CC BY-NC-ND
0
Florian Réveilhac
University of Marburg, Germany; florian.reveUhac@gmail.com
*Tubar(i)-: a divine epithet reflected in Luwic onomastics
This article extends and discusses a study by M. Valério (2015), which deals with several groups of personal names from southern Anatolia. It is proposed here to reconstruct one onomastic stem, *tubar(i)-, common to several names from different language corpora belonging to the Luwic sub-group (Luwian, Carian, Pisidian) and Greek epigraphic sources from southern Anatolia (from Caria to Cilicia). This prolific element is associated with various divine names or epithets in compounds, which suggests that it corresponded to a divine title. Its meaning can be reconstructed as "battle companion, comrade-in-arms", originally qualifying various deities whose role was to guard their protégé on the battlefield.
Keywords: Luwic languages; Anatolian languages; Luwian language; Lycian A language; Lycian B language; Carian language; Pisidian language; Greek language; onomastics; theo-phoric names; Anatolian religion.
1. Introduction
For a long time, the Lycian place-name Tyberissos has been associated with the name of a Lycian hero, Toupepig indirectly referenced in a local myth documented by Stephanus of Byzantium.1 In a complementary study to the present article, I sought to ascertain the validity of this connection — whether Toupepic; was genuinely linked to the toponym or if it was a derivative form originating from it.2 During this investigation, I uncovered an onomastic stem *tubar(i)-attested in numerous Luwic anthroponyms, which appear not only in Luwian-, Carian-, or Pisidian-language records, but also in Greek sources from southern Anatolia. The extensive scope of these forms precluded a comprehensive analysis within the previous article, primarily dedicated to the toponym Tyberissos. Hence, the primary goal of the current work is to delve into this deeper exploration.
The present contribution lists and analyses all the names containing the stem *tubar(i)-across various corpora from southern Anatolia (§ 2). It further proposes a correlation between this onomastic stem, on the one hand, and Carian names in °8ubr-/°8ybr- as well as °v5upepi- / °v5o/(u)pa/e/npa- (in Greek transmission), on the other hand. While these names were dealt with separately in Valerio 2015, it is argued that [d] in the latter group is an allophone of /t/ in the position after [n] or [r] (§ 3). After examining a few anthroponymic compounds where *tubar(i)- appears as the first member (§ 4), the paper provides an overview of all personal names formed using this morpheme, categorizing them based on the accompanying element.
1 This paper has been written under the auspices of the Alexander von Humboldt Fellowship, which enabled my extended academic stay at the University of Marburg. At various stages of its development, it greatly benefited from the comments of Ignasi-Xavier Adiego, Alcorac Alonso Deniz, H. Craig Melchert, Enrique Nieto Izquierdo, Diether Schurr, and Ilya Yakubovich, to all of whom I am grateful. Ilya Yakubovich deserves additional thanks for his stylistic advice and the Russian translation. However, I am solely responsible for any errors that may be found.
2 For a discussion about the heronym and a full analysis of the toponym, I refer to Reveilhac forthc. a. The reader should be aware that sections 5, 6 and 7 of the present article are adaptations of passages already included in the article on Tyberissos. These paragraphs are, in fact, key points in the argumentation, which I feel justified in repeating for the sake of clarity.
Journal of Language Relationship • Вопросы языкового родства • 21/3-4 (2023) • Pp. 259-281 • © Florian Reveilhac, 2023
It becomes apparent that, in the majority of instances, *tubar(i)- is linked to a theonym or a divine epithet, possibly functioning itself as a substitute for a divine name (§ 5). The article endeavours to establish that this stem originally represented a divine title, later adopted as a divine hypostasis. This title might have initially meant "battle companion, comrade-in-arms", referencing the protective roles of tutelary deities on the battlefield — a concept echoed in Hittite and Luwian texts concerning various deities (§ 6). The concluding section compares this interpretation with instances of Lyc. B tuburiz, mentioned three times on the Xanthos Stele, albeit within obscure contexts (§ 7).
2. The onomastic stem tubar(i)-
While the Lycian hero Toupepig is only known from late Greek sources, the stem *tubar(i)- is far better attested. Numerous personal names from southern Anatolia contain elements akin to the heronym conveyed by Stephanus of Byzantium. These potential candidates, found in various sources such as Hieroglyphic Luwian, Carian, Pisidian, and Greek inscriptions, are presented in Table 1. It is important to note that, at this stage, the comparison is purely formal and establishes an initial working hypothesis that needs further investigation.
Region (language of attestation) Personal name City Period Reference
Central Anatolia (Hieroglyphic Luwian) /TuPar(i)-/ (tu-pari, nom. sg.) Hattusa (Bogazkoy) ? BoHa 19, no. 467
D-tybr- Thebes ?6lh-5lh c. BCE E.Th 2.1; Adiego 2007: 96
Egypt Dybr- Thebes ?6lh-5lh c. BCE E.Th 5.1; Adiego 2007: 97
(Carian) Ksa-tybr- Thebes ?6lh-5lh c. BCE E.Th 2.2; Adiego 2007: 96
Kud-tubr- Thebes ?6lh-5lh c. BCE E.Th 9; Adiego 2007; 99
Caria (Greek) Zayyo-xpnQLÇ Arlisseis ?354 BCE I.Carie hautes terres: 91.3
Lycia (Greek) *EQ|ja-T0|30QLÇ (gén. EQ|aaT0|30QL0ç) Tlos 1s' c. BCE TAM II: 550.21
Pisidia (Pisidian) T[|3]£QL-? Solufar 2nd-3rd c. AD N 15; Brixhe 2016: 84
Pisidia (Greek) neQxa-xoupaQLç Termessos Imp. I.Termessos Suppl. IV: 145.1, 2-3; 146.1
Table 1. Personal names potentially formed with the stem *tubar(i)-
The name /TuPar(i)-/ attested in Hattusa in hieroglyphic transmission might be a direct match of Toupepig's underlying name, but its isolated character hardly allows us to go any further. Another simple name, dating to the Roman period in Pisidian records, is T[p]epi-, although unfortunately incomplete. Its restauration, based on Cilician compounds of Greek transmission in Tpep° (see Section 4 below), gains additional support from the Carian name Eayyo-xpnpig. To progress further, it is imperative to examine compound names, primarily in Greek transmission, which tend to be more understandable than those from Carian sources.
The vocalism of °xopopig in Epjaa-xopopig reflects a specific evolution found in Lycian, also observable in the toponym Tuburehe(/i)- contrasted with its Greek equivalent Tupepiaaog. The latter is likely adapted from a more archaic dialectal form (as suggested by Reveilhac forthc. a).3 While the first element of Ep^a-xopopig is easily identifiable as the name of the Moon-god Arma (cf. Ermme-nene/i-, Ap^a-niag etc.), that of nepxa-xoupapig is less apparent. However, the same base might be present in the Lycian place-name napxa^aaog4 and the probable Lycian title parttala- (TL 35.14 and 16) of uncertain meaning, which does not significantly aid in its comprehension. Should one accept the equivalence between the Lycian name napxaaig and the divine epithet CLuw. Parattass(i)-, a title found twice for the Storm-god with a likely meaning of "of Impurity" (based on CLuw. paratt(a)- "impurity"),5 the hypothetical Lycian stem-form *partta- might have been directly used to form compounds, similarly to Luw. /piha-/ "brightness, splendour" (Piha-muwa-, Piha-nan(i)-) compared to *pihassa/i- "of brightness, splendour" (Pihassa-muwa-, Pihassa-walwi-).
In the name Eayyo-xp^pig, a syncope in the second member's first syllable is plausible, as it is well attested in Carian for unaccented vowels (Adiego 2007: 241-242). The isolated first element, Eayyo-, is also found in the simple personal name Eayywg (KPN § 1369; Halicarnas-sus, 5th c. BCE). Although its nature and meaning require further clarification, a speculative hypothesis can be put forward: if the theonym dSalupiya- (van Gessel 1998: 370), documented once in a Hittite cult inventory,6 indeed corresponds to a compound (Satzname?), the first element being the name of the Hurro-Hittite goddess Salu-, wife of Kumarbi (Laroche 1946: 57; van Gessel 1998: 370), then a similar analysis might apply to dSanhupiya-, referring to a deity honoured in the northern regions of the Hittite Empire (van Gessel 1998: 371; Taracha 2009: 102). Thus, a hypothetical theonym *dSanhu- could be inferred from dSanhupiya- and linked to Eayywg and Eayyo-xp^pig.
In the Carian corpus, the identification of a similar element is supported by D-tybr-, where the first element can be traced back to *Ida-. This connection is evident in other Carian and Lycian forms such as Car. D-quq—> I5a-yuyog or D-wsoA-/Id-usol—> IS-uaawAAog.7 Similarly, in Ksa-tybr- one can isolate the stem Ksa-, which is potentially linked to the group of Luw. */hassa-/ "hearth" (inferred from CLuw. hassanitt(i)- "hearth"), Hitt. hassa- "hearth" and Lyc. xaha- "id.". An alternative hypothesis is to link it to the element found in the patronymic of Thales, 'E^a-iauqg,8 and in E^a-poag/A^a-pwg.9 The combination of the element E^a- with -|auqg (^ muwa- "power, drive")10 hints at a possible association with a deity, although a de-
3 This evolution proceeds from a progressive assimilation, favoured by the presence of the bilabial fricative rendered by (b). There is also a matching phenomenon of regressive assimilation -e-u- > -u-u-, as in the case of urublije- "monument" as opposed to erublije-, attested once and probably derived from the verb eruwe- "to build" (Martinez Rodriguez 2019: 221).
4 This place name is documented in the Life of Saint Nicholas of Sion (Anrich 1913: § 57.16-17); the saint lived in Lycia under Justinian and his hagiography was written shortly after his death (Anrich 1917: 536). On the value of this Byzantine text for the study of local toponymy, see Robert 1955: 197-208.
5 KUB 7.14(+) i 2-3 and KBo 29.30+ iii 6' (CTH 694.1). See Laroche 1946: 69 (with an outdated semantic connection); van Gessel 1998: 350; Mouton and Yakubovich 2021: 32. Cf. CLuw. («)paratta- "impurity" (vel sim.), Mel-chert 1993: 167.
6 About this text (CTH 510), see Cammarosano 2015: 221.
7 The same element is found in Carian in forms such as D-biks- or D-bkm- (Adiego 2007: 334) and in Lycian in
the simple name EiSa (KPN § 451-2) and in compounds such as Ida-maxxza—> ISa-|ja^ag or Idazzala--> EiSa-
aaaAag (vs. Zzala—> EaAag or Ep^a-aaAag); see Reveilhac 2018: 501-502.
8 D.L. 1.22, 29: Sud. s.v. 0aA|g; St.Byz. s.v. MiA^xog.
9 E^apoag: Balboura, 2nd c. BCE; A^apwg: Pisidia, 2nd c. BCE.
10 On this lexeme, see now Valerio 2023.
finitive etymology remains uncertain at this stage. Regarding Kud-tubr-, the element Kud-might mirror Luw. */huda-/ "haste, swiftness", documented in Hittite as Glossenkeilwort in the form huda- and as a derivational base in CLuw. hutarlann(i)- and hutarli(ya)-, and Lyc. *xuda-"haste", deduced from the agent noun xddaza- "slave" and the personal names Xudali(je)-, Xudalijë- and Xudara- (Rieken & Sasseville 2022). However, accepting this link requires assuming an evolution PLuw. *hu > Car. ku, in addition to the established change into qu (e.g. *Tarhunt- > Trquô-, *huha- "grandfather" > quq-), which demands further supporting evidence. If this hypothesis were correct, Kud- might represent an epithet that is close in meaning to Hittite nuntar in nuntaras dLAMMA "Tutelary Deity of Swiftness" (see van Gessel 1998: 694 for the attestations).
Yet, it is crucial to examine the connection between the elements tybr- and tubr-, whose vocalism differs. To consider them as potential variants of the same morpheme, it is necessary to assume an effect of i-umlaut triggering the fronting of [u] > [y] in tybr-: PLuw. *[tubari-] > PCar. *[tyberi-] > Car. [tybr-]. Such a phenomenon is known in Carian (as well as Lycian) for other vowels, as seen in ted- "father" (cf. Lyc. tede/i-) vis-à-vis Luw. /tad(i)-/ "id." or in en-"mother" (cf. Lyc. ëne/i-) vis-à-vis Luw. /ann(i)-/ "id." (Adiego 2007: 259). Adiego (2007: 257) suggested that the same pattern accounts for the ethnic suffix -y/yn- (e.g. kbdyn- "Caunian") derived from PLuw. *-wanna/i- (cf. Luw. /-wanna/i-/, Lyc. B -wnne/i-, Lyc. A -nne/i-).11 Melchert (2021: 108) gives arguments against the conditioned fronting of [u] in [y] that can be countered: 1) ybt need not correspond to the preterite "(he) offered" (< *uboto, cf. Lyc. ubete), but could just as plausibly represent the present in *-ti "(he) offers" (as already proposed by Adiego 2007: 259); 2) the analysis of tyn as the infinitive of a form related to Luw. /tuwa-/ relies on the overall interpretation of the inscription C.Ha 1, which is not based on communis opinio (other hypotheses exist, see Adiego 2007: 283-284); 3) the final morpheme (better than "suffix") °eym- in Paraeym- probably represents a participial stem in *-mma/i- subject to i-mutation, as suggested by the name OQi5-nu|aiç (vis-à-vis Aqt-^u^oç), which would justify the fronting of [u] — [y].12
Efforts have been taken to establish connections between the aforementioned names to other names from the Carian corpus and the Greek corpus in Asia Minor. The most comprehensive exploration on this subject, to the best of my knowledge, is found in Valério 2015, which builds upon earlier research concerning the Linear A du-pui-re and the Hittite t/labarna-(Valério 2007). In this contribution, several pages are devoted to Carian names in °DUbr-,13 Lycian names in °v5up£Qi-, and Cilician names in °v5o/(u)pa/e/nQ«-, as well as Lycian Tubure/i-, °xopoQL- (Valério 2015: 333-339). Acknowledging the challenge posed by the interpretation of the personal names from the Carian sources, the paper eventually arrives at the conclusion that reducing all the concerned personal names to a singular and variable element *DUbr- is unfeasible. Valério 2015 suggests instead the stem *(°)y/yb(y)r-, associated with the alleged element *(°)up£Qi- identified in Neumann 2007: 399. Following Neumann's proposal, Valério cautiously attempts to link this virtual element with the group of Hitt. warr- "to help" and HLuw. /warri(ya)-/ "id.", albeit acknowledging a major phonetic challenge: such a hypothesis would require admitting not a proto-form *warri- but **uwarri- > *uweri- (Valério 2015: 335). This sug-
11 The development must have been as follows: *-wanni (common gender direct case) > (*-wni >) *-uni- > -yn-(in the Egypto-Carian system)/-yn-. In addition to the rule [a] —> [e] / (C) _.C[i], there would therefore also be [u] —> [y] / (C)_.C[i] or, to subsume both, a more general phonological rule V[+back] —> [-back] / (C)_.C[i].
12 It is tempting to see the element °eym—> °r|U|jog / °r|U|jig as a cognate of the stem found in the Lycian personal names Uweme/i- and Un-uweme/i-: it would then represent the participle of a verb probably meaning "to look (favourably)" (CLuw. awa- "to see"; Lyc. A and B uwe- "to look favourably"; Lyd. ow- "id.", according to Sasseville 2020: 289-290).
13 Where D = voiced or voiceless dental, and U = [u] or [y].
gestion is also based on the assumption that an intervocalic *w can be strengthened into [p]. However, similar to the situation in Lycian, there is a lack of such a phonetic process in Carian.14
One argument put forward to advocate the Carian anthroponymic (and lexical) stem *(°)y/yb(y)r- is the existence of the simple personal name in the genitive Ybrss (C.Hy 1, 7-8). Nevertheless, unlike the previous forms, it represents an s-stem. This anthroponym finds a plausible avatar in the name Oppaaig attested in Cilicia (LGPN 5b; Olba, 1st c. BCE), easily aligning with the larger group of personal names in uppara-/o(u)npa-/o(u)ppa- (Houwink ten Cate 1961: 162-164; Melchert 2013: 44). It is conceivable that the [y] might have arisen due to the i-mutation affecting the suffix *-assa/i- (as seen in Oppaaig). After excluding Ybrss from the equation, a Carian stem (°)tUbr- (with U = [u] or [y]) emerges, showing similarity to °xopopig/ °xoupapig/°Tpnpig, or even to the Luwian name /TuPar(i)-/, and the Pisidian name T[p]epi-.
The case of Car. Dybr- presents a more complex challenge. The initial (d) might have emerged from an interchange with (t), following the neutralization of the two underlying phonemes in the initial position, if one follows a plausible hypothesis that /d/ and /b/ may have undergone a devoicing in this position in Carian.15 Another possibility is that it could be secondarily derived from the allomorph °dybyr- found after [n] or [r], as in Ar-dybyr- (see Section 3 below). A third solution could be regarding this name as a compound D-ybr-, where the element *Ida-, previously identified in D-tybr-, is associated with the stem *ybr- found in Ybrs-, which I propose to connect to the uppara-/o(u)npa-/o(u)ppa- group. To explain the initial [y], one might refer to rarer forms where this same stem seems to have a final -i, such as the names Oppi-^oxng/Oppi-^WTng (Pisidia). However, a caveat must be expressed: all the compounds in this group display the element uppara-/o(u)npa-/o(u)ppa- as the first member, with the notable exception of KivSu-onpag/KiS-ouppng. While this doesn't necessarily refute this interpretation conclusively, additional unambiguous parallels would be required to reinforce it.
3. An allomorph after [n] and [r]
So far, I have intentionally omitted several names in °(n)-dUb(V)r(a/i)-,16 due to their specific examination in Valerio 2015. The names concerned are listed in Table 2.
This list does not include the Lycian name group Xntabure ^ KevSepopa/KivSapupig/ Kevepopig, of which KaSopopig represents an additional variant with nasal reduction (see below, in this section) and vowel labialisation next to [b] or [p]. These forms mirror a compound Xntab-ure- (cf. Lyc. B xntaba- and Luw. /ura-/ "great"), similar to HLuw. MAGNUS.REX in structure and meaning (Yakubovich 2017: 46; Reveilhac 2018: 437).
14 The outcome of PAnat. *w in Carian is certainly not clear, but it is likely to produce [w] or [u] in certain sequences: e.g. muo- vis-à-vis Luw. /mawa-/ "four" (Adiego 2019a: 18-20, 25-26; followed by Simon in Bauer et al. 2022) and kSous, if it reflects an earlier *hantawa-, as is reasonable to assume (see Simon 2022a for a detailed discussion). The data are more numerous for Lycian: PAnat. *arowa- "freedom" > Lyc. arawa-, PAnat. *gwow- "ox" > Lyc. wawa-, PAnat. *duwV- "to place" > Lyc. tuwe-. Pace Sasseville 2021: 177, it is methodologically problematic to rely on the Lycian B form xntaba- to demonstrate the strengthening of intervocalic *w into [|3] in Lycian B given that its exact meaning and formation are unknown. It is on the basis of this argument that I reject the connection sometimes made between the name of Tyberissos and HLuw. /tuwaris(a)-/ (Réveilhac forthc. a).
15 Cautiously in Adiego (2007: 246). However, this seems to be confirmed by the group of names based on the root *bhëh2- such as Pikre-, Pik(a)rm- etc.
16 Here, U = rounded vowel going back to *u, and V = any vowel. The unorthodox notation of this sequence has a purely conventional and practical function, in order to group together different linguistic and graphic variants of the same stem.
Region (language of attestation) Personal name City Period Reference
Caria (Carian) SmSybr-?17 Halicarnassus ? C.Ha 1; Adiego 2007: 114
SoSubr- Krya ? C.Kr 1.2; Adiego 2007: 158
Caria (Greek) z£q|j£5u|3£qoç Caunos ?354 BCE I.Carie hautes terres : 90.18 (S£g>|j£SuQog); 91.6
Lycia (Greek) Eç>|jav5o|3£QLç (gen. EQ|^ttv5o|3£ÇM.oç) Phellos 2nd-1s' c. BCE SEG 53: 1696.4
Eç>|jav5u|3£QLç Limyra Imp. Wörrle 2012: 71. 4; 73.1; 3; 81.5; SEG 54: 1406.2
Sav5up£QLç Typallia Imp. Peterschen and von Luschan 1889: 149, no. 180.8; SEG 6: 624.4
n£QTC£V5u|3£QLÇ Arykanda Imp. I.Arykanda: 143.2; 59; 128.5.; 136
Pisidia (Greek) MLv5upnpaÇ (MLv5u|3r|QaÇ) Selge 1s' c. AD Bean and Mitford 1970: 201a.4
Isauria *Sav5o|3r|QaÇ (gen. Sav5o|3r|Qou) Iotape Imp. CIG: 4413d.2
Taç>Kuv5|3£Qç>aç Isaura Nea Imp. Sterrett 1888: 181.4
Cilicia Iv5o|3aç>aç Kestros 1s'-2nd c. AD Bean and Mitford 1970: 167.2; 168.2; 182
IvôopnpaÇ Lamos Imp. Bean and Mitford 1970: 193.2
*Mav5ou|3iç>aç (gen. Mav5ou|3iç>ou)18 Diokais-areia/Olba Byz. I.Westkilikien Rep: 33
*MoTov5o|3£ç>aç (gen. motov5o|3£qou) Olba 1s' c. BCE I.Cilicie: 11 A II.39
*Pwv5o|3£Qç>aç (gén. pwv5o|3£qqou) Olba 1s' c. BCE I.Cilicie: 11 A II.37
Pwv5|3£Qç>aç Olba 2nd c. BCE Heberdey and Wilhelm 1896: 155 A 11.34; 69; 84; 100
Table 2. Personal names in °(n)-dUb(V)r(a/i)-
17 In the form SmSybrs, which appears at the beginning of the inscription, the final -s can theoretically belong to the stem of the name (cf. Ybrss in C.Hy 1a) or represent an inflectional suffix. The analysis depends closely on the interpretation of the inscription in which the form occurs and on the value of the s-case itself. On these thorny issues, see especially Adiego 2007: 283-284 and Melchert 2021: 111-112.
18 Nominative Mavôoupipoç in LGPN 5b. But in view of the other occurrences in the region, one would rather expect a final -aç.
After refuting the existence of a *DUbr- stem in Carian,19 Valério (2015: 335-338) offered an alternative account. He suggested that the alleged stem °v5up£Qi-/°v5opnQa- and its variants should be interpreted as a reflection of Lyc. *ntuweri-, which is inferred from the hapax ntuweriha, found on one of the faces of the Xanthos Pillar (TL 44b.57): se=dewë : zxxaza : se ntu-weriha : ade "he made a gift to the fighters and nt.". The Lycian term is analysed as an adjective derived from a substantive *ntuweri-, which consists of the prefix nte- "in, within" and *weri- < *warri- "help" (cf. Hitt. warr- "to help", HLuw. /warriya-/), with e > u in contact with w. This etymology finds support in the Hittite expression anda warissa- "to come to one's aid". The concept of "assistance, help" (vel sim.) that emerges from this etymology fits well the context of the aforementioned Lycian passage and would not be unjustified in theophoric personal names such as EQ^av5up£Qiç/EQ^av5op£Qiç, interpreted as possessive compounds "Who has the assistance of X", where X represents a theonym like the Moon-god Arma.20 Without disputing the meaning of the Lycian lexeme, the details of its formation are of major importance for the present purpose. Ntuweriha is an action noun derived from the verb *ntuwerih-, which is structurally comparable to Hitt. (anda) warissa- (Sasseville 2023; contra Neumann 2007: 253254): on this basis, it would be necessary to reconstruct ntuweriha- (Lyc. A), *ntuwerisa- (Lyc. B), or even *ntuwarisa- (PLyc.) stems, which can hardly be aligned to the onomastic stem. Additionally, a significant phonetic challenge arises: the glide [w] is never rendered by a (p) in Greek onomastics, but rather surfaces as (u), as in Hlmmidewe—> EA|ai5aua and Tewinaza—> Teuivaaaç, or as a hiatus after u, as in Xuwata—> Koaxa (Réveilhac 2018: 360-361; Adiego 2020b: 51). Despite early fricativisation of Anatolian Greek [b] into [p] and then [v], potentially leading to the use of (p) to render the glide, it is implausible that Greek adaptations of ntuweriha- would exclusively display a (p). As a result, the association between the alleged stem °v5up£Qi-/°v5opnQa- etc. and Lyc. ntuweriha- must be dismissed.
To elucidate these personal names, a more economical explanation might be proposed, suggesting their incorporation into the group of compounds having *tubar(i)- as their second member.21 In this scenario, the initial dental may have undergone voicing assimilation following the final nasal of the first member, a phenomenon observed, e.g., in TaQKu^-piype^iç and Pw(|a)-piYQ£|aiç vs. niypa^iç. Let us begin by interpreting the simpler compounds:
— EQ|aav-5op£Qiç/EQ|aav-5up£Qiç ^ /Arman(t)-/ (Moon-god),22 thus a variant of Ep|aa-
TOpOQLÇ;
— Taç>Kuv-5p£QQaç ^ /Tarhun(t)-/ (Storm-god), with syncope of the initial vowel in the second member (as seen in Pwv5p£Qpaç below);
— Pwv-5op£QQaç/Pwv-5p£QQaç ^ /Run(ta)-/ (Protective Stag-god).
The devil's advocate could argue that the proposed segmentation in these theophoric compounds is unverifiable, given that all three theonymic stems end in -nt-. The case of n£QTC£v5up£Qiç might be, however, less ambiguous, if the first member indeed reflects Lyc. (A and B) perepn "further(more)" (perhaps to be understood as divine epithet), as in n£QTC£vnviç (vis-à-vis N^viç ^ nëne/i-) and n£Qrc£vvuv£|aiç (vis-à-vis Ov£|aiç).
This hypothesis could also shed light on the formation of other anthroponymic compounds, although the identification of their first member is not always straightforward. As for
19 See already the embarrassment of Schurr (2013: 28) to account for the formation of several Carian names.
20 This was the analysis adopted in Reveilhac 2018: 490-491. Compare with the numerous Anatolian compounds of the type X-muwa- "Who has the might of X" (Houwink ten Cate 1961: 166-169; Laroche 1966: 322-324; Melchert 2013: 33-34).
21 The Cilician data in -ag suggest that *tubara- replaced *tubar(i)- in Cilicia.
22 Another possibility, suggested to me by Ilya Yakubovich, would be to see in the first member the cognate of Car. armon "interpreter", then perhaps used as a divine epithet.
MLvSupnQaç and its likely variant MavSoupipaç, the first member can receive two interpretations. One interpretation would be to regard Miv- and Mav- as variants of M^v-, reflecting the name of the Moon-god Men, with one variant displaying iotacism — a common feature in the imperial period in Anatolia -23 and the other retaining the older vocalism (Sittig 1911: 153-155; Masson 1990: 327-328). The worship of Men is well-documented in certain Anatolian regions, particularly in Roman Pisidia, where MivSupnQaç24 is attested. If this interpretation holds, MivSupnQaç/MavSoupipaç could stem from a secondary adaptation of the group of AQ|aaT0p0Qiç/AQ|aav50p0Qiç/EQ|aav5up£Qiç, substituting Men's name for that of Arma, who is also a Moon-god. An alternate analysis would be to see the first member as a *mVnd- element, supported by other anthroponyms: Lyc. Mntete(/i)-, MevSiç (LGPN 5b; Lycia, Olympus, 3rd c. AD)25, Miv5r)ç (see KPN § 920)26, MevSeaiç (LGPN 5b; Cilicia, Kestros, 1st-2nd c. AD), or MlvSqlwv (LGPN 5b; Milyas, Soklai, 4th-5th c. AD). The stem of these names may be tentatively related to Lyc. minte/i- (cf. Gk. |aiv5iç, laevSixnç), denoting a local authority sometimes linked to deities (cf. mahai mintehi "of the gods of the minte/i-"; TL 58.5), or alternatively to Lyc. mëte-"damage, harm" (cf. Hitt.-Luw. mantalla/i- "poisonous, slanderous", Lyd. mëtr(i)-/mëtl(i)- "harm").
The first element of EavSupepiç and SavSop^Qaç has been associated by Neumann (2007: 144-145) with Lyc. B xzzata- (TL 44d, 44), potentially underlying the Greek name of the city of SàvGoç (otherwise called arnna- in Lycian A), although this remains uncertain. The same stem appears in the Pamphylian anthroponym EavSapoiCaç (LGPN 5c; Kibyra Mikra, imp.), which can be segmented as EavSa-poiCaç thanks to the comparison with PoiCaç (Cilicia)/PoiCoç (Lycia)/PwlClç (Pisidia).27 Based on this, it is tempting to place the morpheme boundary between Sav5-, on the one hand, and -up£Qiç/-opnQaç, on the other. However, factoring in EQ|aav-5op£Qiç/EQ|aav-5up£Qiç vis-à-vis Ep^avSaç, TaQKuv-Speppaç vis-à-vis TaQKovSaioç, and Pwv-5op£QQaç/Pwv-5p£QQaç vis-à-vis PwvSaç, it is conceivable that the first element in Sav-5up£Qiç/Sav-5opnQaç vis-à-vis EavSa-poiCaç shows variation */ksan(t)-/ ~ */ksanta-/. This stem might tentively correspond to HLuw. /hasa-/ "abundance": it would reflect a derived possessive adjective *has-ant(i)- "having abundance", to be linked to the Palaic adjective hasawant-, used as a divine epithet of the Hattian deity Kamama (van Gessel 1998: 216),28 who belongs to the group of the protective dKAL deities (Laroche 1973: 85-86). Also belonging to the dKAL group, the Luwian god Runtiya is also presumably hidden behind PoiCaç: in the compound SavSa-poiCaç, the Luwian protective deity too appears to be associated with abundance.
The names from Cilicia IvSopapaç and IvSop^Qaç contain a first element that can be linked to several simple names from the same region, namely IvSaç, IvS^ç, or even IvSouç (see LGPN 5b: 213-214). These names have a root that resembles the Luwian adverb inta "below".29 This adverb is known in Cuneiform Luwian as an epithet of the Sun-god Tiwad, invoked in association with his celestial counterpart [sarr]i Tiwat-, meaning "Tiwad above"
23 See also the Pamphylian onomastic element Meiva-, MLva- representing the name of Men (Brixhe 1976: 29 and 136).
24 On the cult of Men in Pisidia, see especially Labarre 2010. Note that the divine name may have been converted into a personal name in Pisidian, as indicated by several attestations of Mr|VL- in the Pisidian corpus.
25 Despite the doubts expressed by Zgusta (KPN: 317, fn. 219), there is no need to resort to a Thracian interpretation of this name (as a variant of the well-known BevSig).
26 Procopius of Caesarea, De bellis, 7.36.26, as the name of an Isaurian.
27 Perhaps SavSa, attested in an Athenian inscription (IG II2 7926, 2nd c. BCE) as the name of a Galatian from Ancyra, should also be added, but this remains uncertain (KPN § 1060).
28 See Sasseville 2022, for a discussion about the Palaic word, with references.
29 I thank Ilya Yakubovich, who suggested this connection.
(KUB 32.10+ obv. 12'; Yakubovich and Mouton 2023, vol. 1: 209). This duality between a solar deity "above" and a solar deity "below" is likely explained in KBo 29.3+ ii 25-26, where the Sun-god (of Heaven) is associated with the living, while the Sun-goddess of the Earth is associated with the dead (Mouton 2023: 128). The root Iv5- likely reflects the epithet "below", which has replaced the divine onomastic sequence /inda Tiwad/. Therefore, IvSopapaç/ IvSopnQaç is based on a formula meaning "(Tiwad) Below tubar(i)-".
In the case of Z£Q|a£5up£Qoç, one has to assume the disappearance of the preconsonantal nasal that triggered the voicing of [t] into [d]. Such an evolution is well documented, particularly in Greek adaptations of Anatolian names, such as K£§npnç vis-à-vis K£v5npnç/K£v5£pnç (cf. Lyc. B xntaba-). The plausible first element of this compound might be traced in the name Z£Q|aouv5iç (Lycia),30 which may in turn result from the anthroponymic conversion of *z(e)rmmnt(i)- < *zarmant(i)-. As per Neumann (1976: 139-141), such a form could be based on the lexeme *zarma- < zalma- "protection",31 identified in the Cilician names TpoKO-CaQ^aç, Pw-CaQpaç and Ia-C«Q|aaç (cf. Luwian personal names of the type /Tarhu-zalma-/).32 *Z(e)rmmnt(i)- might therefore represent a possessive adjective in -nt(i)- meaning "having protection", unless it is the former participle of a verbal stem.33 In the latter scenario, considering the close association between the element *zarma- and divine names, *z(e)rmmnt(i)- might correspond to a divine epithet meaning "protecting".
Carian documents also provide two names of more ambiguous interpretation: Soôubr- and Smôybr- (if this interpretation is favored over that of a stem Smôybrs-),34 where the grapheme (5) indicates the biphonemic sequence /nd/, like in the Storm-god's name Trqô/Trquôe (cf. Luw. /Tarhunt-/, Lyc. A Trqqnt-; see Adiego 2020a: 105). Soôubr- might potentially be a compound with the name of the War-god Sanda, widespread in southern Anatolia,35 as first element. A specific relative chronology of sound chages is, however, required to justify the initial fricative [s] and the vocalism of the first member: */Sanda tubar(i)-/ "Sanda tubar(i)-" > */Snd(a) tubr-/ (syncope and apocope) > */Sn(d)tubr-/ (palatalization of the sibilant)36 > /Sndubr-/ [Sondubr-] > /Sondubr-/ spelled SoSubr-. The final stage would correspond to the phonologisation of the "excrescent" vowel derived from the syllabic nasal, a process recently highlighted by Adiego (2020a: 115-118), elucidating variants such as PnusoA/PunwsoAs (^ novuawAAoç).
Regarding Smôybr-, it has previously been linked to the name E£|a£v5noiç found at Istlada (Lycia), rightly interpreted as derived from the divine name Eo|a£v5iç/Eou|a£v5iç (Schurr 2013: 28). The latter is documented on various votive stelae from Central Lycia, indicating one
30 Pace Valerio (2015: 335), who regards the pair Zep|je5upepog/Zep|jouv5Lg as a hint that there was a stem Zep|j-, possibly extended into Zep^e5-, without however succeeding in offering a comprehensive explanation. Adiego (2019b: 147, fn. 2) prefers not to include the name Zep^ouv5ig in his study of names in -zarma-, -sarma-, -Cap^ag.
31 The meaning of zalma- is inferred from the graphic alternation attested for the personal name Iyarrazalma-/ Iyarra-PAP- (Laroche 1966: nos. 430 and 433), since the sumerogram PAP/PAB is used in Hittite texts as the equivalent of pahs- "to protect".
32 Nevertheless, Adiego (2019b: 157-159) prefers to remain cautious about the origin of the stem -Cap^ag, which could also reflect the Luwian contracted variant Sarma- of the divine name Sarruma-, following the analysis of Houwink ten Cate 1961: 134-136).
33 Indeed, it cannot be ruled out that the element zalma- is verbal, in which case all the names X-zalma- with X = divine name would be Satznamen meaning "X, protect (him)!". Such an explanation would then shed light on forms such as HLuw. /Tarhu-zalma-/, TpoKo-Cap^ag, or Pw-Cap^ag, whose first element would correspond to the vocatives /Tarhu(n)/ and /Ru(n)/ (see Adiego 2022: 81-82).
34 Cf. fn. 17 above.
35 On the place of this divine name in southern Anatolian onomastics, see Houwink ten Cate 1961: 136-137.
36 Compare with sar- < *sri- (Adiego 2007: 261).
or several places of worship dedicated to this deity. It has been suggested elsewhere (Reveilhac forthc. c) that the divine name Eo|aev5ic;/Eou|aev5ic; should be seen as an ancient epiclesis corresponding to CLuw. zammant(i)-/zamnant(i)-, also used as a Glossenkeilwort in Hittite contexts. Literally meaning "having the zam(ma)n-", this adjective, mostly employed in ritual contexts, could convey the active meaning of "damaging" or the passive meaning of "damaged", both possibly referring to the supernatural impact on the patient in purification rituals. The name Ee|aev5noig likely denotes a possessive derivative in *-assa/i-, implying "Belonging to Soumendis / the zammant(i)- deity", 37 while SmSybr- might reflect the sequence *Zammant(i)- tubar(i)-.
Nevertheless, it is essential to exercise caution in the last two cases. If a second member °y/ybr- is aknowledged in Car. Dybr- (see Section 2 above), then it could also potentially be found in SoSubr- and SmSybr-. These two names, therefore, remain equivocal, although, as noted earlier, the prevalence of compounds with uppara-/o(u)npa-/o(u)ppa- as the first member makes the interpretation with *tubar(i)- more plausible in my perspective.
Moreover, the voicing of [t] into [d] extends beyond the position after [n(d)] and likely encompasses the position after [r], as observed in Lycian, evident in examples like Natr-bbijeme/i-(^ AtcoAAo-Sotoc;) compared to Mahane-pi[jeme/i]- ^ Mava-ra|a[i]g and pije- "to give" (Reveilhac 2021: 566). This pattern appears in variants of the same name, as depicted in Table 3 below.
Egypt (Carian) Ardybyr- Memphis ?6lh-5lh c. BCE E.Me 52; Adiego 2007: 72
Caria (Greek) Ap5u|3epoç Halicarnassus 5 th—4 th c. BCE SEG 43: 713C.2; 9; 36
Table 3. Two variants of a personal name in °r-dybVra/i-
In the Carian name Ardybyr- ^ Ap5up£Qoç, the first member can also be found in A[r]bikarm- vis-à-vis Pik(a)rm-. This element is ambiguous and might correspond to: 1) an adverbial element meaning "high" (cf. Luw. /aru-/);38 2) the Carian appellative eri- "companion" and Lycian °eri- "id. (?)" (cf. Hitt. ara-);39 3) a divine name, reflecting dAra/i- found in the Hittite documentation (KBo 52.56 ii 14; cf. van Gessel 1998: 45).
4. The stem *tubar(i)-as first member in compounds
Houwink ten Cate (1961: 159) proposed an interpretation of Tp£Qaonxaç, Tp£Qn|awoiç, Pwv5p£QQaç, and TaQKuv5p£QQaç, presenting them as compound names containing an element akin to Luw. *tapara- "ruler". Nevertheless, Valério (2015: 342) rightfully dismissed this interpretation, based on the comparison with Luw. *uppara- which became 0(u)npa- in Hellenistic forms, implying that one would expect *tupara- > **TVppa-. Valério suggested viewing the first member of Tp£Qaonxaç and Tp£Qn|awoiç as a reflection of a Proto-Anatolian element
37 Like in the Lyc. B personal name Masasa/i- "Belonging to the god(s)".
38 See also Houwink ten Cate (1961: 172-173), where the adverbial element er(i)- is identified in several names, although some of them actually display another adverbial element, hri, such as Ap^oag/Ep^oag, reflecting Hrmmuwe- and having Zepi|juac; as a Lycian B avatar (Reveilhac 2018: 486).
39 For the Carian form, see Simon 2016-2017. The Lycian °eri-, which is not assured, could be the second member of a possible compound tidere/i- "collacteus", with in the first member *tide- "teat" (see Melchert 2004: 66; and Neumann 2007: 360-361, with discussion and references). Adiego (2022: 83) suggests interpreting the Cilician names TapKuapig and TpoKoapig as Satznamen reflecting a clause with a vocative *Tarhu, aris "Tarhunt, (be) his comrade!", but it could also be *Tarhu, ari "Tarhunt, raise (him)!".
*d(u)wara- "long-lasting" (< PIE * dwh-ro- "long") or a Luwian adjective *tuwala- (cf. Hitt. tuwala-"distant") > *t(u)wara- (Valerio 2015: 344-345).40 While this proposal is ingenious, its foundation is tenuous due to the absence of additional lexical or anthroponymic data. Moreover, the semantic aspect is not as apparent as suggested by the author. The proposed translations "Man of long-lasting might" and "Long-lasting man" for Tpepnpwaig and Tp^paonxag respectively are not conclusive, since compounds with a second member °muwa- and °aonxag41 are usually theophoric, which suggests the need for a divine element (name or epithet) as the first element.
The association of two Carian forms with this small group of names from Cilicia, Tbridbd-in Memphis (E.Me 42, 2; Adiego 2007: 66), and *Topopopag (gen. Topopopo)42 from a Greek inscription in Halicarnassus (5th-4th c. BCE), suggests that the stem *tubar(i)- might serve as the first member of a compound. Although the integration of the first is provisional until the second virtual member is positively identified,43 *Topopopag can be interpreted as derived from *Tubur-ura/i- meaning "Great(est) tubar(i)-" vel sim. This formation finds parallels,44 indicating the plausibility of the stem *tubar(i)- serving as a first member of a compound.
If one assumes a syncope of the vowel in the first syllable of the stem,45 as observed in Eayyo-xp^pig, Pwv-Speppag (next to Pwv-Sopeppag), and Pis. T[p]epi- (see Section 2 above), then the stem Tpapa-/Tpepn-/Tpepa- documented in Cilicia (and possibly in Car. Tbr°) might be a variant of tubar(i)-. This phenomenon, quite widespread in names from Cilicia (as seen in Pwv-Speppag), could either be a pan-Luwian or a regional (post-Luwian) feature. Consequently, the interpretation of these anthroponyms could be as follows:
— Tpapa-^oxag < */Tubara-muwatta-/ "Tubar(i)- (of) Conquest" or "Having the Might of tubar(i)-" (cf. HLuw. /muwatta-/);
— Tp£pn-|awoig < */Tubara-muwassa/i-/ "Mighty tubar(i)-" or "Tubar(i)- of Might" (cf. HLuw. /muwassa/i-/);
— Tpep-ao^xag < */Tubara-assatta-/ "Tubar(i)- (of) Peace" or "Having the peace of Tubar(i)-" (Luw. in Hittite transmission *assatta-).46
5. Synthesis on the compounds with *tubar(i)-
So far, I have endeavoured to demonstrate the existence of the stem *tubar(i)- in southern Anatolian onomastics. This stem, which has sometimes undergone phonetic variations, clearly appears in compounds as a second member after a first member ending in a vowel (e.g. Epjaa-
40 The first hypothesis is suggested with caution, however, since "no direct descendant of P.-Anat. *dwara-is attested among the Anatolian languages".
41 Contrary to what has long been claimed, names in °oT|Tag do not correspond to Luwian compounds in /°zid(i)-/, but are actually in °aor|xag. See fn. 46 below.
42 Better than *Topopopog (SEG 43.713A; cf. LGPN 5b).
43 The second member could be related to the Lycian place-name Idebessos (ISepr|oorog, ISpr|oorog, ESepr|oor6g; see KON § 361).
44 Cf. Lyc. Xntabure—> KevSepopa/KivSapupig/Kevepopig. See Section 3 above.
45 For syncope in the languages of the Luwic subgroup, see Melchert 1994: 275-276 (Luwian, more limited than Lycian) and 318-321 (Lycian); Hajna! 1995: 175-188 (Lycian); Adiego 2007: 241-242 (Carian).
46 Contra Houwink ten Cate (1961:171-172), the names in °or|xag (Cilicia) or °aaxr|g (Pisidia) from Greek sources cannot correspond to Luwian compounds with /zid(i)-/ "man" as second members for phonetic reasons (nature of the dental, vocalism of the stem). On this crucial issue, see now Reveilhac forthc. b, where a segmentation °aor|Tag/°acraTr|g/°aouTag is proposed, and this element is analysed as the abstract noun "peace" (Luw. *assatta-, Lyc. B asata-, Lyc. A ahata-) used as a divine attribute.
Topopig). I have argued that this same stem displays an allomorph with initial [d] when it comes after a first member ending in [n] (e.g. Ep|aav-Sopepig/Ep|aav-Supepig) or [r] (e.g. Ar-dybyr-/Ap-Supepog). I have also suggested identifying *tubar(i)- as the first member of (rarer) compounds, in full form in Topop-opag (Caria), but with syncope in a few Cilician names (e.g. Tpapa-^oxag). If one accepts this argumentation, it is possible to make use combinatorial analysis in order to interpret the meaning of the stem under discussion. Below one finds a summary of the compounds with *tubar(i)-, ordered according to the associated stem type.
Personal name Divine name Deity's function
Ar-dybyr-Ap-Sußepog Ara/i-? (or *ar- "super") ?
D-tybr- Ida- ?
Ep|aa-To|3oQLg Eg>|jav-So|3£QLg Eg>|jav-Su|3£QLg Arma(nt)- Moon-god
Pwv-Soßeppag Pwv-Sßeppag Runt(iy)a- Protective Stag-god
Zayyo-xßnpL? *Sanhu- (cf. dSanhupiya-)? ?
?SoSubr- Sanda- War-god
TapKuv-Sßeppag Tarhu(nt)- Storm-god
Table 4. Type 1, X-tubar(i)-, where X = divine name ^ "(Deity) tubar(i)-"
Personal name Divine epithet Epithet's class and meaning Parallels
Zep^e-Sußepog *zarmant(i)-? Participle/adjective: "protecting"? Zep|jouvSig
Iv-So|3apag Iv-So|3r|9a? Luw. /inda/ Adverb: "below" IvSag IvSr|g IvSoug
Ksa-tybr- ? ? 'E^a-|jur|g E^a-|3oag?
Kud-tubr- Luw. */huda-/, Lyc. *xuda-? Substantive: "haste, swiftness" KouSeig?
Mav-Sou|3upag Miv-Su|3r|9a? Lyc. minte/i-or mete-? Substantive: (official institution) ou "harm, damage" Lyc. Mntete(/i)-MevSig MivSr|g MevSecxg
Moxov-Soßepag ? ? MouxvoLT^g? (vs. Mo[K]-oLT^g)
Sav-Sußepig Sav-5oßnPa? *hasant(i)-? Adjective: "having abundance"? SavSa-poiCag
nepnev-SußepLg Lyc. perepn Adverb: "further" nepnevr|vig
nepxa-ToußapLg CLuw. paratt(a)- Substantive: "impurity" napTaaLg Parattassa/i-Place-name napTariccrog
?SmSybr- CLuw. zammant(i)- Adjective: "damaging; damaged" Divine name Zo|jevSig/Eou|jevSig Ze|jevSr|cng
Table 5. Type 2, X-tubar(i)-, where X = divine epithet ^ "([Deity +] divine epithet) tubar(i)-"
Personal name Divine epithet Epithet's class and meaning Parallels
T|3eQ-aar|Taç Luw. *assatta-, Lyc. B asata-, Lyc. A ahata- Substantive: "peace" MLQ-aaqxag OpQ-aaqxag/ OnQ-aaaxnç
Tbr-idbS- ? ? Place-name I5e|3r|GGoç?
T|3aç>a-|joTaç *muwatta- Substantive: "conquest" Oupa-^ouxaç AQGi-|jouTaç 0|3ç>a-|jouTaç
Tpe^n-^w^LÇ *muwassa/i- Adjective: "mighty" or "of might" nLYQa-|JOUGLÇ OuTCQa-|JWGLÇ
To|3oQ-oç>aç *ura/i- Adjective: "great" Tiwata-ura-
Table 6. Type 3, tubar(i)-X, where X = divine epithet ^ "([Deity +] divine epithet) tubar(i)-" (same as Type 2) or "[Deity+] tubar(i)- (divine epithet)"
It quickly emerges that this stem is associated with various theonyms, but also with other lexemes belonging to different grammatical classes (nouns, adjectives or adverbs, when their identification is assured), used as likely divine epithets. On this basis, two hypotheses are con-ceivable:47
1) *Tubar(i)- is a theonym, as compounds of types 2 and 3 seem to suggest, and Toupepig, indirectly attested as the name of a local hero or deity in Lycia, would reflect this theonym. In this case, type 1 compounds (theonym + *tubar(i)-) would be dvandvas combining two divine names, in line with what we already find in Anatolian onomas-tics:48 in cuneiform sources, e.g. dSIN-dU (Laroche 1966: no. 138) "Moon-god (and) Storm-god"; in Hieroglyphic Luwian, e.g. /Sanda-tiwad(i)-/ (TUNP 1.1; Hawkins 2000: 155) "War-god (and) Sun-god"; and in alphabetical documentation, e.g. Ap^a-pwvCag (LGPN 5b), "Moon-god (and) Protective Stag-god". However, apart from Stephanus of Byzantium's late and indirect attestation, there is no other testimony of this word as a theonym in the various corpora, suggesting rather a secondary use as a divine name. Furthermore, the only avatar of this stem in the lexicon of a Luwic language, that is Lyc. B tuburiz, is in all likelihood an appellative used in the plural (see Section 7 below).
2) *Tubar(i)- is originally a divine epithet, associated with various deities in compounds of type 1. In type 2, therefore, the divine epithet used as first element would be a hypostasis, i.e., it would replace a divine onomastic sequence theonym + epithet. A similar process is frequently attested elsewhere, cf. pihassa/i- "luminous; (subst.) lightning", first used as a characteristic of the Storm-god (van Gessel 1998: 791) before constituting a hypostasis (e.g. Pihassa-muwa- "Having the Might of [the Storm-god of] Lightning") or Lyc. *xttbile/i- "destroyer" (cf. xtta(i)- "to harm, to damage"), associated with the Moon-god in anthroponymic compounds (e.g. Ep|ae-KTipiAig/Ep|aa-KxipiAog/Ep|aa-KxupeAig), to finally replace the whole divine onomastic formula in the personal name KxipiAag "(Moon-god) The Destroyer" (Adiego 2022: 84-86). In type 3, finally, the elements are reversed in relation to type 2, since *tubar(i)- is used as first member, in association with another epithet: either the meaning of such a compound is equivalent
47 The claim that the stem tubar(i)- refers to a deity finds independent support in the fact that it serves as the base for the place name Tyberissos, according to a mode of formation that finds parallels in Anatolia: see Réveilhac forthc. a.
48 On these personal names, see in particular Laroche 1966: 282; Zehnder 2010: 58-59; Melchert 2013: 36; Ya-kubovich 2013: 103-104.
to type 2 or, which seems more likely to me, *tubar(i)- is to be interpreted here as a hypostasis and stands in for a formula theonym + *tubar(i)-. This would then constitute the intermediate stage between its use as an epithet and its use as a divine name, which is attested later in Lycia in the form Toupepig.
6. A warlike epithet?
The only attestation of the stem under discussion in the lexicon comes from Lycian B, in the form tuburiz, unfortunately mentioned in passages that are far from clear. As a result, it is necessary first to consider the semantics of the anthroponymic compounds listed in the previous section before attempting an etymological analysis of *tubar(i)-.
The association of *tubar(i)- with elements of such diverse nature suggests that it must not be a specific divine epithet: the underlying lexeme rather represents some kind of generic divine title or designates a quality compatible with different deities. In this respect, the previously listed names can be compared with the numerous Luwian and Lycian compounds having nan(n)(i)- "lord, leader" as second member,49 particularly as several of these compounds have an identical first member: Ep|aa-Topopig/Ep|aav-Sopepig/Ep|aav-Supepig vs. Hier. /Arma-nan(n)(i)-/, Lyc. Ermmenene/i—> Epiae-vqvig etc.; TapKuv-Speppag vs. Cun. mdU-SES, Hier. TONITRUS-FRATER2 /Tarhu(nta)-nan(n)(i)-/; neprcev-Supepig vs. neprce-v^vig (^ Lyc. *Perepn-nene/i-).
In order to precisely define the meaning to *tubar(i)-, it is beneficial to compare it with the matching lexical items and attempt a morphological analysis. The relevance of a Luwic lexical group containing the seme [strike] becomes immediately apparent:
— HLuw. /tub(a)-/ "to strike down"; /tub(a)i-/ "to strike repeatedly"; /tubas-/ or /tubassa-/ "battle";
— CLuw. dupi-/dupai- "to strike, touch, punish";
— Lyc. A and B tub(e)i- "to strike down"; tube- "attack, battle (?)".
These words, which are based on a Proto-Anatolian stem going back to the PIE root *(s)teup- "to strike", reflect two types of semantic development (see Sasseville et al. 2023):
1) [STRIKE] > [PUNISH; destroy]. This semantic value is attested in Luwian and Lycian in the context of curse formulae to designate the divine punishment to which any dese-crator is exposed. Thus, in Hieroglyphic Luwian:
KULULU 5 (= Hawkins 2000, p. 485-486) :
§ 4 za-ti-za-pa DOMUS-na-za kwa/i-sa a+ra/i-na CUM-ni i-zi-i-ri+i
[•••]
§ 8 wa/i-tu-ta za-zi DEUS-ni-zi wa/i-la "PES"-tu
§ 9 wa/i-na tu-pa-tu
§ 10 wa/i-tu-u [a-la/i-ma-za ARHA] DELERE-nu-tu-u
"(he) who will cause ARA(NA) to these houses, [...] for him may these gods come mortally! Let them strike him down, let them destroy his name— "
49 On this noun, see the discussion in Oreshko 2014. Contra Reveilhac (2018: 479-480), the translation by "lord, leader" on the basis of the connection with CLuw. nana- "to lead" (compare Latin dux) seems now plausible to me. It is also implicitly adopted in Adiego (2019b: 158). Nevertheless, the precise connection between this noun and the name of the "brother" poses a challenging issue that cannot be addressed within the scope of this present paper.
In this passage, the formula wa/i-na tu-pa-tu /w=an tubantu/ "let them strike him" is integrated into a network of divine threats. Similar formulae can also be found in Lycian epitaphs. Here are a few examples:
TL 72.2+3-4 (bilingual):
m=ëne mahai : tubeiti : nelez[i]
"the gods of the agora will destroy him"
OL 0£OL I ànOÀ£G£LaV : OL àYOQaLOL
"may the gods of the agora destroy (him)"
TL 80.3-4 :
m=ene I [Trq]as tubidi se Malija hrixuwama
"Trqqnt and Maliya the Supervisor will destroy him"
TL 93.3 :
m=ëne Trqas tubidi se muhai huwedri : "Trqqnt and all the gods will destroy him"
2) [strike] > [attack; fight]. This meaning is admittedly less certain than the previous one, since it is generally inferred from the warlike context for both Lyc. A tube- (TL 29.10; TL 44b.56) and HLuw. tù-pa-sa6-ti /tubas(s)adi/, which is a hapax:
TOPADA § 17-18 (= Hawkins 2000: 451-461) :
wa/i-mu â-mi-sa4 DOMINUS-ni-sa (DEUS)TONITRUS-zi/a-sa8 (DEUS)SARMA-sa8 (DEUS)L198-sa6 (DEUS)BOS.L206.PANIS-sa«-ha PRAE-na L179-ia-tax Iwa/i-mi-tax tù-pa-sa6-ti wa/i^-su-ha "my lord Tarhunt, Sarruma, Ea and (god) X ran before me, and I succeeded by battle"
The form of the noun here is ambiguous, since it could theoretically correspond to a neutral s-stem /tubas-/ derived from the verb or to an adjective /tubassa/i-/ derived from of a PAnat. stem *tub(a)- "strike, attack".
To this little dossier one can add ToupaGGiç,50 a Carian divine name of Greek transmission, which could originally be an epithet equivalent to the Luwian form. If the latter is indeed to be assigned to the stem /tubassa/i-/, as argued in Simon 2016, who believes that ToupaGGiç can be interpreted as "(God) of Smiting" or "(God) of Fighting". A similar semantic ambiguity exists for a related Luwian term known from Hittite sources, namely the abstract *dupattar-(gen. adj. acc. dupattanassin), only used as an epithet of the Storm-god: does the epithet designate a god who strikes with lightning, one who fights, or even one who protects from smiting?51 All are theoretically possible. Indeed, on the one hand, the Annals of Mursili II relate how the Storm-god defeated the king's enemy by striking the land of Arzawa with his lightning;52 on the other, it is common for gods to assist their protégé in battle. This applies not only to the Storm-god, but to all protective deities, as is shown clearly in the Annals of Hattusili I,
50 Gen. Tou|3aamo[g]: Kizil et al. 2015: 379, 389.15.
51 See the full discussion, with references, in Simon 2022b.
52 CTH 61.I : mahhan=ma iyahhat nu GIM-an INA URSAGLawasa ärhun nu=za dU NIR.GAl EN-YA parä handandätar tekussanut nu GlSkalmisanan siyäit nu GlSkalmisanan ammel KARASöLA-YA uskit KUR URUArzauwa=y(a)=an uskit nu GlSkalmisanas pait nu KUR URUArzauwa GUL-ahta SA mÜhha-LU-ya URUApäsan URU-an GUL-ahta (KBo 3.4.1519; Goetze 1933: 46-47) "Wie ich aber marschierte, wie ich da zum Gebirge Lawasa gelangte, da zeigte der stolze Wettergott, mein Herr, seine göttliche Macht, und einen Donnerkeil schmetterte er hin. Und den Donnerkeil sah mein Heer. Auch das Land Arzawa sah ihn, und der Donnerkeil ging hin und traf das Land Arzawa. Auch des Uhha-LU-is Stadt Apasa traf er." (transl. A. Goetze).
where the Sun-goddess of Arinna is reported to have "run before" (p[é]ran huwais) the king on the battlefield, thus ensuring the surrender of the enemies.53 The common motif of the deity "running before" on the battlefield is also found in the hieroglyphic Luwian inscription from TOPADA mentioned above.
In the case of tubar(i)-, the semantics of the stem *tub(i)- is also ambiguous, but can possibly be clarified by its formation. There are two possibilities:
1) either an adjectival derivation in -ra/i- < *-ro- on a reconstructed stem *tuba- "strike" or "battle", but this type of denominal formation seems unproductive in the synchrony of the Anatolian languages, although it does exist;54
2) or a compound *tub-ar(i)- "companion of tuba-", with the second member *ar(i)-, virtual Luwian equivalent of Hitt. ara- c. "friend, companion". Such a formation would then be comparable to Lyc. A tidere/i- "collacteus/-a, foster brother/sister", whith *tide- "teat, breast" (cf. Hitt. tëta(n)-, CLuw. titan- < PIE *dhehi-i- "to suckle")55 as first member. For such a compound, the meaning "battle, fight" is then much more relevant than "strike", so *tub-ar(i)- would designate the "battle companion, ally (au^^axoç), comrade-in-arms".
From a semantic perspective, this second hypothesis sheds light on the numerous uses of the epiclesis alongside various deities, each representing distinct entities tasked with accompanying a protégé in warfare. It has been observed that the intercession of tutelary deities is mentioned several times in Hittite and Luwian texts. This suggests that anthroponymic compounds containing *tubar(i)- resonate with this divine function, encompassing both a martial and protective role.
A typological parallel is possible with the Greek texts, where warlike epicleses are not exclusive to Athena or Ares, as illustrated by some inscriptions. For example, the votive inscription from the Cyrenaean Paniskos I.Pan du desert: 3 (Wadi Bir el-Aïn, Egypt, Ptolemaic period) displays a coherent network of warlike epicleses applied to several deities (Mother of Gods, Ares, Zeus, and Pan), who are asked by the dedicator to take part in the military expedition he is about to lead (see Bonnet & Galoppin 2021: 7-11).
It is not rare to find military or political titles used as divine epithets in Anatolian and Near Eastern languages. In some instances, these titles even replaced the original theonym.56 The transformation of the epithet tubar(i)- into a divine name is suggested by anthroponymic compounds of type 3 (as discussed in section 5 above), where the stem is used as the first
53 CTH 4.II : LUGAL.GAL Tabarnas NARAM dUTU URUArinna nu=mu=za=kan a[nda ginuwas=s]a[s hal]is[t]a? nu=mu kessa[rta harta?] n=as=mu ME-ya p[e]ran huwais nu INA URUNinassa ME-ya paun nu=mu mahhan LUMES URUNi-nassa menahhanda auer nu <KA.GAL№A> EGIR-pa hessir (KBo 10.2 i 27-32; Imparati 1965: 46-47; Melchert 1978: 8-9) "The Great King, the Tabarna, beloved of the Sun-goddess of Arinna — she placed him on her lap — she took me by the hand and ran before me in battle. In the city of Ninassa I went to battle, and when the men of Ninassa saw me facing them, they reopened the city gates."
54 Note the possible existence of a *-ro- derivation in the Hittite hapax tupran (c./n. acc.) " Herzklopfen (?)" (Oettinger 2001: 461, fn. 15), which could reflect a primary PIE derivation *(s)tup-ro- "striking" (cf. lat. stuprum "disgrace, violation etc.").
55 A similar analysis is cautiously proposed by Billing 2023 for the Lyc. A hapax tuweri (TL 84.5), whose meaning is unclear. According to the author of the eDiAna entry — and quite convincingly, in my opinion — it could be a compound *tuw-eri- litt. "companion of the offering" (cf. Lyc. B tuwe/i- "offering" and °ere/i- "companion", perhaps in tidere/i-).
56 The reader can find a survey of well-known examples (such as Bel "Lord" first associated with the Babylonian god Marduk, before he became the object of a specific cult until Late Antiquity) in Yakubovich 2021: 239-240, which also convincingly suggests that the theonym 'EvudAiog constitutes a borrowing from a Lydian epiclesis enwaAa- to be reconciled with the Hittite (probably borrowed from Luwian) annawali-/annauli- "co-ranked, peer, colleague".
member along with a divine epithet. Additionally, indirect evidence for this evolution comes from the mention of Toupepig as a local hero or deity in Lycia by Stephanus of Byzantium.
7. Lyc. B tuburiz
In the Lycian B lexicon, a direct reflection of *tubar(i)- is evidenced by the term tuburiz, occurring three times in the poem from the Xanthos Pillar (TL 44c.53; c.57; d.31-32), engraved on the funerary monument of a ruler of Xanthos (TL 44c.32-d.71). This text, flanked by a lengthy Lycian A inscription (TL 44b.1-c.19) and a Greek epigram (TL 44c.20-31), remains the most cryptic one among the three inscriptions due to the scarce attestation of Lycian B: beyond it, only one other text, the poem of Antiphellos (TL 55), is known in this language. Despite multiple studies dedicated to TL 44c.32-d.71, the precise content of this text remains ambiguous, making it unrealistic to use it alone as definitive evidence for the proposed hypothesis. However, it is crucial to ensure that the context in which tuburiz appears is compatible with the proposed meaning.
Prior research has primarily associated tuburiz with the toponym Tuburehe(/i)-/ Tupepiaaog, with limited advance beyond this link, as noted in Neumann 2007: 385.57 Among recent contributions, there exist two conflicting interpretations. Schurr (2016: 174, 190-191), inspired by a suggestion by Zimmermann (2003: 293), draws a parallel between the poetic formula trmmiliz : tbiple : trpple : tuburiz (TL 44c.51-52) and the relationship between the mythical figures Toupepig and Tepjaepig indirectly reported by Stephanus of Byzantium. This connection leads Schurr to conclude that tubur(i)- refers to Tyberissos as an ethnonym. Nevertheless, the association between trmmiliz and Tep^epig, forming the basis of this parallel, is approximate and, thus, fragile.
Offering an alternative perspective, Sasseville 2021 proposes to identify a lexical network in the poem associated with offerings made by the ruler to appease enraged deities. By considering context and etymological parallels, Sasseville suggests three offerings: waxssa-"bread" (cf. Hitt. NINDAwagessar "bread bites"), mur(i)- "bunch" (cf. Hitt. muri- "bunch, grape") and tubur(i)- "vine" (cf. HLuw. /tuwaris-/ "vineyard"). While this proposal provides a comprehensive and coherent interpretation supported by etymology, it remains challenging in a context where circular arguments can arise. The analysis leans heavily on the assumption that waxssa- designates the "bread", which is contentious and, in my opinion, highly uncertain. Additionally, the link between tubur(i)- and HLuw. /tuwaris(a)-/ is, in my view, phonologi-cally unacceptable (see fn. 14 above). Overall, the interpretation remains open to alternative viewpoints.
Addressing all the issues raised by this text is unfeasible at the present stage. Instead, it is possible to offer a few hypotheses regarding the occurrences of tuburiz. As observed, the Xan-thos poem is part of a coherent text group lauding the military deeds and piety of a ruler, likely Gergis-Xeriga. Notably, the Greek epigram connects the king's achievements closely with the divine world, attributing his conquests to Athena "destroyer of cities", acknowledging divine rewards, offering numerous tributes to Zeus, and culminating with the erection of a stele in dedication to the Twelve Gods. The Lycian A section also implies a votive inscription, mentioning several identifiable deities. One can make an educated guess that the Lycian B poem
57 Apud Neumann, ibid., V. Sevoroskin points out to the similarity of tubur(i)- with the supposed second member of anthroponyms of Greek transmission such as ApSupepig, Ep|aaxopopig or nepnevSupepig, without, however, making the connection explicit.
addresses a comparable subject, which is supported by the use of theonyms (e.g. Natri-, Trqqnt-), heronyms (Zrppedun-), and military terms (e.g. laxa- "military campaign") within the text.
Regarding tubur(i)-, it solely appears in the plural form tuburiz (nominative or accusative):
a) trmmiliz: tbiple: trpple: tuburiz: pduradi: xuzruwetiz (TL 44c.52-53);
b) murenedi: tuburiz : uple<s>iz (TL 44c.57);
c) mqrimiz ntuwiteni: uplesiz waxssadi: tuburiz (TL 44d.29-31).
In each instance, tuburiz is associated with adjectives such as trmmile/i- "Lycian" and xuz-ruwat(i)- "who has xuzr-". Both in b) and c) uplese/i- appears: this possessive adjective has not yet found a satisfactory translation but its stem is well rooted in local onomastics, as illustrated by the names OnA^g (Lycia, Kibyratis and particularly Pisidia), OnAeaig (Cabalis, Pisidia),58 OrcAeaiavog (Pisidia), OnAaviAig (Kibyratis), ApTL^-onA^g (Cabalis), and nava^u-onA^g (Kibyratis, Pisidia).59 The other terms remain equally obscure: the noun waxssa- in c), here in the ablative-instrumental (as murenedi and perhaps pduradi), is also a well-known element used in the regional onomastics: Lyc. Waxssepddime/i-, Waxssebe-; Car. Uksmu-/Waksmu-, Uksi-/Uqsi-; Oua^a^ouag/Oua^a^wg (Isauria, Cilicia), Oua^aig (Isauria).
Comparing a) (extended to the following lines) to a passage in Lycian A is allowed by common place-names displayed in both sections:
Lyc. A (TL 44b.47-56)
s=ed=de: ahata ha l[de] ene : qla=bi: ehetehi: se mahana : ehet l[he] Arnna : Tuminehi: Kerddi: Xakbi: epi=dl[e n]temle (e)si tama: s=e=we=nepe: astte: Trmm lis: s=ed=de=tuwete: kumezija : tere tere l Trqqnti: pddatahi: qnnakba : xrsseni: eh l bi: taba-haza : kumezija : padritahi: Arnlna Tuminehija : kumezija : Xakbija : kumelzi[j]a : se tukedri: Kerddi: ade: urublije l Hatahe : tubehi: prnnezi: se lihbeze eh[b]l ije He has left peace60 in the precinct of Peace and to the gods of Peace in Xanthos, in Tymnessos, in Kerththi, in Kandyba. (There) a temple is a building: Lycia had not (yet) made them. Everywhere he erected sanctuaries to the local Trqqnt (Storm-god) (and) to his twelve personal celestial gods, sanctuaries in Aphrodision (?) in Xanthos, Tymnesian sanctuaries, Kandybian sanctuaries and in Kerththi he made a monument to Handa (War-god) of Battle for his family and his lihbeze.
Lyc. B (TL 44c.51-54)
se=de keri trisu : qnnatbisu : prete l laxadi zreteni: sebe=be qirze : utakija Trmmiliz : l tbiple: trpple: tuburiz : pduradi: xuzruwetiz :) l Xazbi: Tuminesi: hntawa : Kridesi:
The mention of three cities, Tymnessos (Lyc. B Tuminesi; A Tuminehi), Kandyba (Lyc. B Xazbi; A Xakbi) and Kertthi (Lyc. B Kridesi; A Kerddi) in both passages justifies a comparison, although their contents are not identical.61 Despite a lack of comprehension for several terms, I propose the provisional translation:
58 This form probably reflects the anthroponymic conversion of the genitival adjective uplese/i-.
59 KPN § 1097. The exact references of the anthroponyms are listed in LGPN 5b (Lycia) and 5c (Kibyratis-Cabalis, Pisidia). Note also the strong presence of Greco-Anatolian noms d'assonance based on the similarity of the indigenous stem with Greek names: e.g. 'OnAwv (73 individuals in southern Anatolia vs. 10 in the rest of the Greek world covered by the LGPN), OnAwviavog (Pisidia), OnAwvig (Pisidia), DnA^Tiavog / DnA^Tiav^ (Pisidia). On the question of Greco-Anatolian noms d'assonance, see most recently Reveilhac in press.
60 The meaning "peace" for Lyc. A ahata- and Lyc. B asata- has recently been challenged by Melchert forthc., suggesting "good fortune" instead. However, the traditional meaning is defended in Reveilhac forthc. b.
61 According to Craig Melchert (pers. comm.), the division of the sequence must be considered. Indeed, the symbol :) likely serves to separate prosodic units, making it unclear whether the three toponyms should be linked
Three times in the country he pre- twelve (divine) protectors during the expedition and he carries (?) as utakija- of the designated properties Lycian tubur(i)-s xuzruwet(i)-s in double, in triple in Kandyba, in Tymnessos (and) a statue of the god Handa in Kerththi.
The formula qnnatbisu [...] zreteni "twelve times... a divine protector" might be understood as a poetic expression akin to Lyc. A qnnakba: xrsseni: ehbi "to his twelve personal gods", thus indicating the Twelve Gods protecting the king, hence a translation "twelve divine protectors". As for the group Trmmiliz: [...] tuburiz [...] xuzruwetiz, it might correspond to another poetic periphrase referring to the same deities, i.e. literally "the Lycian (divine) allies who have xuzr-". Syntactically, this group would be on the same level as hntawa (acc. sg.), likely derived from the Lycian theonym Hata- (< Sanda-), which would therefore exceptionally retain the initial aspirate typical of Lycian A (in order not to distort the divine name), with the -wa- suffix (< -wo- + *-h2-)62 and therefore literally meaning "that which is like Handa", a poetic turn corresponding to Lyc. A tukedri [...] Hatahe "statue of Handa".
Overall, while subject to speculation, this analysis attempts to unify tuburiz within both martial and religious themes, showcasing its relevance in varying contexts.
8. Conclusions
The study yields several significant conclusions. Firstly, the existence of a Proto-Luwic stem *tubar(i)- is affirmed through personal onomastics, indirectly evidenced in the place names Tuburehe(/i)-/Tupepia(a)og and the divine name or heronym Toupepig. This stem manifests diversely in various languages, with variations such as /TuPar(i)-/ in Hieroglyphic Luwian, °tubr-/°tybr- in Carian, x[p]epi- in Pisidian, °xopopig in Greek from Lycia, °xoupapig in Greek from Pisidia, and °xpnpig in Greek from Caria.
Moreover, a broader range of personal names reveals alternative forms of this onomastic stem: in Carian names as °8ubr-/°8ybr- and °dybyr- and in Greek-transmitted names as °Supepi-/°So/(u)pa/e/npa and °Supepog. The initial dental of *tubar(i)- undergoes voicing assimilation when in contact with the final [n] or [r] of the first member in these instances. This comprehensive explanation unifies and interprets multiple personal names that were previously explained through distinct and sometimes speculative processes.
Furthermore, the element *tubar(i)- appears as the first member in some compounds, albeit less frequently and often in syncopated forms: Tbr° and Topop° in Caria, Tpep°, Tpapa° and Tpepn° in Cilicia.
An overview of these occurrences, detailed in Sections 2 to 4, reveals that this stem is consistently associated with a theonym or a divine epithet. Through the combinations in an-throponymy, it is suggested that *tubar(i)- represents a divine title, which evolved into a divine hypostasis and then into a divine name or heronym, as implied by Stephanus of Byzantium's account about Toupepig.
to the previous clause, particularly given the blank at the end of line 53. Yet, incorporating this sequence into the subsequent clause is challenging due to the numerous nouns it contains. As a result, the sequence at the start of line 54 might be an enjambment, possibly emphasized within the broader poetic structure.
62 The -wa- suffix, attested in several Anatolian languages (Hittite, Luwian, Lycian, and Carian), was used to form words denoting a social category (Rieken & Sasseville 2014). However, it originally had an "equitative" function, which can still be seen in certain nouns such as prnna-wa-, lit. "that which is like a house (prnna-)", hence "tomb-house, mausoleum" (Rieken & Sasseville 2014: 310).
Etymologically, *tubar(i)- likely relates to the Luwian /tub(a)-/ "to strike" and Lycian tube-"battle". It might be a compound /tub-ar(i)-/ meaning "battle companion, comrade-in-arms". This concept echoes Hittite and Luwian texts describing deities engaging in battles alongside their protégés.
Finally, tubar(i)- finds a direct avatar in the Lycian B lexicon, with tuburiz. This word potentially holds the meaning of "(divine) allies" in the poem on the Xanthos Pillar, considering the martial and religious context, although this interpretation remains uncertain.
The prominence of tubar(i)- in Luwic onomastics suggests its use in defining and designating deities, evident in personal names across Luwian, Lycian, Carian, and Pisidian collections, Greek inscriptions from southern Anatolia, the heronym Toùpepiç from indirect sources, and the place name Tyberissos (see Réveilhac forthc. a).
References
Adiego, Ignacio-Javier (Ignasi-Xavier). 2007. The Carian Language. Leiden / Boston: Brill.
Adiego, Ignasi-Xavier. 2019a. A Kingdom for a Carian Letter. In: Ignasi-Xavier Adiego, José-Virgilio García Tra-bazo, Mariona Vernet, Bartomeu Obrador-Cursach, Elena Martínez Rodríguez (eds.). Luwic Dialects and Anatolian Inheritance and Diffusion: 11-50. Barcelona: Universitat de Barcelona. Adiego, Ignasi-Xavier. 2019b. The Survival of the God Name Sarruma in Cilician Names in the Greek Sources. Altorientalische Forschungen 46: 147-160. Adiego, Ignasi-Xavier. 2020a. Consonant Clusters, Defective Notation of Vowels and Syllable Structure in Car-omemphite. In: Ronald I. Kim, Jana Mynárová, Peter Pavúk (eds.). Hrozny and Hittite: The First Hundred Years. Proceedings of the International Conference Held at Charles University, Prague, 11-15 November 2015: 95-119. Leiden / Boston: Brill.
Adiego, Ignasi-Xavier. 2020b. Adaptación griega de nombres licios, adaptación licia de nombres griegos: aspectos fonológicos. In: C. Ruiz Darasse (ed.). Comment s'écrit l'autre? Sources épigraphiques et papyrologiques dans le monde méditerranéen antique: 43-60. Bordeaux: Ausonius éditions. Adiego, Ignasi-Xavier. 2022. Luwian Tarhunaza-, Cilician Tç>OKOvaÇaç, TQiKovaÇaç. Indogermanische Forschungen 127: 75-90.
Anrich, Gustav. 1913-1917. Hagios Nikolaos: Der Heilige Nikolaos in der griechischen Kirche. Texte und Untersuchungen,
2 volumes. Leipzig: B. G. Teubner. Bauer, Anna H., Elisabeth Rieken, Zsolt Simon, David Sasseville, Andreas Opfermann. 2022. Proto-Anatolian *miéu-/miu-' (eDiAna-ID 1440). In: eDiAna. Available online at: www.ediana.gwi.uni-muenchen.de/dictionary. php?lemma=1440 [accessed on 19/10/2023]. Bean, Georg. E., Terence B. Mitford. 1970. Journeys in Rough Cilicia 1964-1968. Wien / Graz: Böhlau Verlag. Billing, Oscar. 2023. Lycian A tuwere/i- (eDiAna-ID 1952). In: eDiAna. Available online at: www.ediana.gwi.uni-
muenchen.de/dictionary.php?lemma=1440 [accessed on 19/10/2023]. BoHa 19 = Herbordt, Suzanne. 2005. Die Prinzen- und Beamtensiegel der hethitischen Großreichszeit auf Tonbullen aus dem
Ni§antepe-Archiv in Hattusa (Bogazköy-Hattusa Ergebnisse der Ausgrabungen 19). Mainz am Rhein: von Zabern. Bonnet, Corinne, Thomas Galoppin. 2021. Divine Names on the Spot: Towards a Dynamic Approach of Divine Denominations in Greek and Semitic Contexts. In: Thomas Galoppin, Corinne Bonnet (eds.). Divine Names on the Spot. Towards a Dynamic Approach of Divine Denominations in Greek and Semitic Contexts: 1-17. Leuven / Paris / Bristol, CT: Peeters. Brixhe, Claude. 1976. Le dialecte grec de Pamphylie: documents et grammaire. Paris: Adrien Maisonneuve. Brixhe, Claude. 2016. Stèles et langue de Pisidie. Nancy: A.D.R.A.
Cammarosano, Michele. 2015. Foreign Gods in Hatti. A New Edition of CTH 510. KASKAL. Rivista di storia, ambi-
enti e culture del Vicino Oriente Antico 11: 199-244. CIG = 1860-1890. Corpus inscriptionum Graecarum. Berlin: Officina Academia/Reimer. CTH = Laroche, Emmanuel. 1971. Catalogue des textes hittites (2nd ed.). Paris: Klincksieck.
eDiAna = Miller, Jared, Elisabeth Rieken, Olav Hackstein, Ilya Yakubovich (eds.). 2017-. eDiAna. Digital Philological-Etymological Dictionary of the Minor Ancient Anatolian Corpus Languages. Available online at: www.ediana.gwi.uni-muenchen.de.
van Gessel, Ben H. L. 1998-2001. Onomasticon of the Hittite Pantheon. 3 vols. Leiden / Boston / Köln: Brill.
Goetze, Albrecht. 1933. Die Annalen des Mursilis. Leipzig: Hinrichs.
Hajnal, Ivo. 1995. Der lykische Vokalismus. Methode und Erkenntnisse der vergleichenden anatolischen Sprachwissenschaft, angewandt auf das Vokalsystem einer Kleincorpussprache. Graz: Leykam.
Hawkins, J. David. 2000. Corpus of Hieroglyphic Luwian Inscriptions. Vol. 1: Inscriptions of the Iron Age. Berlin / New York: De Gruyter.
Heberdey, Rudolf, Adolf Wilhelm. 1896. Reisen in Kilikien, ausgeführt 1891 und 1892 im Auftrage der Kaiserlichen Akademie der Wissenschaften. Wien: C. Gerold's Sohn.
Houwink ten Cate, Philo H. J. 1961. The Luwian Population Groups of Lycia and Cilicia Aspera during the Hellenistic Period. Leiden: Brill.
I.Arykanda = §ahin, Sencer. 1994. Die Inschriften von Arykanda. Bonn: Habelt.
I.Cilicie = Dagron, Gilbert, Denis Feissel. 1987. Inscriptions de Cilicie. Paris: De Boccard.
I.Carie hautes terres = Debord, Pierre, Ender Varinlioglu. 2001. Les hautes terres de Carie. Bordeaux: Institut Ausonius.
I.Pan du désert = Bernand, André. 1972. Pan du désert. Leiden: Brill.
I.Termessos Suppl. = iplikçioglu, Bülent, Güler Çelgin, A. Vedat Çelgin. 1991-2007. Epigraphische Forschungen in Termessos und seinem Territorium, 4 volumes. Wien: Verlag der Österreichischen Akademie der Wissenschaften.
I.Westkilikien Rep. = Hagel, Stefan, Kurt Tomaschitz. 1998. Repertorium der westkilikischen Inschriften. Nach den Scheden der kleinasiatischen Kommission der Österreichischen Akademie der Wissenschaften. Wien: Verlag der Österreichischen Akademie der Wissenschaften.
IG = Deutsche Akademie der Wissenschaften zu Berlin. 1913-. Inscriptiones Graecae. Berlin: De Gruyter.
Imparati, Fiorella. 1965. L'autobiografia di Hattusili I. Studi Classici e Orientali 14: 40-76.
KBo = 1916-1921. Keilschrifttexte aus Boghazköi. Leipzig/Berlin: Gebr. Mann-Verlag.
KON = Zgusta, Ladislav. 1984. Kleinasiatische Ortsnamen. Heidelberg: Winter.
KPN = Zgusta, Ladislav. 1964. Kleinasiatische Personennamen. Prag: Verlag der Tschechoslowakischen Akademie der Wissenschaften.
KUB = 1921-. Keilschrifturkunden aus Boghazköi. Berlin: Akademie-Verlag.
Kizil, Abuzer, Patrice Brun, Laurent Capdetrey, Raymond Descat, Pierre Fröhlich, Koray Konuk. 2015. Pidasa et Asandros: une nouvelle inscription (321/0). Revue des Études Anciennes 117: 371-409.
Labarre, Guy. 2010. Le dieu Mèn et son sanctuaire à Antioche de Pisidie. Bruxelles / Fernelmont: E.M.E. & InterCommunications.
Laroche, Emmanuel. 1946. Recherches sur les noms de dieux hittites. Revue hittite et asianique 7: 7-139.
Laroche, Emmanuel. 1966. Les noms des Hittites. Paris: Klincksieck.
Laroche, Emmanuel. 1973. Études de linguistique anatolienne. Revue hittite et asianique 31: 83-99.
LGPN 5b = Balzat, Jean-Sébastien, Thomas Corsten, Richard W. V. Catling (eds.). 2013. Lexicon of Greek Personal Names. Vol. 5b: Coastal Asia Minor: Caria to Cilicia. Oxford: Clarendon Press.
LGPN 5c = Ballzat, Jean-Sébastien, Richard W. V. Catling, Édouard Chiricat, Thomas Corsten (eds.). 2018. Lexicon of Greek Personal Names. Vol. 5c: Inland Asia Minor. Oxford: Clarendon Press.
Martinez Rodriguez, Elena. 2019. Phonotactics of the Lycian Labial Glide Clusters. Indogermanische Forschungen 124: 29-229.
Masson, Olivier. 1990. Onomastica Graeca selecta, I et II (introduction et index par C. Dobias et L. Dubois). Nanterre: Université de Paris X-Nanterre.
Melchert, H. Craig. 1978. The Acts of Hattusili I. Journal of Near Eastern Studies 37/1: 1-22.
Melchert, H. Craig. 1993. Cuneiform Luvian Lexicon. Chapel Hill: self-published.
Melchert, H. Craig. 1994. Anatolian Historical Phonology. Amsterdam / Atlanta: Rodopi.
Melchert, H. Craig. 2004. A Dictionary of the Lycian Language. Ann Arbor / New York: Beech Stave Press.
Melchert, H. Craig. 2021. A Possible New Greco-Carian Contact Phenomenon. In: M. Bianconi (ed.). Linguistic and Cultural Interactions between Greece and Anatolia. In Search of the Golden Fleece: 108-115. Leiden/Boston: Brill.
Melchert, H. Craig. Forthc. The Ässiya Deities. To appear in: The proceedings of the 1st workshop The Gods of Anatolia and their Names held in Barcelona, May 17th-19th 2023.
Mouton, Alice, Ilya Yakubovich. 2021. Where Did One Speak Luwili? Geographic and Linguistic Diversity of Luwian Cuneiform Texts. Journal of Language Relationship 19/1: 25-53.
Mouton, Alice. 2023. Solar Deities in the Kuwatalla Ritual Tradition: Epithets and Functions. In: Livio Warbinek, Federico Giusfredi (eds.). Theonyms, Panthea and Syncretisms in Hittite Anatolia and Northern Syria. Proceedings of the TeAI Workshop Held in Verona, March 25-26, 2022: 127-140. Firenze University Press.
Neumann, Günter. 1976. Zu einigen hethitisch-luwischen Personennamen. Zeitschrift für vergleichende Sprachforschung 90/1: 139-144.
Neumann, Günter. 2007. Glossar des Lykischen (überarbeitet und zum Druck gebracht von Johann Tischler). Wiesbaden: Harrassowitz.
Oettinger, Norbert. 2001. Hethitisch -ima- oder: Wie ein Suffix affektiv werden kann. In: Gernot Wilhelm (ed.). Akten des IV. Internationalen Kongresses für Hethitologie, Würzburg, 4.-8. Oktober 1999: 456-477. Wiesbaden: Harrassowitz.
Oreshko, Rostislav. 2014. The Strange Case of Dr. FRATER and Mr. DOMINUS: a Re-Consideration of the Evidence Concerning Luwian nani-. In: Piotr Taracha (ed.). Proceedings of the Eighth International Congress of Hitti-tology, Warsaw, 5-9 September 2011: 614-631. Warsaw: Agade.
Réveilhac, Florian. 2018. Contact linguistique et emprunts onomastiques entre grec et lycien: apports à la phonétique et à la morphologie. PhD dissertation. Sorbonne Université, Paris.
Réveilhac, Florian. 2021. Geminate Consonants in Lycian. A Twofold Interpretation. In: Annick Payne, Sarka Vel-harticka, Jorit Wintjes (eds.). Beyond All Boundaries: Anatolia in the First Millennium BC: 551-575. Leuven / Paris / Bristol, CT: Peeters.
Réveilhac, Florian. Forthc. As you like it: les noms d'assonance lyciens-grecs. Esquisse d'une méthodologie. In: Alcorac Alonso Déniz, Julian Méndez Dosuna, Enrique Nieto Izquierdo, Gilles van Heems (eds.). Contact linguistique en Grèce ancienne: diachronie et synchronie (CoLiGA): 77-101. Lyon: MOM Éditions.
Réveilhac, Florian. Forthc. a. Tybérissos, un lieu divin : sur les traces d'une épiclèse en Anatolie méridionale. In: Alcorac Alonso Déniz, Enrique Nieto Izquierdo (eds.). Noms de lieux en Méditerranée antique: approches linguistiques. Nancy: A.D.R.A.
Réveilhac, Florian. Forthc. b. Anatolian Names in °aor|Taç/°aoaTr|ç/°aoaTaç;, CLuw. *assatta-, Lyc. B asata- and Lyc. A ahata-. To appear in: News from the Lands of the Hittites 8 (2024): 197-213.
Réveilhac, Florian. Forthc. c. Give a God a Bad Name and Beg Him: the Lycian Theonym zou^£v5lç/zo^£v5lç. To appear in: Proceedings of the 1st workshop The Gods of Anatolia and their Names held in Barcelona, May 17th-19th 2023.
Rieken, Elisabeth, David Sasseville. 2014. Social Status as a Semantic Category of Anatolian: The Case of PIE *-uo-. In: H. Craig Melchert, Elisabeth Rieken, Thomas Steer (eds.). Munus amicitiae: Norbert Oettinger a collegis et amicis dicatum: 302-314. Ann Arbor / New York: Beech Stave Press.
Rieken, Elisabeth, David Sasseville. 2022. Proto-Luwian *Hudä- (eDiAna-ID 2107). In: eDiAna. Available online at: www.ediana.gwi.uni-muenchen.de/dictionary.php?lemma=2107 [accessed on 19/10/2023].
Robert, Louis. 1955. Hellenica. Recueil d'épigraphie, de numismatique et d'antiquités grecques, Vol. X. Paris: Adrien Mai-sonneuve.
Sasseville, David. 2021. Bread and Wine in Lycian Poetry. A Contextual Interpretation of TL 44c. Kadmos 60: 173-188.
Sasseville, David 2022. Palaic hasäwant- (eDiAna-ID 338). In: eDiAna. Available online at: www.ediana.gwi.uni-muenchen.de/dictionary.php?lemma=2107 [accessed on 19/10/2023].
Sasseville, David. 2023. Lycian A ntuweriha- (eDiAna-ID 1806). In: eDiAna. Available online at: www.ediana.gwi.uni-muenchen.de/dictionary.php?lemma=2107 [accessed on 19/10/2023].
Sasseville, David, Caroline Schneider, Andreas Opfermann. 2023. Proto-Anatolian *(s)teup- (eDiAna-ID 288). In: eDiAna. Available online at: www.ediana.gwi.uni-muenchen.de/dictionary.php?lemma=2107 [accessed on 19/10/2023].
Schürr, Diether. 2013. Kaunisch-Karisches in Krya: Revision der Grabinschrift und Vergleiche. In: O. Henry (ed.). 4th Century Karia. Defining a Karian Identity under the Hekatomnids: 21-31. Istanbul: Institut Français d'Études Anatoliennes.
Schürr, Diether. 2016. Zum Agora-Pfeiler in Xanthos V: Das Nordgedicht auf Cheriga (TL 44c, 32 ff.). Kadmos 55: 147-196.
SEG = 1923-. Supplementum epigraphicum Graecum. Leiden: Brill.
Simon, Zsolt. 2016. Toubassis. Zum Weiterleben anatolischer Gottheiten in der Eisenzeit. Nouvelles Assyriologiques Brèves et Utilitaires 2016/2: 82-84.
Simon, Zsolt. 2016-2017. Zur Bedeutung von karisch eri-. Münchener Studien zur Sprachwissenschaft 70: 259-264.
Simon, Zsolt. 2022a. Carian kSous (eDiAna-1071). In: eDiAna. Available online at: www.ediana.gwi.uni-muenchen.de/ dictionary.php?lemma=2107 [accessed on 19/10/2023].
Simon, Zsolt. 2022b. Luwian *dupattar/dupattn- (eDiAna-ID 345). In: eDiAna. Available online at: www.ediana.gwi.uni-muenchen.de/dictionary.php?lemma=2107 [accessed on 19/10/2023].
Sittig, Ernst. 1911. De Graecorum Nominibus Theophoris. Halis Saxonum: Karras.
Sterrett, John R. S. 1888. The Wolfe Expedition to Asia Minor. Boston: Damrell and Upham.
TAM II = Kalinka, Ernst. 1920-1944. Tituli Asiae Minoris. Vol. 2, Tituli Lyciae: linguis Graeca et Latina conscripti. Vin-dobonae: R. M. Rohrer.
Taracha, Piotr. 2009. Religions of Second Millennium Anatolia. Wiesbaden: Harrassowitz.
TL = Kalinka, Ersnt. 1901. Tituli Asiae Minoris, Vol. 1, Tituli Lyciae: lingua Lycia conscripti. Vindobonae: A. Hoelder.
Valério, Miguel. 2007. "Diktaian Master": a Minoan Predecessor of Diktaian Zeus in Linear A? Kadmos 46: 3-14.
Valério, Miguel. 2015. Linear A du-pu2-re, Hittite tabarna and their Alleged Relatives Revisited. Journal of Language Relationship 13/4: 329-354.
Valério, Miguel. 2023. Drive for power and the Indo-European source of Luwian muwa-. In: J.-V. Garcia Trabazo et al. (eds.). New Approaches on Anatolian Linguistics: 265-291. Barcelona: Universitat de Barcelona.
Wörrle, Michael. 2012. Die griechischen Sepulkralinschriften von Limyra. In: Jürgen Borchhardt, Anastasia Pekri-dou-Gorecki (eds.). Limyra: Studien zu Kunst und Epigraphik in den Nekropolen der Antike: 411-457. Wien: Phoi-bos-Verlag.
Yakubovich, Ilya. 2013. Anatolian Names in -wiya and the Structure of Empire Luwian Onomastics. In: Alice Mouton, Ian Rutherford, Ilya Yakubovich (eds.). Luwian Identities: Culture, Language and Religion between Anatolia and the Aegean: 87-123. Leiden / Boston: Brill.
Yakubovich, Ilya. 2017. The Luwian Title of the Great King. In: Alice Mouton (ed.). Hittitology Today: Studies on Hittite and Neo-Hittite Anatolia in Honor of Emmanuel Laroche's 100th Birthday: 38-50. Istanbul: Institut Français d'Études Anatoliennes-Georges-Dumézil.
Yakubovich, Ilya. 2021. The Anatolian Connections of the Greek God Enyalius. In: Michele Bianconi (ed.). Linguistic and Cultural Interactions between Greece and Anatolia. In Search of the Golden Fleece: 233-245. Leiden/Boston: Brill.
Yakubovich, Ilya, Alice Mouton. 2023. Luwili. Hittite-Luwian Ritual Texts Attributed to Puriyanni, Kuwatalla and Silalluhi (CTH 758-763), Vol. 1: Edition and Commentary, Vol. 2: Discussion and Glossary. Wiesbaden: Harrassowitz.
Zehnder, Thomas. 2010. Die hethitischen Frauennamen. Katalog und Interpretation. Wiesbaden: Harrassowitz.
Zimmermann, Martin. 2003. Hafen und Hinterland. Wege der Akkulturation an der lykischen Küste. Vorbericht über die Feldforschungen in den zentrallykischen Orten Tyberissos und Timiussa in den Jahren 1999-2001, mit einem Beitrag von Oliver Hülden. Istanbuler Mitteilungen 53: 265-312.
Флориан Ревеяк. Божественный эпитет *tubar(i)- в лувийской ономастике.
Отправной точкой данной статьи является исследование М. Валерио (2015), посвященное нескольким группам личных имен из южной Анатолии. Здесь предлагается восстановить одну ономастическую базу, *tubar(i)-, общую для нескольких имен из различных языковых корпусов, принадлежащих к лувийской подгруппе (лувийский, карийский, писидийский) и греческим эпиграфическим источникам из южной Анатолии (от Карии до Киликии). Данный продуктивный элемент ассоциируется с многочисленными божественными именами или эпитетами в составе композитов, что указывает на то, что мы имеем дело с божественным титулом. Его значение можно восстановить как "боевой спутник, соратник", изначально характеризовавший различных богов, чья роль заключалась в охране своих протеже на поле боя.
Ключевые слова: лувийские языки; анатолийские языки; лувийский язык; ликийский A язык; ликийский B язык; карийский язык; писидийский язык; греческий язык; ономастика; теофорные имена; анатолийская религия.
