2nd proofs PAGE PROOFS © JOHN BENJAMINS PUBLISHING COMPANY https://doi.org/10.1

2nd proofs PAGE PROOFS © JOHN BENJAMINS PUBLISHING COMPANY https://doi.org/10.1075la.252.26tru © 2018 John Benjamins Publishing Company Chapter 26 Some Celto-Albanian isoglosses and their implications John Trumper University of Calabria Çabej (1969) originally posited a small number of Albanian-Celtic-Germanic isoglosses. Apart from a more detailed discussion of besa and njerí, we highlight in greater depth the areal diffusion of the I-E diffusion of Albanoid bërrakë, e blertë, brī and dritë. We add in-depth observations on the Celto-Albanian binomial ardracht (Old Irish) – dritë (Albanoid), where the Celtic terms involved are traceable to Gaulish dercos and uodercos of texts and inscriptions. Such isoglosses are developed here and tend to substantiate Hamp’s view that the Albanoid Urheimat was originally Central-North Europe rather than its present day Mediterranean Heimat. Keywords: Albanian, Celtic, Germanic, lexical isoglosses 1. Celto-Albanian isoglosses Çabej (1969) is perhaps the first systematic attempt to come to grips with the prob- lem of possible Germano-Albanian and Celto-Albanian isoglosses. The solution of these would help to establish an Urheimat for Proto-Albanoid groups in Central Europe, north of their actual territory. Originally, Jokl (1923) had underpinned isoglosses between Albanoid and Baltic languages, Greek and Armenian, but had overlooked possible relations with Celtic and Germanic. The only case not dealt with by Çabej in those discussed below is that of Bri, which he had hoped to re- solve in a later moment (Çabej 1969: 186).1 Huld (1984) criticized these first stud- ies, rejecting all the lexical isoglosses except drekë, dele, and shpen[d], overall a rather facile criticism, for motives I shall try to elaborate. Trumper (1999, 2002) 1. “Hoffe ich an anderer Stelle zu handeln” . 2nd proofs PAGE PROOFS © JOHN BENJAMINS PUBLISHING COMPANY 380 John Trumper had suggested other cases, such as besa, a central topic in Albanian culture.2 There is an obvious connection with Old Irish bés[s], of similar meaning, < Old Irish or Late Gaulish bessu (Virgil Grammaticus). The problem here is that of a possible double origin, either from *BEND-TUH- ‘binding’ or from *BEID-TUH- ‘per- suasion’ , ‘confidence’ etc., so that Pokorny in IEW 117 and IEW 127 (1959) gave a double listing for Albanian besë –a, without treating Celtic cognates. Both Old Irish béss and Middle Breton boas (Ernault 1895: 72) are comparable outcomes.3 In fact both Hamp’s considerations and Çabej (1982: 110) rightly conclude that a reasoned choice between *BEND-TUH- and *BEID-TUH- is not possible, even though Meyer (1891: 33) had tried to argue in favour of the first, Huld (1984: 143) in favour of the second (< *BID-TIH- a reduction of *BEID-TIH-), without much logic or success. LEIA B−43 seemed to think that all British Celtic forms were incomparable (“les formes brittoniques sont en tout cas aberrantes”), though only Welsh moes seems to have the wrong vocalic outcome (it seems to suggest a Proto-Celtic *BAISS- not *BEISS-). Whichever of the two possible etyma is ap- propriate, it cannot be denied that there is an undoubted relationship between Old Irish béss, Middle Breton boas and Albanian besë. More doubtful is the relation of Welsh moes with all these, notwithstanding GPC’s comments, probably based on Stokes-Bezzenberger (1894: 174)‘s original hypothesis (base *bêssu- as in Virgilius Grammaticus > Irish bés, “cymrisch moes aus *boes” etc.). Pisani (1959) had also tried to establish more lexical relations between Albanian and North Europe Indo- European, even if occasionally ignoring macroscopic cases like njer, njerí, forget- ting Celtic outcomes of the IE base *HaNER- (rather more than those usually con- sidered). Apart from the cases in the poetical sections of the Llyfr Coch Hergest quoted for the early thirteenth century in GPC 2571A-B, we find the Middle Welsh couplet ener/ ner, in the same period, in the poetry of the bard Cynddelw 16. 206, as coupled ener/ ner in 26. 40–41,4 in the mid twelve hundreds in the bardic po- etry of Meilyr’s sons Gwalchmai (9. 120) and Elidir Sais (17. 46), of Gwalchmai’s son Einion (28. 40), and of the bard Dafydd Benfras (26. 49) “O wythlonder, ner, nerth a gedwy” (in fierceness a prince, [your] strength holds still strong), (27. 28) 2. It is present in the first Italo-Albanian texts, e.g. Matranga’s Dottrina Cristiana 1592, folio 30 v. 8: see Sciambra (1964: 58). 3. Irish bés could well be from an earlier *beiss, Breton boas could be the outcome of either *beiss or baiss (Proto-Celtic *ei and *!i < *ai merge in Breton and Cornish but not in Welsh, which presents *ei > ‘wy’ , *ai > *!i > ‘oe’). 4. 26. 40–41 (page 327) “Gwas a las o Leissyaun ener/ Llary Ennia6n, llyw llwyrdaun llwr6 ner” (A youthful prince who slew the Lleision, / generous Einion, a leader, perfect his gifts as prince). 2nd proofs PAGE PROOFS © JOHN BENJAMINS PUBLISHING COMPANY Chapter 26. Some Celto-Albanian isoglosses and their implications 381 “Och am ner, muner mynogrwydd” (lament for a prince, a noble lord),5 in the mediaeval Welsh translation of the Parvum officium beatæ Mariæ V. (Gwasanaeth Meir, matins 10. 266), up until Dafydd ap Gwilym in the thirteen hundreds (poem 5. 17), thence up to modern usage.7 It would not appear to occur in Old Welsh, though the Old Welsh section of the Canu Taliesin (Williams’s edition) VIII. 3 has “gweleis i rac neb nym gweles/ pop annwyl. ef diwyl y neges” (neb for nep would be strange in Old Welsh!8), which would have more sense if the reading were “gweleis i rac ner nym gweles/ pop annwyl. ef diwyl y neges” . One might also add, from the Old Welsh sections of the Aneirin, Williams’ corrective reading of line 381 as budic ener enhy (victorious [was] a brave prince) instead of the incomprehen- sible reading budic e ren eny of Evans’ diplomatic edition. Thus the Old Welsh use of Ner/ Ener is directly comparable with Old Irish nár ‘generous, magnanimous’ as princely attributes (with secondary lengthening), and the metaphorical use as ner ‘wild boar’ (Gamkrelidze-Ivanov 1995: 158, 703; Mallory & Adams 1997: 366). The comments in Loth (1924, 207–208) on Ner and its IE relations, adding Breton nerein “fortify” (Vannetais dialect), still hold good.9 The E- in Ener, like Greek Ἀ- in ἀνήρ, supports the hypothesis of a word-initial laryngeal, the whole an areality that links Albanian njerí not only with Greek and Armenian (as well as Indo- Iranic) but also with Celtic congeners. Central cases to the argument, apart from Albanoid outcomes of *BEID-TU−/ *BEND-TU- and HaNER-, which I propose discussing, are the following. 5. See also 29. 24, 29. 100, 31. 22, 35. 4 etc. 6. This seems to be the first religious use of Ner as Lord = God (literally Nef Ner ‘Lord of Heaven’). 7. Up to the use of Ener ‘Lord’ in post-1600 Welsh Protestant hymnology. 8. Williams is following Evans (1910)‘s diplomatic version, v. 18–19, page 62 “g6eleiʃ i rac neb nym gweles/ pop ann6yl, ef diwyl y neges” . He admits, in his note, however, that in this orthogra- phy ‘p’ and ‘r’ are difficult to distinguish and concludes “barnaf mai ner oedd y gwreiddiol yma, ‘o flaen (yr) arglwydd’ …” (I judge that ner was the original [reading] here: ‘in front of/ before the prince’), comparing the sense with line 14 of the same poem where we find ‘glyw’ , lord (“yscaw- ydawr y rac glyw gloyw glas gwen”: shield before a splendid lord with [his] grey smile). 9. Fleuriot (1964: 266) had already linked Welsh and Breton forms with Irish ner, near, while Schrijver (1995: 350) had already postulated the word-initial laryngeal of H2NER-, H2NER-TO- for the Celtic outcomes. 2nd proofs PAGE PROOFS © JOHN BENJAMINS PUBLISHING COMPANY 382 John Trumper 2. Bërrakë -a ‘bog’ Bërrakë -a ‘bog’ (Meyer 1891: 39), Italo-Albanian Birrakë (Giordano 1963: 31, 33; Raimundi 2001: 271; Baffa 2009: 23). Meyer (1891) proposed as origin the IE colour adjective base *BLEI-DRO-, from IEW 155–156. Huld (1984: 45–46) ac- cepted this hypothesis, which I would exclude on semantic grounds. I would also exclude Orel (1998: 23)‘s derivation bërrakë < birë ‘cavity’ both on formal (pho- nological) and semantic grounds. Çabej (1969: 176) and Çabej (1976: 215–217; 1981: 214, 254; 1987: 232) offered two hypotheses: (1) a Turkish loan (< bare, bi- ral), (2) a Western IE isogloss, starting from Hesychius’ βραγός· ἕλος, of Galatian origin. He considered similar Slav terms congeners. Skok 1. 109–110 thought the Albanian, as well as late Greek μπάρα, borrowings from Slav (Meyer’s first hy- pothesis), which he proposed as co-radicals, without the theme’s being doubled, of Sanskrit barburá and Greek βόρβορος < IEW 482 *GwOR(-GwOR-). Had Skok been on the right track, then one would have to suppose Southern Slav conge- ners to have been Greek borrowings (< βόρβορος). Were one to follow the Turkish hypothesis, the only solution would be in terms of Starostin- Dybo- Mudrak 2. 904–05 Proto-Altaic *MĀRO- > Proto-Turkish *BOR- ‘marsh; muddy terrain’ etc. However, an –AK- formant would seem decidedly more IE than Altaic. In IE terms *MRK- > *MERK-, *MREK- (Pokorny 739–740) (1) ‘rot; putrefy’ , (2) ‘putrid uploads/Litterature/ trumper-pdf.pdf

Tags
littérature proofs irish celtic welsh albanian
Documents similaires
  • 22
  • 0
  • 0
Afficher les détails des licences
Licence et utilisation
Gratuit pour un usage personnel Attribution requise
Partager