Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
FIVE CELTIC TOWN-NAMES IN CENTRAL SPAIN Leonard A . Curchin University of Waterloo, Ontario Se proponen etimologías célticas para explicar los antiguos topónimos Attacum, Bravum, Centobriga, Ocelodurum y Sentica. The author proposes Celtic etymologies for the ancient town-names Attacum, Bravum, Centobriga, Ocelodurum and Sentica. The central plateau of the Iberian Peninsula boasted in antiquity many towns with Celtic names. Some of these, such as Segobriga and Uxama, have a transparent Celtic etymology; in less obvious cases, consultation of Holder' s treasury (which must be used with caution) will usually reveal roots and parallels 1 . But there are others that have not been satisfactorily explained: 1. A TTA CUM. Forrns: A ttacon (Ptol. 2.6.57), A ttacc(ensis) (CIL II 4189). Holder lists the name without offering an etymology; Schulten calls it Iberian, from the same root as the river A tax (Aude) in southern Gau1 2. However, the Celtic personal name A tto/A tta occurs frequently in the epigraphy of Gaul and central Spain3, and the suffix -acum (Latinized form of Gaulish *-acon; the equivalent neuter I A. Holder, A lt-Celtischer Sprachschatz, 3 vols. (Leipzig 1896-1907), hereafter cited as "Holder". 45 HABIS 27 (1996) 45-47 LEONARD A. CURCHIN ending in Hispano-Celtic would be *-acom) is common in Celtic place-names in Gaul and Britain. *A tt-acom therefore appears to be a Hispano-Celtic toponym formed from a personal name. 2. BRA V UM. Form: Brauon (Ptol. 2.6.51). Holder' s gloss "Bovon?" is unnecessary 4 . Montenegro rightly points out that the root cannot be Latin, but his suggested parallel, Greek brabeion "victory prize", is fantastic 5 . Rivet and Smith derive the British toponym Bravoniacum from *bravon, *braun "quern", alluding either to a quernstone quarry or to a place full of quern-shaped stones 6 . This should also be the root of Ptolemy' s Brauon (Hispano-Celtic *Bravom). The name might equally refer to a quern-shaped hill on which the town was situated. 3. CENTOBRIGA . Form: Centobriga (Val. Max. 5.1.5). Celtic *cento/*canto- means "hundred" 7 , but "hundred-fortress" makes little sense. Nor is Schulten's explanation "Burg des Cento" very enlightening 8 : the name Cento is unknown in Spain. However, there is another Celtic root *cento-l*cinto-, meaning "first" (Irish cétne, Breton kent, Welsh cynt, Gaulish names in Cintu-), from 1E *ken- "arise, begin" 9 . Thus Centobriga ought to mean "foremost fortress". 4. OCELODURUM. Forms: Octoduron (Ptol. 2.6.49), Ocelo Duri (kin. A nt. 434.6, 439.10), Ocelodorum (Rav. 319.4). Schulten, followed by Tovar, retains Ptolemy's reading, on the grounds that *octo- is good Celtic m . However, the element *ocelo-, meaning "headland, promontory, spur" occurs frequently in Celtic toponymy, e.g. A launocelum, lulocelum, Cinctocelum". Ocelus is also attested as an indigenous theonym in Britain and Lower Germany, syncretized with Mars u . Moreover, though octo- may be Celtic, it is difficult to see how it could change to ocelo- in classical sources i3 ; the likelier corruption would be for Celtic 2 Holder vol. I, col. 274; A. Schulten, "Attacum", RE 11,2 (1896) col. 2153. 3 GIL XIII, Indices p. 26; J. Untermann, Elementos de un atlas antroponímico de la Hispania antigua (Madrid 1965) 63-64. 4 Holder vol. I, col. 622. 5 A. Montenegro Duque, Historia de Burgos, 1 (Burgos 1985) 246. 6 A. L. F. Rivet and C. Smith, The Place-names of Roman Britain (London 1979) 275-76. They do not cite our Spanish example. 7 Irish cét, Welsh cant, Breton kant, Hispano-Celtic cantom (Botorrita tablet, line A4), cf. Latin centum. 8 A. Schulten, Geschichte von Numantia (Münich 1933) 67. 9 9 H. Lewis and H. Pedersen, A Concise Comparative Celtic Grammar, 3rd ed. (Güttingen 1974) 192; D.E. Evans, Gaulish Personal Names (Oxford 1967) 179. 10 A. Schulten, "Octodurum", RE XVI1,2 (1937) col. 1868; A. Tovar, lberische Landeskunde 11,3 (Baden-Baden 1989) 358. 11 Rivet and Smith, Place-names 246. 12 Holder, vol. II, col. 827. 13 There is a legitimate Octodurum in Switzerland, but in that case the sources (Caesar, Pliny, Ptolemy, hin. A nt., Rav., Tab. Peut., Not. Gall.) unanimously read "Octo-". Recollection of this Octodurum may have misled Ptolemy (or his copyists) in the spelling of the Spanish toponym. 46 FIVE CELT1C TOWN-NAMES IN CENTRAL SPAIN ocelo- to become octo- by attraction to the Greco-Roman numeral. But while Ocelo- is to be preferred, I cannot support Cortés' view that Ptolemy's reading is a corruption of Ocelo Duri 14. That form appears only in the Antonine Itinerary and seems to reflect popular etymology, deriving the toponym from the river Durius (some 19th century scholars went so far as to read Ocellum Duri, as if the name were completely Latin 15 !). Both Ptolemy and the Ravenna cosmographer suggest that the correct ending is -durum (*duro- "fort"), attested as a Celtic suffix in such toponyms as Boiodurum and Sorviodurum in Upper Germany, Brivodurum, Divodurum and Icciodurum in Gaul, and Mutudurum in Celtiberia (-um being a Latinized spelling of Gaulish -on, Hispano-Celtic -om) 16. The name should therefore be understood as *Ocelo-durom, "promontory fort". 5. SENTICA . Form: invariably Sentice (Ptol. 2.6.49; Itin. A nt. 434.3; Rav. 319.8), but final -Ti in Ptolemy can represent a (e.g. Cordube in 2.4.9), and in the Antonine Itinerary the next line is "Salmantice", i.e. Salmantica in the locative; so the nominative form is pilobably Sentica. Holder relates this toponym to Old Irish setig "female companion" 17 ; one could also consider Celtic *seno- "old", whence Old Irish sentu "age". However, the likeliest derivation is Celtic *sento- "path" (seen in toponyms Clau-sentum, Gabro-sentum; cf. French sentier) from IE *sento"go, travel; way", plus the toponymic suffix -ica as in British A esica,etc. (cf. Old Irish -ech, Breton -ek) 18. Thus Sentica should mean "trail place", i.e. road station, as befits its appearance in the itineraries. 14 M. Cortés y López, Diccionario geográfico-histórico de la España antigua, I (Madrid 1835) 225, followed by Holder and others (cf. Tovar, loc. cit.). 15 E.g. P. Madoz, Diccionario geográfico, histórico y estadístico de España, XVI (Madrid 1850) 501; C. Fernández Duro, Memorias históricas de la ciudad de Zamora, I (Madrid 1882) 112. 16 J. Whatmough, The Dialects of A ncient Gaul (Cambridge 1970) 563 (s. v. -doro-); A. Schulten, "Mutudurum", RE XVI,! (1933) col. 979. On -otn as the neuter nominative singular ending of o-stem nouns in Hispano-Celtic, see J.F. Eska, Towards an Interpretation of the Hispano-Celtic Inscription of Botorrita (Innsbruck 1989) 160. 17 Holder, vol. II, col. 1501. 18 Rivet and Smith, Place-names 309 (sento-), 242 (-ica); on the former, cf. J. Pokomy, Indogermanisches etymologisches W arterbuch (Bern/München 1959) 908. The nominative *Gabrosentum is reconstructed by Rivet and Smith from attested forms Gabrosenti, Gabrocentio. On -icol-ica toponyms in Spain, see U. Schmoll, Die Sprachen der vorkeltischen Indogermanen Hispaniens und das Keltiberische (Wiesbaden 1959) 53; and on the use of this suffix in Hispano-Celtic ethnic adjectives, cf. Eska, Towards an Interpretation, 156-57. 47