CC BY
12
L. Repansek
FLUVIUS ARGAO, QUIS IN SINUM ARGO FLUIT? 1
В статье на основе детального морфологического анализа обсуждается связь между названием реки Argaone (совр. Драгония) в Равенн-ской «Космографии» и письменно зафиксированым не ранее 16 в. названием залива Largon(e), в который эта река впадает. На обширном индоевропейском материале рассматривается круг проблем, связанных с суффиксальным комплексом *-аиоп- в древних гидронимах.
Ключевые слова: Argaone, Largon(e), топонимия, словообразование.
The modem day Dragonja (It. Dragogna, older flume d'Argogna) is a Histrian river flowing into the Adriatic at the Gulf of Piran (Sln. Piranski zaliv, Cr. Piranski zaljev ~ Savudrljska uvala, It. Vallone di Pirano). Its ancient name has come down to us as Argaone (Rav., Cosm. IV, 36, 75; see also Vedaldi Iasbez 1994: 122), which quite unproblematically points to the older nasal-stem *Argaon-, with nom. sg. Argao(n)* of probable but unknown currency. Contrary to the communis opinio (see Ramovs 1927: 22; ESSZI 122; Roncevic Brozovic 1995: 23; Simunovic 2013: 201), the Slavic continuant goes back to a later adstratally influenced renovation of the original phonetic substitute at best, but in any case cannot represent an old loan (similarly Furlan 2002: 31, ft. 4) as the latter would result in Sl. **Dargujna > **Dragyna uel sim. (i .e. presupposing an older Romance *Dargon(')V as its source). However this may be, it is clear that the name of the river as it stands is in itself the result of a later renaming process within Romance from the original Argao(n)* ~ Argaone to the descriptive and analytical D'Argone - the logical starting point for a number of ensuing reinterpretations and alterations 2 - beyond any doubt reflecting the transference of prominence from the river and/or its
1 The text has been prepared with the input system ZRCola (http://ZRCola.zrc-sazu.si) developed by Dr Peter Weiss, fellow of the Scientific research centre at the Slovene Academy of Sciences and Arts (http://www.zrc-sazu.si).
Draconius fluvius, Dragone, Dragugne, Dragogna*, Dragujna etc. (Furlan 2002: 31; SVI I: 147-148 with older bibliography; Simunovic 2013: 201).
mouth (by way of the pars pro toto principle) to the bay in which it flows. Indeed, the Gulf of Piran is recorded on 16th c. maps as Largone ~ Largon (cf. Pietro Coppo, Histriae Tabula, dated 1525)3, i. e. L'Argon(e), the name usually being interpreted as a variant or a corruption of sorts of the river-name (see B. Benussi. L'Istria sino ad Augusto. Triest, 1883: 13, 144 (non vidi); Rosamani 1990 s.v.). Although it is perfectly feasible that an underlying Romance *Argone represents a phonetic continuant of *Argaone (be it via the monophthongization of -ao- in hiatus or, more straightforwardly, synaloepha - see Furlan 2002 forth.), it would nevertheless be intriguing, leaving aside the rather disturbing temporal difference in the attestations, to review the later/modern naming model (based on the spatial/possessive rather than the descriptive criterion) as reproducing the historical relationship, thus obviating the two-way naming process RN t GN and uncomfortable historical phonology. However, this alternative opens up the difficult question of historical morphology (more precisely word/name-formation) behind the hydronym and the nature of its relation to the admittedly conjectural * Argon-.
The phonetic reality behind the spelling -ao-, which as is well known goes back to *-auo ~ *-auon- (Krahe 1942: 209-218 (esp. 213ff.), Schmid 1969: 135), is confirmed at least by the Pannonian hydronym Nedao (lord. 49,260), possibly LN Azao (iA 246,3) if not refuted by Odiabo in Notitia Dignitatum (33,8), which would point to an older *-auo (~ *-auon-) or *-auum (see esp. Anreiter 2001: 2324, but cf. Falileyev 2014 forth., sub Azao). This allows for the reconstruction of a nasal stem with possible but unattested length of the suffixal vowel. As the relation of *Argauon- to the merely putative *Argon- can only be assessed upon perfect understanding of the historical morphology behind the former derivative, it is necessary to provide convincing and diachronically rooted arguments for the shape of the hydronym's derivational base and the function of the derivational suffix(es) in both synchronic and diachronic perspectives.
The underlying lexeme is usually identified as the PIE adjective *H2erg'- (Krahe 1942: 211-12, IEW: 64ff., NIL: 317ff.), meaning 'white, shiny', secondarily also 'fast', be it in its full or zero grade: *H2erg'- > *arg— *H2rg'- > *arg- (cf. Arsia, a rather wide-spread river-name with two Histrian 4 continuants, which is commonly
3 Courtesy of Dr. Drago Kladnik and Dr. Metka Furlan, cf. http://db. nsk.hr/HeritageDetails.aspx?id=917.
4 In purely geographical terms!
brought into connection with OInd. ársati = Hitt. ars-: a-ar-as-zi 'flows' (Krahe 1964: 64) and as such must go back to *H1rs-ia, with #HR- > #ar-), thus at face value arriving at something like "The shiny (river)" uel sim. This is problematic, however, as apart from Gr. ápyó^, which on the other hand is nearly universally equated with OInd. rj-rá- < *H2rg'-ró- by dissimilatory loss of *r (i.e. since Wackernagel 1897: 9, cf. Rix 1992: 18, see NIL 312, and most recently Vine 2011: 1 for full(er) bibliography), there is no trace of a thematic derivative of *H2erg'- in the available comparanda, so that an underlying *arg-o- 'shiny, white' cannot be indiscriminately assumed for the hydronym at question. Moreover, a few comparable derivatives in *- arn(n)- point to their deverbative origin, a case in point being the probably agentival Nedao (see sup., to *ned- 'tonen, drohnen', LIV2 448), cf. also Patau(i)um (possibly to *petH2-'ausbreiten', LIV2 478-79 - Anreiter 2001: 114), however the likes of Timauus, Sarauus, Iuuauus/um, or Ouilaua are silent. These various toponymical bases in *-au(i)o- dispersed throughout the wider Eastern Alpine and Pannonian areas, surely belong to different linguistic strata and at present cannot be hoped to yield to closer scrutiny. Nevertheless, a basic *-au- can be abstracted from cases where it is clearly suffixal such as in Iuu-auum ~ Iuu-auus vs. Iu-enna, both to *iuu-. These cases must represent an older *-a-uo-rather than the synchronically motivated *-au- (pace Anreiter 2001: 24 et passim) from *-o-uo- (cf. Drauus, Sauus, both from *-ou-o-with expected a < *o either universally or before (heterosyllabic) *u)6 or, as at least some of these sequences were demonstrably short (e.g. Pataua, PetauiV ^ Sl. *Pitauiu > *Pbtu2ib!), from *-a-uo- < *-H}-uo-, cf. *petH2-uo- > *peta-uo- (for deverbative abstract/action nouns in *-uo- compare OInd. é-va- < *H¡ei-uo-, Lat. riuus < *H3r(e)iH-uo-, PSl. *piuo < *p(e)iH3-uo- etc.) "(IdgGr II/1: § 128). Within the limits of viable semantic motivation a possible candidate for *Argauon- as a deverbative formation would perhaps be *H3reg'-(LIV2 304-5) 'gerade richten, ausstrecken', but one would first need to prove on purely morphological grounds that the derivative as it stands cannot be of denominal origin and provide convincing argumentation in favour of the phonetically required zero-grade! As regards the former, an overview of the derivational nest of PIE *H2erg'- in fact reveals a fully fledged system of "Caland"-type
5 Krahe 1942: 209, more recently Anreiter 2001: 246-47.
6 Cf. Krahe 1942: 151, Anreiter 2001: 14.
alternations in *H2(e)rg'-i- *H2erg'-i-n6-; Risch 1974: 99, 154) ~ *H2rg'-ro- ~ *H2erg'-ent- (^ *H2rg'-nt-6-), with the addition of what seem to be primary derivatives *H2erg'-es-1 and probably also *H2erg'-uer/n- (as the basis of the secondary *H2(e)rg'-ur/n-, cf. tpn. Arg-yr-unt-um, etn. Apy-up-i-voi).8 Hom. apy^, -ex- 'shining' < H2rg'-e(-)t- may or may not be a denominal dental stem (see Schindler 1976: 351, Rieken 1999: 86-88). In any case, typology here allows for the reconstruction of a thematic o-grade *H2org'-6-(cf. *krouH2-6- to *kreuH2-, *H1roudh-6- to *H1reudh- etc.), less so an e-grade thematic derivative *H2erg'-o- (~ *Hireudh -o- ?) as would of course be the face-value reconstruction of Gr. dpyo^ if indeed original. If so, one would then be required to account for the deadjectival function of *-äuon-. If original length in the suffix is maintained, *-äuon- could very well represent a simple nasal extension (with individualizing/substantivizing force of the OHG bero type) to *-ä-uo-, a secondary suffix abstracted from -eH2-stems (cf. PSl. *lQkawb 'malus, malignus' to *lQka 'dolus'(3CC3 XVI: 143) as in PSl. *bordatt = Lat. barbatus to *bhardheH2- etc.),9 which is immediately comparable in form and function to the Balto-Slavic deverbative and denominal *-ä-uo-/-ä (as already convincingly argued by Schmid - Schmid 1969: 126-27, 131; IdgGr II/1: §125).
The main problem with this external comparison, however, is that it is exactly the adjectival function of the suffix, as arguably also employed in the case of *Argäuon-, which is difficult to account for. Indo-Iranian denominative adjectives in *-uan- (< *'-uon, *-un-' ~ *-un-', *-uen(-i) - AiGr II/2: §§718ff.) - "primary" deverbal agent nouns of the yajvan-type are largely internally Indo-Iranian and surely secondary - are of little use here because their function as a fairly productive means in the formation of adiectiva abundantiae may have an altogether different source of analogical spread10 (cf. snavan- 'Band, Sehne' < *sneH1-uon- to *sneH1-ur/n- (YAv. snäuuars); gravan- 'Mühlstein' < *gHreH2-u-on- etc., also note that the feminine forms utilize *-uar-i, to which cf. *pi-uan- 'fat', f. *pi-uar-T, < *piH-uon-, an originally amphikinetic internal derivative)11.
7 But cf. Stuber 2002: 50-55.
8 Cf. NIL s.v., Krahe 1942: 213.
9 Cf. Gr. PN Maxarov (: Maxn 'battle') etc. (Risch 1974: 57).
10 The exocentric character of which is obvious from the consistent lengthening of the vowel before the suffix (cf. satya-van-, AiGr II/2: §718e).
11 AiGr loc. cit., IdgGr II/1: §§232-35.
However, on the strength of Hom. onarov 'comrade' < *onaprov, which seems to have been built to a primary feminine o-stem derivative *sok4-eH2- to *sek4- 'follow' (Schindler 1975: 63, Schmid 1969: 137, Risch 1974: 57 (cf. Schwyzer 1977: §III/29a), Rasmussen 1989: 160-61), it would be possible to recognize in *-a-uon- a formal individualisation of the underlying *-a-uo-, i.e. *-a-u-on-12, and predict its spread to denominal and deverbal agentival use, cf. Hom. SiSu^aove 'twins' to clearly adjectival SlSu-^o- 'dual' (GEW s.v.), Lith. zinovas 'sciens' (: zinoti) ^ gerovas 'drinker' (: gerti -Otr^bski 1965: §151-57) etc. The possessive force of the denominal *-uo- (for the suffix' function in secondary derivatives cf. Gaul. oronym 'Apvopa = *Abn-o-ua13 to Celt. *abon- 'river', av. pafnava-to par^na- 'feather', OInd. kesava- 'hairy, long-haired' to kesa- etc.) is better preserved in the likes of PSl. *kfi>uaub (: *kry, -bu) 'bloody' (< *-a-uo-!), *rQkam (: *rQka) 'sleeve' ~ Lith. rankove, and nomina loci such as *DQbraua (tpn.) (: *dQbrb 'дубовый лес').14 However, apart from the Lith. type sventove 'sacred spot' (: sventas 'sacred'), which on the strength of senove 'vetustas' to senas 'seneX may have its source in abstract nouns, the suffix at question seems not to have been deadjectival. This is expected since according to the model proposed above *-a-uo- would have had a clearly adjectival function, a fact that receives additional support in the nasal extension *-a-u-on-, in which the deadjectival function of the individualizing *-on- must undoubtedly be recognized. As such, the RN *Arg-auon- (: *Argo-) is then perhaps best seen as a simple univerbization, cf. PSl. hydronymic *Cbrnaua (cf. Sln. RN Crnava -ESSZI: 101) to *сьтъ, i.e. < *Cbrna reka, = *Cbrnica, rather than a possessive derivative. As no direct connection between Argaone and 16th-century L'Argon(e) can be demonstrated on morphological grounds, even if the latter be old and could in theory go back to the individualized/substantivized *Arg-on- ^ *Argo-, the forged juxtaposition in the title is rendered unlikely on grounds of its improbable anachronism.
12 But see IdgGr II/1 §234c, AiGr II/2 §718, Schindler 1975: 63, who connect these derivatives with the Ind-Ir. maghavan-/magauuan- type (agentival) adjectives.
13 I.e. for the expected athematic *abn-ua, see Ziegler 2003: 292.
14 ESSZI: 111, ЭСС^ V: 95.
Abbreviations
Cr. = Croatian; GN = name of gulf; Gr. = Old Greek; hdr. = hydronym;
Hom. = Homeric Greek; Ir. = Old Iranian; It. = Italian; Lat. = Latin, Lith. =
Lithuanian; OInd. = Old Indic; PN = personal name; RN = river name;
Pann. = Pannonian; (P)Sl. = (Proto-)Savic; Sln. = Slovene; tpn. = toponym;
YAv. = Young Avestan.
Bibliography
ЭССЯ - Этимологический словарь славянских языков: праславянский лексический фонд / отв. ред. О. Н. Трубачев. М.: «Наука», 1974-.
AiGr II/2 -Wackernagel J., Debrunner A. Altindische Grammatik. Bd. II/2: Die Nominalsuffixe (von A. Debrunner). Göttingen, 1954.
Anreiter 2001 - Anreiter P. Die vorrömischen Namen Pannoniens. Budapest: Archaeolingua, 2001.
Brozovic Roncevic 2005 - Brozovic Roncevic D. Iz istarske hidronimije // Folia onomastica Croatica. 1995. T. 4. P. 19-30.
ESSZI - Snoj M. Etimoloski slovar slovenskih zemljepisnih imen. Ljubljana: Zalozba Modrijan in Zalozba ZRC, ZRC SAZU, 2009.
Falileyev 2014 - Falileyev A. In search of the Eastern Celts. Budapest, 2014, forth.
Furlan 2002 - Furlan M. Predslovanska substratna imena v slovenscini // Jezikoslovni zapiski. 2002. # 8/2. P. 29-35.
GEW - Frisk Hj. Griechisches etymologisches Wörterbuch. Bde. I-III. Heidelberg, 1960, 1970, 1972.
IdgGr II/1/1 - Brugmann K. Vergleichende Laut-, Stammbildungs- und Flexionslehre nebst Lehre vom Gebrauch der Wortformen der indogermanischen Sprachen II: Lehre von den Wortformen und ihrem Gebrauch. 1. Teil: Allgemeines. Zusammensetzung (Komposita). Nominalstämme, Strassburg, 19062.
IEW - Pokorny J. Indogermanisches etymologisches Wörterbuch. Bern; München: Francke, 1959-1969.
Krahe 1942 - Krahe H. Beiträge zur illyrischen Wort- und Namenforschung // Indogermanische Forschungen. 1942. Bd. 58. S. 132-152, 209-138.
Krahe 1964 - Krahe H. Unsere älteste Flussnamen, Wiesbaden, 1964.
NIL - Wodtko D. S., Irslinger B, Schneider C. Nomina im indogermanischen Lexikon. Heidelberg: Winter, 2008.
Otr^bski 1965 - Otr^bski J. Gramatika j^zyka litewskiego. T. II. Nauka o budowie wyrazow. Warszawa, 1965.
Ramovs 1927 - Ramovs F. O naravi psl. tort- in tert- v praslovenscini // Casopis za slovenski jezik, knjizevnost in zgodovino. 1927. T. 6. P. 2226.
Rav. - Ravenna Cosmography. Ravennatis Anonymi Cosmographia et Gudonis Geographica, Vol. II of Itineraria Romana / J. Schetz (ed.). Stuttgart (reprint), 1990. P. 1-110.
Rasmussen 1989 - Rasmussen J. E. Studien zur Morphophonemik der indogermanischen Grundsprache. Innsbruck, 1989 (= Innsbrucker Beiträge zur Sprachwissenschaft 55).
Risch 1974 - Risch E. Wortbildung der homerischen Sprache. Berlin; New
York: Walter de Gruyter, 1974. Rix 19922 - Rix H. Historische Grammatik des Griechischen. Darmstadt, 19922.
Rosamani 1990 - Rosamani E. Vocabolario giuliano: dei dialetti parlati nella Venezia Giulia, in Istria, in Dalmazia, a Grado e nel Monfalconese. Triest, 1990. Schmid 1969 - Schmid W. P. Zur Geschichte des Formans *-äuon-/-äuo-
/-ä // Indogermanische Forschungen. 1969. Bd. 74. S. 126-138. Schindler 1975 - Schindler J. Armenisch erkn, griechisch oSwq, irisch idu
// Zeitschrift für vergleichende Sprachforschung. Bd. 89/1. S. 53-65. Schindler 1976 - Schindler J. On the Greek type innsuq // A. Morpurgo Davies, W. Meid (eds.). Studies in Greek, Italic, and Indo-European Linguistics offered to Leonard R. Palmer on the occasion of his seventieth birthday. Innsbruck, 1976. P. 349-352. Schwyzer 1977 - Schwyzer E. Griechische Grammatik I: Allgemeiner Teil,
Lautlehre, Wortbildung, Flexion. München, 1939, 19775. Stüber 2002 - Stüber K. Die primären s-Stämme des Indogermanischen.
Wiesbaden: Dr. Ludwig Reichert Verlag, 2002. SVI I-II - Bezlaj F. Slovenska vodna imena. T. I-II. Ljubljana: SAZU, 1956-1961.
Simunovic 2013 - Simunovic P. Predanticki toponimi u danasnjoj (i povi-jesnoj) Hrvatskoj // Folia onomastica Croatica. 2013. T. 22. P. 147-214. Ziegler 2003 - Ziegler S. Bemerkungen zum keltsichen Toponym Abnova/Abnoba // Historische Sprachforschung. 2003. Bd. 116/2. S. 290-294.
Vedaldi Iasbez - Vedaldi Iasbez V. La Venetia orientale e l'Histria. Le fonti letterarie greche e latine fino alla caduta dell'Impero Romano d'Occidente. Rome: Edizioni Quasar, 1994. Vine 2011 - Vine B. On dissimilatory r-loss in Greek // Th. Kirsch, Th. Lindner (eds.). Indogermanistik und Linguistik im Dialog. Akten der XIII. Fachtagung der Indogermanischen Gesellschaft vom 21. bis 27. September 2008 in Salzburg. Wiesbaden, 2011. S. 1-17. Wackernagel 1897 - Wackernagel J. Vermischte Beiträge zur griechischen Sprachkunde, Basel, 1897.
L. Repansek. Flavias Argao, quis in sinum Argo fluit?
Careful morphological examination of a possible possessive relationship between the peculiar river-name Argaone in the Ravenna Cosmography and the name of the gulf in which it flows, attested as Largon(e) no sooner than in 16th century sources, is employed as an excuse to investigate the hydronym's ancient word-formation^ in the context of the ubiquitous and problematic suffixal conglomerate *-auon- from the wider Indo-European perspective.
Key-words: Argaone, Largon(e), toponymy, word-formation.
