Empire romain. Transformations, adaptation, évolution, Bordeaux, Ausonius, 2011, pp. 37‑87.
Dana 2012a: D. Dana, “La différenciation interne de l’onomastique thrace”, in T. Meißner (dir.), Personal names in the Western Roman world, Berlin, Verlag für Kunst & Kulturwissenschaften, 2012, pp. 223‑245.
Dana 2012b: D. Dana, review of “P.A. Dimitrov, Thracian language and Greek and Thracian epigraphy”, Ancient West & East 11, 2012, pp. 344‑347.
Dana 2014: D. Dana, Onomasticon Thracicum: répertoire des noms indigènes de Thrace, Macédoine Orientale, Mésies, Dacie et Bithynie, Athens, KERA, 2014.
Dana 2016a: D. Dana, “Onomastique indigène à Byzance et à Cyzique”, in M. Dana, F. Prêteux (dir.), Identité régionale, identités civiques autour des détroits des Dardanelles et du Bosphore (ve siècle av. J.‑C.-iie siècle ap. J.‑C.), Besançon, PUFC, 2016, pp. 47‑68, https://www.persee.fr/doc/dha_2108-1433_2016_sup_15_1_4073 (accessed 21/07/2022).
Dana 2016b: D. Dana, “Droit de réplique”, Studia Antiqua et Archaeologica 22, 2016, pp. 95‑102.
Dana 2017: D. Dana, “Hellénisation par suffixation: noms non grecs et suffixes grecs”, in A. Alonso Déniz, L. Dubois, C. Le Feuvre, S. Minon (dir.), La suffixation des anthroponymes grecs antiques (SAGA). Actes du colloque international de Lyon, 17‑19 septembre 2015, Université Jean-Moulin‑Lyon 3, Geneva, Droz, 2017, pp. 201‑223.
Dana 2019: D. Dana, “Onomastic interactions. Greek and Thracian names”, in R. Parker (dir.), Changing names: tradition and innovation in Ancient Greek onomastics, Oxford, Oxford University Press, 2019, pp. 167‑194.
Dana 2020: D. Dana, “Les noms d’assonance thrace: des miroirs culturels”, in C. Ruiz Darasse (dir.), Comment s’écrit l’autre? Sources épigraphiques et papyrologiques dans le monde méditerranéen antique, Bordeaux, Ausonius, 2020, pp. 61‑80.
Detschew 1957: D. Detschew, Die thrakischen Sprachreste, Vienna, Rohrer, 1957.
Dimitrov 2009: P. Dimitrov, Thracian language and Greek and Thracian epigraphy, Newcastle upon Tyne, Cambridge Scholars Publishing, 2009.
Dubois 2010: L. Dubois, “Des anthroponymes en ‑ους”, in R.W. Catling, F. Marchand (dir.), Onomatologos. Studies in Greek personal names presented to Elaine Matthews, Oxford, Oxbow books, 2010, pp. 398‑421.
Duridanov 1995: I. Duridanov, “Thrakische und dakische Namen”, in E. Eichler, G. Hilty, H. Löffler, H. Steger, L. Zgusta (dir.), Name studies: an international handbook of onomastics, vol. I, Berlin/New York, De Gruyter, 1995, pp. 820‑840.
Georgiev 1983: V. Georgiev, “Thrakische und dakische Namenkunde”, ANRW II/29/2, 1983, pp. 1195‑1213.
Guijarro Ruano 2019: P. Guijarro Ruano, “The northern coast of the Aegean Sea: a linguistic overview”, in M. Benedetti, C. Bruno, F. Logozzo, L. Tronci (dir.), Multilingualism in ancient Europe: Greek as language in contact, published in Linguarum Varietas 8, 2019, pp. 63‑84.
Guijarro Ruano 2023: P. Guijarro Ruano, “Dialect contact and Koineization: the case of the Greek colonies of Aegean Thrace”, in A.C. Cassio, S. Kazcko (dir.), Alloglо̄ssoi. Multilingualism and minority languages in Ancient Europe, Leiden/Boston, De Gruyter, 2023, pp. 115‑137.
Guijarro Ruano forthcoming: P. Guijarro Ruano, “Déplacements et contacts culturels dans le miroir de l’onomastique du sanctuaire d’Apollon à Zôné”, in R. Oreshko, F. Réveilhac (dir.), Déplacements, migrations et contacts culturels en Méditerranée orientale dans le miroir de l’anthroponymie, Paris, Sorbonne Université, forthcoming.
Hălmagi 2015: D. Hălmagi, review of “D. Dana, Onomasticon Thracicum: répertoire des noms indigènes de Thrace, Macédoine Orientale, Mésies, Dacie et Bithynie”, Studia Antiqua et Archaeologica 21, 2015, pp. 259‑271.
Hatzopoulos 2000: M.B. Hatzopoulos, “L’‘histoire par les noms’ en Macédoine”, in S. Hornblower, E. Matthews (dir.), Greek personal names. Their value as evidence, Oxford, Oxford University Press, 2000, pp. 99‑117.
Hatzopoulos, Loukopoulou 1996: M.B. Hatzopoulos, L. Loukopoulou, Recherches sur les marches orientales des Téménides, Athens, KERA, 1996.
Kretschmer 1896: P. Kretschmer, Einleitung in die Geschichte der griechischen Sprache, Göttingen, Vandenhoeck & Ruprecht, 1896.
Lehmann 1917: K. Lehmann, “Inschriften in Konstantinopel”, MDAI(A) 42, 1917, pp. 185‑190.
Lhôte 2007: É. Lhôte, “Typologie des anthroponymes en ‑υς”, in M.B. Hatzopoulos (dir.), Φωνῆς χαρακτὴρ ἐθνικός. Actes du Ve Congrès international de dialectologie grecque (Athènes, 28‑30 septembre 2006), Athens, KERA, 2007, pp. 271‑294.
Loukopoulou 2004: L.D. Loukopoulou, “Thrace from Nestos to Hebros”, in M.H. Hansen, T.H. Nielsen (dir.), An inventory of archaic and classical poleis, Oxford, Oxford University Press, 2004, pp. 870‑884.
Masson 1984: O. Masson, “Quelques noms de magistrats monétaires grecs”, RN 26, pp. 48‑60 (reprinted in Masson, OGS II, pp. 427‑439).
May 1966: J.M.F. May, The coinage of Abdera (540‑345 BC), London, Royal Numismatic Society, 1966.
Méndez Dosuna 2007: J. Méndez Dosuna, “Ex praesente lux”, in I. Hajnal (dir.), Die altgriechischen Dialekte. Wesen und Werden. Akten des Kolloquiums Freie Universität Berlin 19.‑22. September 2001, Innsbruck, Institut für Sprachen und Literaturen der Universität Innsbruck, 2007, pp. 356‑383.
Mihailov 1943: G. Mihailov, La langue des inscriptions grecques en Bulgarie. Phonétique et morphologie, Sofia, Universitetska Pečatnica, 1943.
Osborne, Byrne 1996: M.J. Osborne, S.G. Byrne, The foreign residents of Athens: an annex to the Lexicon of Greek personal names, Attica, Leuven, Peeters, 2006.
Panayotou 2007: A. Panayotou, “Greek and Thracian”, in A.‑P. Christidis (dir.), A history of Ancient Greek. From the beginnings to Late Antiquity, Cambridge, Cambridge University Press, 2007, pp. 745‑775.
Parissaki 2007: M.‑G. Parissaki, Prosopography and onomasticon of Aegean Thrace, Paris, KERA, 2007.
Parissaki 2009: M.‑G. Parissaki, “Étude sur l’organisation administrative de la Thrace à l’époque romaine. L’histoire des stratégies”, REG 122, pp. 319‑357.
Proeva 2017: N. Proeva, “On the names of Thracia and Eastern Macedonia”, in H. Popov, J. Tzvetkova (dir.), ΚΡΑΤΙΣΤΟΣ, Сборник в чест на професор Петър Делев, Sofia, Sofia University, 2017, pp. 75‑82.
Robert 1973: L. Robert, “Sur des inscriptions de Délos”, Études déliennes, publiées à l'occasion du centième anniversaire du début des fouilles de l’École française d’Athènes à Délos, Paris, École française d’Athènes, 1973, pp. 435‑489.
Robert 1978: L. Robert, “Documents d’Asie Mineure”, BCH 102, 1978, pp. 395‑543.
Russu 1958: I.I. Russu, “Thracica. Notes d’épigraphie et d’onomastique thrace”, Dacia 2, 1958, pp. 463‑466.
Schönert-Geiss 1987: E. Schönert-Geiss, Griechisches Münzwerk. Die Münzprägung von Maroneia, Berlin, Akademie Verlag, 1987.
Slavova 2006: M. Slavova, “Thracian names written through Greek and Latin letters: a philological approach”, Orpheus 16, 2006, pp. 79‑89.
Slavova 2007: M. Slavova, “Transliteration of Thracian names through the resources of the Greek alphabet: <i> digraphs in the Thracian names”, in A. Iakovidou (dir.), Thrace in the Graeco-Roman world. Proceedings of the 10th International Congress of Thracology, Athens, KERA, 2007, pp. 553‑556.
Slavova 2015: M. Slavova, “The Thracian diphthong /oi/ revisited”, Studia Classica Serdicensia, vol. IV, In honour of Prof. Boris Gerov (1903‑1991), Sofia, Universitetsko izdatelstvo “Sv. Kliment Okhridski”, 2015, pp. 139‑151.
Slavova 2017: M. Slavova, “The inscriptions from the mound necropolis of Duvanlii (Thrace) and their socio-cultural context”, Kadmos 56, pp. 119‑138.
Sowa 2020: W. Sowa, “Thracian”, Palaeohispanica 20, 2020, pp. 787‑817.
Tsiafaki 2020: D. Tsiafaki, “The Northern Aegean”, in F. de Angelis (dir.), A companion to Greeks across the Ancient World, Hoboken, Wiley-Blackwell, 2020, pp. 409‑430.
Tzitzilis 2014: C. Tzitzilis, “Greek and Thracian”, Encyclopedia of Ancient Greek language and linguistics, vol. II, Leiden/Boston, Brill, 2014, pp. 83‑89.
Vlahov 1972: K. Vlahov, “Typologie und Chronologie der Personennamen bei den Thrakern”, Thracia 1, 1972, pp. 249‑262.
Yanakieva 2018: S. Yanakieva, “The Thracian language”, Orpheus 25, 2018, pp. 26‑68.
Notes de bas de page
1 This paper is framed within the Spanish national research projects “Modos de contacto interdialectal en los textos epigráficos del griego antiguo II” (FFI2017-82590-C2-1-P) and “Onomástica y contactos lingüísticos en Griego Antiguo” (PID2020-114162GB-I00). I want to express my sincere gratitude to D. Dana, M.L. del Barrio Vega and the editors for their valuable comments on this work. All inconsistencies and errors remain my own. The following abbreviations are used: ded. = dedication, fem. = feminine, fun. = funerary inscription, gen. = genitive, Gr. = Greek, hon. = honorary inscription, masc. = masculine, nom. = nominative, Thrac. = Thracian.
2 On the geographical definition of this region and its poleis, see Loukopoulou 2004; Tsiafaki 2020.
3 Parissaki 2007, pp. 288‑289.
4 The corpus has been updated with later publications (SEG, BE, etc.). For the sake of clarity, inscriptions from I.Thrake Aeg. are quoted E, no. 1; E, no. 2, etc.
5 Along with the works quoted throughout this work, see Guijarro Ruano 2019 for an overview.
6 OnomThracSuppl is published on the website of HiSoMA (the version accessed in this work is that of December 2020, although, in the meanwhile, a more updated version [version 9] has seen the light in December 2022, which does not alter the conclusions of this research).
7 For an inventory of Latin PNs in Aegean Thrace, not examined in this work, see I.Thrake Aeg., pp. 613‑616; Parissaki 2007, pp. 282‑288; more generally, Dana 2011.
8 54 different PNs from over 92 references.
9 See Guijarro Ruano 2019, p. 74. Other non-Greek PNs belong to minorities: Egyptian names in Ἰσιo and Σεραπo appear between the 2nd c. BC and the 2nd c. AD. On the Egyptian character of Ἄβρατις (E, no. 202, Maroneia, 1st c. BC-1st c. AD) and Ψάμμις (E, no. 38, Abdera, early 4th c. BC), see Parissaki 2007, pp. 119 and 263. For the Galatian Ἀτευρίστου Βρίκκων (E, no. 215, Maroneia, early 2nd c. BC), see Parissaki 2007, pp. 144‑145; and for the Greek, as opposed to Illyrian, character of Ἀ̣ρ̣θ̣έων (E, no. 222, Maroneia, 4th c. BC), Parissaki 2007, pp. 136‑137. We do not consider Hebrew names from early Christian inscriptions (Ἰσαάκιος, Ἰωάννης, Μαρία, etc.).
10 Up‑to‑date accounts of the Thracian language in Brixhe, Panayotou 1997; Panayotou 2007; Tzitzilis 2014; Yanakieva 2018; Brixhe 2018; Sowa 2020. On epigraphic and linguistic research on this specific area, see Guijarro Ruano 2019, pp. 69‑76.
11 Thracian epigraphic texts can be found in Dimitrov 2009; however, his interpretation and linguistic analyses have been strongly questioned (see Dana 2012b). In our corpus, three inscriptions, gathered under the lemmata E no. 376‑378 (Maroneia, undated), seem to be written in an indigenous language identified as Thracian.
12 Previous works on Thracian onomastics (Beševliev 1970; Vlahov 1972; Georgiev 1983; Duridanov 1995, etc.) are of varying quality and must be revised in the light of new findings and theoretical approaches; see Dana 2012a, p. 225; OnomThrac, pp. XII‑XLI. For the problems of geographical limits, see Bouzek, Graninger 2015, pp. 12‑15.
13 On the Ionic varieties, see del Barrio Vega 2015 and 2018; and on contact and Koineization processes, Guijarro Ruano 2023. For a general and updated linguistic description of Aegean Thrace, see Guijarro Ruano 2019.
14 On Western Thrace, see D. Dana’s contribution in this volume, to whom I thank for the image of fig. 1 and for his comments on a previous version of this work.
15 The classification was contested by Hălmagi 2015, pp. 266‑267 (response in Dana 2016b) and by Proeva 2017 (but cf. OnomThracSuppl, p. 2, n. 3).
16 See OnomThrac, pp. LXXV and LXXVIII, where it is said that a general overview of the onomastics of this zone is still lacking.
17 See references in Dana 2012a, pp. 225‑226. Hatzopoulos’ methodology for the analysis of Macedonian names (see Hatzopoulos 2000) lies at the origin of much recent onomastic research.
18 Many PNs are concentrated in E, no. 84 (Topeiros, 46‑54 AD). It contains a list of 33 Thracian strategoi (23 with Roman citizenship). The stone was found in situ, but there are no explicit references to their origin or their districts, although some are known from other testimonies. Other names (Ἐσβενις Ἀματόκου, Καρώσης Αὐλουπόρεος, Ζυκουλήσης Τυρέλσου, Κοισόρμας Σπόκου) seem to be from southern Thracia (Parissaki 2009, p. 330, n. 34). The interest lies in how these anthroponyms were rendered and Hellenised in the local copy of the text.
19 Εσβενις Αματοκου (E, no. 84, Topeiros, 46‑54 AD), Κρονίων Ἐζβενεως τοῦ Τηρου, βουλ(ευτὴς) Μαρωνείτης (E, no. 396, Dione, 235‑238 AD). The PN has been etymologically derived from i.e. *h1k’wos “horse” (cf. Greek ἵππος, Latin equus), which in Thracian would correspond to a form *esvas or *ezvas (perhaps related to a toponym Ασβηνοί). See more references in OnomThrac, p. 185.
20 The term “dithematic” applies to non-simple PNs composed of two lexical items which, in principle, do not seem to be suffixes. See OnomThrac, pp. LXXXVI‑LXXXVII.
21 On this name and its use as an ethnic, see Parissaki 2007, p. 255 with previous bibliography.
22 Μουπορις (cf. Δίας Μουπόρεως, E, no. 273, Maroneia, 2nd c. AD) can be the original name or an abbreviated form of Μουκαπορις (Parissaki 2007, p. 217).
23 On this PN, see Parissaki 2007, p. 245; OnomThrac, p. 308 (not included in LGPN IV). For other simple PNs in ‑ης, see infra § 2.3.3.
24 On Καρδους, see OnomThrac, p. 78. The bithematic Καρδενθης (E, no. 84, Topeiros, 1st c. AD), well attested in Thracia proper, could be related to this series; if ‑δενθης is a variant of ‑δελθης, it could also be a Western Thracian feature (OnomThrac, p. 120).
25 Parissaki 2007, pp. 247‑248, n. 665; OnomThrac, pp. LXXX‑LXXXI. On the frequency of sus‑/συσ‑ in Bithynia, see OnomThrac, pp. 340‑341 with examples in OnomThracSuppl, p. 39.
26 Parissaki 2007, p. 248. Examples in OnomThrac, pp. 332‑333. For other Greek PNs inspired from Asia Minor hydronyms, see Parissaki 2007, p. 268; Dana 2019, pp. 180‑183.
27 Dana 2017, pp. 201‑202; Dana 2019, pp. 171‑172.
28 See supra n. 11.
29 An older introduction to the Hellenization and Latinization of PNs can be found in Beševliev 1970, pp. 38‑46. For Hellenization, see Dana 2017 and 2019.
30 For the alleged etymologies, see Detschew 1957, p. 181; Duridanov 1995, p. 827. See also Brixhe, Panayotou 1997, pp. 192‑193; Brixhe 2018, p. 1852. On the correspondences between translation names and agnomina, see OnomThrac, p. CIV.
31 OnomThrac, pp. 146 and 390; Sowa 2020, p. 793.
32 We follow Dana’s correction (2006, p. 142) for Τε̣ι̣ο̣υτ̣α (vs. Τα̣ρ̣ο̣υλα in I.Thrake Aeg.) due to the absence of parallels of this name. Femenine Τειουτα/Τιουτα/Τυτα is well attested (OnomThrac, p. 372).
33 Detschew 1957, p. 163; OnomThrac, p. 174. On the etymological reconstruction of Ἕβρος, see main references in Yanakieva 2018, p. 30.
34 Yanakieva 2018, pp. 49‑50.
35 On the reading of these unica PNs, see Parissaki 2007, p. 195. A hypothetical connection with Hesychius s.v. Κάπρονται· ἐκαλοῦντο οὕτως οἱ Θρᾷκες, and the epithet Καπρηνός (IGBulg V, no. 5323) would confirm their Thracian origin (OnomThrac, p. 77). ‑βηα/βεη appears (as a suffix/second element) in Δουτουβεη (IGBulg I2, no. 334, Mesambria, 2nd‑1st c. BC).
36 However, Gr. ‑οῦς generally forms feminine anthroponyms following the Greek pattern (cf. Thrac. Βενδοῦς, Δαντοῦς, Νανοῦς and Σευθοῦς in OnomThrac, p. 383). Owing to its recent chronology, could Καρτους be rather considered a masculine in ‑ος with an erroneous spelling? On the diffusion of the suffix ‑οῦς in LGPN IV, see Dubois 2010.
37 Βεισουλας Ἀρτεμιδώρου (E, no. 64, Abdera, 1st c. BC‑1st c. AD), Ἀλέξανδρος Ταρούλου ὁ καὶ Βέβιος, <Ζ>ώσιμος Ταρουλο<υ> (E, no. 379, Maroneia, 3rd c. AD) and Γλαυκίας Ταλουλου Μαρωνίτης (IG II2, no. 9286, Athens, imperial period). On the variants Ταρουλ(λ)ας, Ταλουρας, Ταλουλας, see OnomThrac, p. 349.
38 τ̣ῆ̣ς̣ γυναικὸς Βεσουλας Μοκαπορεως (E, no. 387, Dyme, 1st c. AD). Dana (OnomThrac, pp. 34 and 37) classifies Βεισουλας and Βεσουλα under different lemmata. Could they be associated to PNs in Βεσσός (Bes(s)us) built upon the ethnic?
39 OnomThrac, p. 204.
40 OnomThrac, p. 274. Οn ‑πολις, see Bechtel, Personennamen, pp. 375‑377. Most of ‑πολις PNs are masculine. See also Mihailov 1943, p. 87.
41 On the interpretations of α/η forms, see Yanakieva 2018, p. 41. The distribution of the variants lacks an obvious pattern: while we find Ροιμηταλκας/αλκης (gen. Ροιμηταλκα/αλκου) in Athens and in Ionic centers, Ροιμηταλκης, ‑ου is attested in Thracia (OnomThrac, pp. 293‑296). Be as it may, the alternation λ/ρ is frequently found in some Thracian names (see, for instance, Tarula, Ταρουλας, Ταρουλλας, Ταλουρας, Ταλουλας in OnomThrac, p. 349).
42 Εὐπείθου (E, no. 41, Abdera, ca 350‑300 BC), Διογένου (E, no. 257, Maroneia, 3rd c. BC), Διογέ̣ν̣ο̣υ̣, [Διογέ]νου (E, no. 259, Maroneia, Hellenistic period); more examples in Guijarro Ruano 2023, p. 130.
43 Διουζης is also attested, with a common ‑ζης suffix, see OnomThrac, pp. 145 and 392.
44 OnomThrac, pp. 77‑78.
45 On oβυζo, see OnomThrac, pp. 74‑75; for ι/υ, Mihailov 1943, p. 22.
46 OnomThrac, pp. 105‑106. A unique genitive Δαδεος (Paeonia, 2nd‑3rd c. AD) points to a different nominative Δαδης not yet attested.
47 See Καλλίστρατος Δαδου (Amphipolis, ca 250 BC), Δαδου (Sintike, imperial period) in OnomThrac, p. 105.
48 OnomThrac, pp. 77‑78. The name is Latinised as Cardentes. If related to cardo/καρδo, it could be connected to Καρτους (see supra § 2.1, n. 24).
49 On its Thracian character, see Parissaki 2007, p. 161; OnomThrac, pp. 135‑136.
50 OnomThrac, pp. 135‑136. On Εβρενις and Τραλλεις, see OnomThrac, pp. 174 and 378‑379 respectively.
51 On the Thracian interpretation of Δαλιος, see Parissaki 2007, pp. 155‑156 (with previous references); OnomThrac, p. 109; Guijarro Ruano forthcoming, § 3.2. See infra n. 73.
52 Guijarro Ruano 2023, pp. 126‑129.
53 On the Ionic genitives ‑ιος/‑ιδος in Aegean Thrace, see Guijarro Ruano 2023, p. 121, n. 32.
54 See Μουκατραλις Βειθηους Βοσις and Μουκατραλις Βοσ̣εος (IGBulg III, 2, no. 1690, b19 and c14 [respectively], Augusta Traiana territory, 202 AD). For the different spellings, see Zgusta, Personennamen, § 199.
55 See Αυλουζεις and Αυλουζειδος (OnomThrac, p. 22), but gen. Αυλουζενεω (IGBulg III, 2, no. 1794, Hadrianopolis, 2nd c. AD [date after OnomThrac, p. 19]). On the variants (nom. ‑ζηνεις/ζενις/ζηνης/σανις/ζανος), see OnomThrac, pp. 18‑22 and 406.
56 OnomThrac, p. XCIII; Sowa 2020, pp. 804‑805.
57 Robert, Noms indigènes, pp. 62‑63 and 57‑58, n. 1; Parissaki 2007, pp. 227‑228.
58 Along with Πορκης, see Πόρκος (LGPN III.A, Sicily, 5th c. BC). πορκ[ could be another Samothracian example of Πορκης (Brixhe 2006, p. 139). These PNs, absent from Bechtel, Personennamen, can be semantically associated to the fishing tacks and the fishing practice (cf. the meaning of πόρκης, πόρκος and πορκεύς).
59 Also, Βοσταν Διοκλέους (mentioned by Robert 1978, p. 530, Byzantium, 2nd c. BC). On its possible relation to the second part of the toponym Κηριβωστ̣[η]νοι (IGBulg I2, no. 270, Odessos territory), see OnomThrac, p. 64.
60 On these Ionic names, see Curbera 2013. For its interpretation as Iranian (Βοστακων/Βοσταγων) or Thracian, see references in Parissaki 2007, p. 152.
61 Detschew 1957, p. 17; Parissaki 2007, p. 131.
62 On Ανδραο/Αδραο and β > μ, see Kretschmer 1896, p. 390.
63 The name presents a local spelling which Brixhe renders with <y> for /i/ and /y/. It is also present in πλαyε (no. 15), the same PN with syncope, and in other examples from Zone and Samothrace. For Brixhe 2015, pp. 285‑286, it represents a semitic yod, as it appears in Phrygia and Lemnos. Greek Φίλαιος is attested in Greek coins from Abdera: ἐπὶ Φιλαίō (May 1966, VII, no. 426‑428, 360‑350 BC).
64 The form Oλβε (no. 19) could also be understood in the same sense; see Brixhe 2015, pp. 295 and 301. See Guijarro Ruano forthcoming, § 3.2.
65 See Dimitrov 2009, pp. 13‑17.
66 For Brixhe 2015, p. 216, the loss of final ‑ς is a Thracian grammatical feature, but the name seems to be followed by a patronym κυτονιδες (cf. Ionic ‑ίδης) with a written final ‑ς. This leads Brixhe to justify the exception because of the Greek character of this onomastic formula νεοκλε κυτονιδες. However, we cannot be certain that the loss of final /s/ was a Thracian feature since we have no examples from Thracian. We can detect it in a reduced number of Greek forms. Furthermore, if it is an athematic gen. ‑ιδος > ‑ιδες, as also C. Brixhe proposes, it remains difficult to explain why the weakening of final /s/ did not affect this word; conversely, it is better explained if we take into account that it is not a thematic masculine in ‑ος.
67 In E, no. 410 (Zone, 450‑400 BC), the editors reconstruct Νανη in ]νη [ ]λ̣βίο.
68 OnomThrac, pp. 105‑106.
69 On Microasiatic examples, see Zgusta, Personennamen, § 244.
70 I.Thrake Aeg., p. 518.
71 Also, [Κοι]σορμου (IGBulg IV, no. 2057, Pautalia, 135 AD)? See OnomThrac, p. 88.
72 Beševliev 1959, p. 68; Dana 2006, p. 138; OnomThrac, p. 88.
73 On ΔαλιεΨα (Αρχαία Ζώνη I, no. 20) and the simple (gen.) Δαλιος (E, no. 394, Maroneia, 3rd c. AD), see supra § 2.3.3 and other PNs in Δαλαo. For a more in‑depth study, see Guijarro Ruano forthcoming, § 3.2.
74 Also, Τηρίας, Τήριλλος, Εὐτηρίδας (< τηρός ‘guardian’) in Bechtel, Personennamen, p. 426.
Auteur
-
Paloma Guijarro Ruano
ORCID : 0000-0002-7551-7982
Universidad Complutense de Madrid
Le texte seul est utilisable sous licence Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Dire la ville en grec aux époques antique et byzantine
Actes du colloque de Créteil, 10-11 juin 2016
Liliane Lopez-Rabatel, Virginie Mathé et Jean-Charles Moretti (dir.)
2020
Dérivation nominale et innovations dans les langues indo‑européennes anciennes
Alain Blanc et Isabelle Boehm (dir.)
2021
Stylistique et poétique de l'épigramme latine
Nouvelles études
Daniel Vallat et Florence Garambois-Vasquez (dir.)
2022
Contacts linguistiques en Grèce ancienne
Diachronie et synchronie
Alcorac Alonso Déniz, Julián V. Méndez Dosuna, Enrique Nieto Izquierdo et al. (dir.)
2024
Les représentations du désir féminin dans la poésie latine, de Catulle à Juvénal
Contraintes et libertés de l’espace poétique
Giacomo Dimaggio
2024
Définir l'épopée en Grèce ancienne
Contribution à une histoire de l'idée d'épopée d'Hérodote à Tzetzès
Flore Kimmel-Clauzet
2024
Lector in poematis
Le lecteur dans la poésie gréco-latine
Bruno Bureau, Bénédicte Delignon, Jean-Christophe Jolivet et al. (dir.)
2025
