Post by Teuta1975 on Aug 24, 2023 15:56:46 GMT -5
VNIVERSITAS IAGELLONICA
FACVLTAS PHILOLOGICA
LE INDOEVROPAEA CRACOVIENSIA
IOANNIS SAFAREWICZ
MEMORIAE DICATA
Some remarks on the Albanian vocabulary
in the Palaeo-Balkanic and Indo-European view
IGNACY RYSZARD D ANK A
AND
KRZYSZTOF TOMASZ WITCZAK
JAN SAFAREWICZ (1 904 - 1992), professor of Indo European linguistics at the Jagellonian University (from 1937 onward) and honorary doctor of the universities of Vilnius and Cracow, was the most outstanding Polish scholar working in the domain of Latin and Greek linguis tics, which he studied from the historical and comparative point of view. He left behind a few hundred publications, including several monographs and text-books for students of Classical philology. In addition to the Classical languages, he also investigated Lithuanian and the Slavonic languages.
This volume contains the papers read at a conference held at Cracow in 1993 to...commemorate the first anniversary of the scholar's death. Outstanding Polish linguists as well as specialists in the domain of Inda-European scholarship from different European countries are among the contributors. The papers deal with a wide variety of subjects, covering almost the whole Inda-European language group, from Hittite to Slavonic and Lithuanian. Some authors tackle general and methodological issues. To conclude with, the volume contains a few biographical contributions on Jan Safarewicz as well as a full bibliography of his scholarly output.
1. THE lNDO-EUROPEAN DIALECTOLOGY: LEXICAL EVIDENCE FOR THE PALAEO DALKANIC LANGUAGE LEAGUE
We believe that the primary Indo-European dialects can be di vided into five language groups1 :
(a) northern, i.e. Germanic-Balto-Slavic;
1 T he division of the Indo-European languages suggested here generally follows the classification presented by Danka (1983, 94-114). An innovation is separating the eastern (Indo-Iranian) group from the central group, which is based on a close cognation of t he Indic, Da rdic, I<afir and Iranian languages (expressed in striking phonetic, morphological, lexical, mythological-religious and other peculiarities,
\
Analect a Indoeuropaea Cracoviensia
I. Safare wicz memoriae dicata Ed. W. Smoczynski
Cracoviae: Univcrsitas 1995, pp. 123- 135.
IGNAC Y R.YSZARD DANKA, I<R ZYSZ T OF TOMASZ WITCZAK
(b) western, i.e. Italo-Celtic;
(c) southern, i.e . Anat olian;
(d) eastern, i.e. Indo-Ir anian;
(e) central, i.e. Palaeo-Ba.lkanic 2 .
This last group included languages existing on the Balkan Peninsula (e.g. Thracian, Dacian, Illyrian, or Macedonian) or in areas not far away from there (e.g. Messapic in southern Italy; Phrygian in Anatolia; Armenian in trans-Caucasian regions) in ancient times. From among the Palaeo-Balkanic languages only three a.re still used: Albanian, Armenian and Greek. The others have become irrevocably extinct, leaving, apart from glosses and onomastics, some inscriptional texts ( esp. Phrygian and Messa.pie, to a smaller extent Thracian and Mysian) or no written traces at all (e.g. 11acedonian , Illyrian , Epirotic, or Paeonian).
The central (or Palaeo-Balkanic)group was relatively closely connected wit h the eastern (or Inda-Iranian) one, but it has none of the specific innovations of the eastern group such as MONOVOCALISM (i.e. IE *e * a *o > Indo-Iranian *a ), PER MANE NT SATEM QUALITY, COMMON ETHNIC TERMINOLOGY (* A rya-), SPECIFIC RELIGIOUS VOCABULARY (e.g.
rta- ' cosmic order', the gods Asura-, Mitra-, Nasatiya-, Indra-) and
other features (Burrow 1976, 17-22). On the other hand, the Palaeo - Balkanic group is divided from the eastern one by phonetic archaisms (e.g. preserving the so-called 'prothetic vowels' that have completely disappeared in the eastern, northern and western groups, and have only partly survived in the archaic southern group, see vVitczak 1993) and by lexical correspondences such as3 :
(A) *andheyeti ' blossoms, blooms': Alb. endem 'I blossom, bloom'
II Gk. av{)iw 'id.' II Arm. andem 'I culti vat e';
see Burrow 1976, 7- 36) on the one hand, and on distinct similariti es within the Palaeo-Balkanic group suggesting the existence of a. primeval language league (resembling the present Balkanic league) on the other, cf. Bednarczuk 1986, 510- 511.
2 For the terms "Palaeo-Balkanic languages" and "Pa.la e o -Balkanic philology", see Neroznak 1978, 7- 20.
3 Below we list a number of isoglosses that appear exclusively in the Paleo
-Balkanic area and that are attested at least in three Paleo languages.
Some remra ks on the Albanian vocabulary 125
(B) *argel ya (f.) 'hut or underground dwelling': Alb. ragcil (f .) ' hut' II Rum. argea (f.) 'underground dwelling' < Dacian II Maced. ap,i>.>.a ' vapour -bath' I Kimmer . a p, D,) ,a i (f. pl.) 'underground
dwellings ' (cf. Neroznak 1978, 190);
(C) * autos 'the same:' Alb. vete 'id.' II NP hr . AYTOE, OP hr .
autaj (<lat . sg . fem.) I Gk. av r 6<;, Cret. a f r o<; II Mess. atavetes 'in the same year', cf. Gk. avr o£r i <; ad v. ' id.';
(D) *d2 gherdos/*d2 ghfd-s (f.) '(wild) pear-tree': Alb. dardhe (f.) ' pear-t ree' II Gk. axcpOo<; (f.) and axpa<; ( - a oo<; ) 'wild pear-tree, Pyrus amygdaliformis' II Maced. a1 i p5a (f.) 'common pear-tree,
Pyrus comn1unis' (cf. Huld 1984, 48);
(E) *;J2i fl-s / ;J2i §i- (f.) 'goat': Alb. edh 'kid, young goat' (alternatively from Latin haedus 'id .', cf. Rum. ied) II Gk. att, al,6<; (f.) 'goat' II Arm. aye (i-stem) 'she-goat ' (Huld 1984, 48);
(F) * biik]>o s 'vulva': Alb. bythe 'bottom, arse' II Gk. (Hes.)
f]vrro<; 'vulva' II Arm. puc'. 'id .' (Mann 1984-87, 58);
(G) * clesm6 s (m.) 'fat ': Alb. dhjame 'fat, lard' II Gk. Att. 817µ6<;
m. 'fat' II Arm . tam-uk' 'moisten' (Neroznak 1978, 182; Huld 1984,
60);
(H) *drepano- 'sickle': Alb. Tosk. draper, Gheg. drapen 'sickle'
II Gk . 5pbravov (n.) and 8p£1r6.vr; (f.) 'id.' II Arm. artewan1mk'
(Lambert erie 1983, 21- 22);
(I) *dhdl yeti 'blooms, sprouts' and *dhaler6s adj . 'springing forth, fresh': Alb. clal 'I bloom, go out' (Aor. clola) II Gk. 196.>..>..w 'I bloom,
sprout' (Aor. I k1J17>..a, perf. ri1J11>..a, Dor. Aeol. ri-i?a.>..a), ad j.
19a>..E.p6<; 'blooming, fresh, moist' II Arm. dalar adj . 'green, fresh'
(Hamp 1984; Muller 1984, 100);
(J) *dh rftbhos (m.) ' tomb': Gk. r6'cpo<; (m.) 'id.' II R oum. climb,
do.mb ' h Hlo ck. hill, rising round' < Da.cian II Arm . damban 'tomb' II
Gk. "Pelasgiau" TVf.l/30<; ( m . ) 'tomb, grave' (cf. Pokorny 1959, 248; Mann 1984, 87, 193);
(K) *d 1 dusna (f.) 'pain, ache, grief': Alb. dhune £. 'pah1, ache, grief, suffering, shame , disgrace, affront, violence' I Gk. Aeol. tovva (f.), At t . Ion. oo{;vr; (f .) 'gri ef' II Arm. erkan 'birth-pains; toil, ]abour
in childbirth' (Mann 1984-87, 170);
lGNACY RYSZARD DANKA, KRZYS Z'l'OF T OMAS Z WITCZAK
(L) *a1 rino- ( m . andn.) 'cloud, mist': Alb. Gheg. re (OGheg ren)
'cloud' II Gk. (Hes.) £plvov (n.) 'cloud' II illyr. pivoc; (m .) 'mist' II
1V[ess. p!.vo11 (n.) 'cloud' (Hamp 1957, 80; Neroznak 1978, 163);
(M) *kak6s adj. ' bad': Alb. keq adj. 'bad, evil' II Gk. 1,,0,1,,6c; adj. 'bad' II NPhr. 1,,0,K,ovv (n.) 'harm ' (see Huld 1984, 79- 80);
(N) *k6wilos adj. 'hollow, empty': Alb. thelle 'deep, dark' (as to colours) II Gk. 1,,oZ>.oc; a dj. 'hollow', Myc. ko-wi-ro II Arm. soyl ' cave'
(Huld 1978, 297-299; 1984, 118);
(0) *oler6- (adj.) 'impure, turbid ', (n.) 'dirt, mud, excrements': Alb. lere 'dirt (on the body)' II Gk. o>.£p6c; adj. 'impure, turbid' II
Maced. a>.ipov (n.) 'excrements' (Neroznak 1978, 180);
(P) *0(1.t )-k JJe: '(and) not': Alb. as ' not ' II Gk. ovr£ 'and not' II
NPhr. v - 1,,£? II Arm. oc' ' not';
(R) *a3k'1"-ya 1 (dual) 'two eyes': Alb. sy 'eye(s)' II Gk. ocHJ£ (du.) 'two eyes' II Arm. ac1c' 'e yes';
(S) *;)3ner yos (m.) 'dream': AJb. Task. ender, Gheg. ader 'id.'
II Gk. Att. Ion. OV£lpoc;, Aeol. ovolpoc;, Dor. avo.lpoc; II Arm. anur.f
(o-stem);
(T) *skorodom (n.) 'garlic, Allium sativum': Alb. hurdhe (f.) and
hudher, hudhre (f.) 'id.' II Gk. <71,,6p(o)6ov (n.) II Arm. xstor, NArm.
dial. sxtor 'garlic' (Djahukian 1967, 220).
2. NOTES ON THE SEMANTIC CHANGES IN Tiffi lNDO-EUROPEAN ARBOREAL TERMINOLOGY
This paper aims at discussing PaJaeo-Balkanic and Albanian arbo real terminology with special attention paid to semantic innovations in this :fiel d. Already at the very beginning of Indo-European studies as a scientific discipline, it was noticed that semantic innovations, usually motivated by observable extralinguistic factors, are of particular importance for comparative studies. It has been known for a long time that such words as Greek <.p171 6c; (f.) 'oak' and Latin fagus (f.)
'beech', although stemming from the uniform original Inda-European
etymon *bhag6s (feminine o-stem), differ considerably in their mean ing. Usually it is explained by the fact that beech is very rare in Greece and thus the original name has been transferred to another
Some remarks on the Albanian vocabulary 127
tree, narnely to oak, which 1s quite common there (Walde and Poko rny 1927, 128). An explanation for a similar correspondence between Latin fraxinus (f.) 'ash' and Sanskrit bhu1ja (m.) 'a kind of birch', Ossetic barz 'birch', Pol. brzoza 'id.', etc. (all from IE. *bhrH§- or bherH§ -, orig. 'birch') is that birch is hardly to be found on the Ital ian Peninsula and thus the Romans could have easily transferred its original name to ash, which also has light bark (see Friedrich 1970, 26- 31, esp. 29).
The triple semantic change ' a conifer' > ' a kind of willow' is well documented in the Slavic material, namely:
(1) Russ. dial. bred 'wicker', bred 'Salix caprea', bredina 'wil low, Salix', Ukr . bredina 'Salix caprea', Slovin. bfaza (f.) 'fruit-tree'
< Proto-Slavic *bred5, *bredin a, * bredja (f.) < IE. *bhroid( h)os,
*bhroid{h)ya (f.) 'pine', cf. Alb. bredh 'pine', also 'spruce' and 'larch', Roum. brad 'pine, spruce' (both from IE. *bhroid[h}os ); Lat vjan priede 'pine', dial. 'spruce' (from Baltic *braidja f. with initial p- by a cross with the synonymous Baltic word *peuse f. 'pine'), also pdeds f. 'jd.';
(2) Russ. iva 'willow', Pol. iwa, SC. 'iwa, et c. < PSI. *jiva f. 'a kind of willow' < IE . *twos (f.) 'yew', cf. Gaulish ivos, Fr. if, Welsh yw m. (sg. ywen), Bret. ivin m. (sg. ivinenn), Corn. ew, OCor n. (h)ivin 'yew'; OPruss. iuwis 'yew'; ONord . yr m. 'yew, yew-wood bow\ OE. iv, eov, E. yew, OHG. iwa, G. Eibe, MDu. iwe, Du. ijf 'yew';
(3) Cz. rokyta, dial. also rakyta, Pol. rokita, Russ., Blg., SC. rakita, etc.< PSI. *orkyta (f.) 'a kind of willow, brittle willow'< IE. *arkutos (f.) 'juniper', cf. Gk. o:.pK,cv'l3o<; f. 'juniper or cedar, Iuniperus'; Eteo Cretan &p,croc; 'id.'; Latvian ercis, also ecis ' juniper' .
Three observed semantic changes in Common Slavic are convincingly explained by the fact that the early Slavs, when moving from mountainous areas to plains rich in water, transferred the names of conifers (so numerous in the mountains) to the newly found kinds of willows.
In the Palaeo-Balkanic area, the arboreal terminology has also undergone characteristic semantic changes, in which Albanian participates to a smaller or a greater extent. Therefore, it is enough to examine, firstly, in which pan-Balkanic changes Albanian participates and in which it does not, secondly, in which cases it retains the original n1eaning and in which it does not, and thirdly, which tree terms it borrows and which it does not, to gather often invaluable information on the prehistory of Albanian in the Palaeo-Balkanic and ludo-European contexts.
3. 'BEECH' IN lNDO-EUROPEAN AND PALAEO-BALKANIC
Within the Palaeo-Balkanic languages, one can see two characteristic semantic changes concerning the arboreal terminology:
(a) The old appellative for 'beech' (IE. *bhag6 s f.) is used in the Palaeo-Balkanic area as a na111e for 'oak', cf. Greek <p'Tf,6c;, Dor. cpii 1 6c; (f.) 'Valonia oak, Quercus aegilops', Albanian bunge (f.) 'a kind of oak, Quercus esculentis' (from * bhiigna 4, see Huld 1981, 306; Adams 1985, 82), Phrygian Ba,a.Zoc; 'oak-god' (epitheton of the sky
-god, cf. Greek (Zcvc;) ?Pa,wva'ioc;, Umbrian (Iuppater) G,abovius'oak-god '5) .
The reason for this semantic change is not quite clear. Usu.ally its appearance is explained by scarcity or lack of beech. However, such an explanation is hard to accept as beech can be found almost all over the Balkans (except for the eastern and southernmost parts 6) in fair abundance, and the semantic change, as we can see, has a
4 It should be added that the Albanian word for ' Querc us esculentis', bunge, attests the presence of the velar stop *g in the Indo-European etymon * bhag6s (f.) 'beech'.
5The Indo-European sky-god *Dyiws is closely connected with the oak tree,
which used to be his sacred tree. Already in the Inda-European epoch, there appeared the epitheton *Perk'lun6s, literally 'oak-god>, which was derived from IE.
* perk'!Jv.s f. 'oak' (cf. Lat. quercus f. 'id .' ) and especially accepted in the northern
group. The divine epitheton * Perk'*iin6s (cf. ONord . Fj9rgynn, Lith. Perkilnas, OP ru s. Percunis and Slavic *Peruna/* Peryna with an irregular sound change under the influence of the verb * perq 'I strike') has supplanted the original name
of the sky-god (IE. *Dyiws or * Dyiws p.12 tir) in thenorth. There is no source
justification for the assertion that there were two different Inda-European gods having those names (i.e. * Dyiws and * Perk'lun6s) as the ranges of appearance of the two gods are disjointed and mutually complementary, which is a proof that
*Perk'Jl,un6s is only a local (i.e. northern) hypostasis of *Dyiws.
6See the map 16 in Friedrich (1970, 113).
Some remarks on the Albanian vocabulary 129
Pan Balkanic range. Probably we have to do with a non-conditioned semantic divergence that has occurred within the Palaeo-Balkanic group and has been generally accepted 7 .
(b) The semantic change 'beech, Fagus' > 'oak, Quercus' has caused an "empty place" (a free semanteme) in the microsemantic system of the particular Palaeo-Balkanjc languages. To fill it in, two independent semantic changes have taken place in the Palaeo Balkanic area:
(1) The meaning of 'beech' has been gjven to the old appellative form of 'ash, Fraxinus' (IE. *osk6s and *oskewa f.) . This process has appeared in the three basic Palaeo-Balkan languages:
Alb. ah (f.) 'beech' (< IE. *osk6s f. 'ash-tree', cf. Okel. askr m. 'ash, ashwood spear', OE. asc, etc. < Germanic *askaz m., Skt. ak$d
m. 'the tree Elaeocarpus ganitrus' (from *osk6s with the metathesis of -sk- > -ks- );
Gk. ofia and o(u17 (f.) 'beech' (from IE. *osJ..ewa f. 'asht
Arm. usi 'a deciduous tree', perhaps 'beech' (Djahukian 1967, 255), Mod. Arm. hosi 'a particula.r bush' (from IE. *oskew<i f. ' ash'). All other ludo-European stocks have retained the original mean ing, e.g. Lat. ornus 'ash', Slavic *( j)aseno 'ash' (both from PIE.
*H3osH1 enos ), see Pokorny 1959, 782.
(2) To denote 'beech' in other Palaeo-Balkanic languages, the old Indo-European etyrnon for 'elder', b(o)u§h6s (cf. Slavic *buzo beside
*bozo 'Sambucus nigra') has been used. This process can be proved for 1vi ysian (cf. µv(J6c; ' beech'8) and perhaps for some Armenian dialects (cf. Mod. Arm. piik(i in the Karabakh dialect).
7 Walde-Pokorny's explanation, cited in the proper text , was rightly abandoned by Rudnicki (1956, 128). A basis for I.he Palaeo-Balkanic shift from 'Fagus' to ' Quercus (esculent.is)' "".ould have been the edibility of both Aesculus acorns and beechnuts. They may be rendered into soap and, because they are sweet and tasty, into oils for cooking, as is in fact done in southern France. A similar explanation was suggested by Bender (1922, 30). According to him, the Greek shift took place in two stages, first from the beech to the <'native sweet chestnut that is still so characteristic of northwestern Greece", and then to the oak. "In any case, all three trees, beech, oak, and sweet chestnut , bear edible nuts, and belongs to the same family, Fagacaeae" (Bender 1922, 30). Compare also Friedrich (1970, 109).
8 See Georgiev (1981, 125). For the Mysian (or Lydian) provenance of µva6<:;
'beech', see Gusmani (1964, 276).
l GN AC Y R YS ZARD D ANK A, KRZYSZTOF TOMASZ WITCZAK
Both semantic processes discussed in points 3b (1- 2) can be convincingly applied to the period of existence of the Palaeo -Balkanic linguistic league.
4 . ' A SH-TREE' lN THE BALKANIC LANGUAGES
Further semantic changes took place independent ly from one an other, and particular Palaeo-Balkanic peoples introduced specific and divergent innovations to complement the semanteme ' ash':
(a) Albanian borrowed the termfrasher (Tosk.) / frashen (Gheg.) ' ash' from Latin fraxinus (f.) 'id.'. It is impossible to find out what term was used to describe a given tree jn the pre-Roman period;
(b) Greek adapted an old Indo-European word for ' Ostrya carpinofolia', thus Gk. µc:,>..i a (f . ) ' a sh ' continues IE. *melisa or
*melisya (f.) ' Ostrya carpinofolia' , cf. Alb. rnelleze, mellinje (f.) 'id.', Inda-Iranian *rnarisa - (the word attested in Fenno-Ugric *maris 'lime-tree': Mord. M. mar s, maras 'elm, ash'; Vot. dial. b§r jz, berji,
ber ji,- pu [pu ' tree , wood'L ber ji , berii , berzi 'lime-t ree' ; Zir. beris
'id.')9•
(c) To describe ' a.sh', Armenian used the original term for 'elder tree, Sambucus nigra L.': Arm. hac'i (gen. pl. hac'eac') 'ash' < IE.
*a2 kpewa or * d 2 k]>ewos (f . and m.) 'elder ', cf. Gk. 0:11,ria. (f.) and
cr.11,rio<; (m.) 'Sambucus nigra', Dacian <7i /3a., seva 'id .'. For further cognates, see Witczak (1992).
5 . ' E LDER-TREE' IN THE BALKANJC LANGU AGES
In the Palaeo-Balkan area, both primaeval Inda-European appellatives for 'elder-tree, Sambucus nigra', IE. * d 2kpewa and *bough6 s
(f.), have survived, but in most cases they have undergone a semantic change.
(a) The original meaning of the firnt appellative is preserved in Greek and Dacian (see above, Sect. 4c). In Armenian this appellative
9 An alternative etymology of the Fenno-Ugric words is suggested by Redei (1986, 55), who assumes a borrowing from Middle Iranian * barza- ' birch'.
Some remarks on the Albanian vocabulary 131
means 'ash-tree', whereas in Albanian and in Bessan it has undergone a deeper semantic divergence: 'Sambucus nigra' > 'Ilex aquifolium'
> 'Tussilago farfara', cf. Bessan ao-a 'colt's foot, Tussilago farfara ' (from *ao-ia and IE. * 2kpewii ), Alb. ashe, ashja ' Ilex aquifolium' beside 'Tussilago farfara.'. The semantic change is fully motivated by the similarity of sharp-pointed leaves of all these plants.
(b) In the Palaeo-Balkanic area, the alternative word for 'elder
-tree', IE. *b( o)ugh6s (f.), has been transferred to 'beech', cf. Mysian
µvo-6<; 'beech' (see Sect. 3b).
(c) Dacian has introduced an alternative word for ' Sambucus ni gra', oAµa (f.)10 , which, however, still has the original Indo-European etymon of 'elm-tree' (IE. */m os (f.), cf. Lat . ulm1.ts f. 'id.', 0Ir. lem
< Celtic *limos, Slavic * jblmo) .
(d) The native form of the Albanian appelJat ive for 'elder-tree' is not known. In the contemporary literary language, similarly as in dialects, the only term to appear is shtog 'Sambucus nigra', borrowed from Vulgar Latin *sab'ltcus < Lat. sambiicus, cf. Roum. soc ' elder
-tree'.
,6 THE ALBANIAN ARBOREAL TERMINOLOGY AS AN EVIDENCE FOR THE ORIGIN
OF THE ALBANIANS
The observations described in points 3- 5 are fairly important for determining the connexons between Albanian and ancient languages of the Balkans. It is worth noticing that Albanian and Mysian have dHferent (although native and primaeval) terms for 'beech' (cf. Alb. ah 'beech', orig. 'ash-tree' vs. Mys. µvo-6<; 'beech', orig. 'elder-tree'),
which should exclude the possibility of the Mysian origin of Albanian . Moreover, there is a great semantic and phonetic divergence bet,.,ween Dacian and Albanian as far as the ludo-European term *d2 kj:>ewii
(f.) 'elder' is concerned, which speaks against the Dacian hypothesis (cf. Dae. o-i/3a, seva 'Sambucus nigra' vs. Alb. ashe 'Ilex aquifolium; Tussilago farfara'). On the other hand, one should note the similarity
10 Neroznak (1978, 61) supposes that the Dacian word 5>..µcr is a borrowing from Latin ulmus 'elm'. Such an interpretation, however, has no semantic, phonological, or morphological justification.
IGNACY RYSZARD DANKA, KRZYSZTOF TOMASZ WITCZAK
between Alb. ashe and Bessan &a& covering both semantics (thesame im1ovation) and phonetics11 .
Albanian has kept Indo-European terms applying to particular conifers such as:
(a.) ' yew': Alb. Gheg. bershe 'stinking juniper ' (< IE.
*ebhur-isyo-, cf. Ofr. iber m. 'yew', Gaulo-Romance *eburos 'id.', MHG. eber-boum and German Eber-esche ' mountain-ash', all from IE. *ebhuros 'yew', see Pokorny 1959, 334);
(b) ' pine': Alb. bredh (<* bhroid{h}o-, cf. Roum. brad 'pine, spruce', Lat v. prieds f. and priede f. 'pine\ Russ . bred, dial. bred, bredina ' willow, Salix', Slovinian braza 'fruit-tree' (see Sect. 2);
(c) 'ftr ': Alb. hale f. 'black pine, Pinus nigra' (from IE. *adhla or
*adhlis f. 'fir, Abies', cf. OPr us. addle 'fir, spruce', .Jatvingian egle 'fir-tree' , Ljth. egle, dial. agle, Lat v. egle; OPol. jedl, jedla, Pol. .iodla 'fir', Cz. jedle, Polabian jadla, Russ. elb, dial. elina; 0 Ir . aidlen, MCorn. adlen 'Abies', Bret . aedlen, edlen 'fir';
(d) 'spruce' : Alb. pishe (< *pit-so-, cf. Gk. 'lfiTvc; f. 'pine, spruce\ Skt . pztu-dciru ( n.) 'a kind of pine, or resinous tree', see Huld 1983, 303). An alternative etymology connects the Albanian word with Latin picea f. 'spruce';
(e) 'juniper': Alb. dellinje, dellenje, delli (< IE . *(s)ghelin yo- 'ju niper', cf. Gk. ( Hes.) <JXi>..ivo<; 'wild cypress or juniper';
(f) 'larch': Alb. bre, breni 'spruce, fir' (< IE . * brin Hu- 'larch' , cf. Romance *brinu- 'larch', attested in Italian dialects: brenge, brendze (Valle dell' Orca) , brenva (Val'soana), brengola. (Canavese); Pol. dial. brzyn, brzym (Silesia), Cz. dial. brim 'larch', Sln. brin 'juniper' or 'larch, Larix meleze', SC. dial. brina (f.) ' juniper' or 'spruce, Picea ' , Sln. brina : brinja (f.) 'a conifer', usually 'juniper', dial. 'spruce' and also ' pine' .
(g) Only Alb. rrobu.ll ' Pinus leucodermus' is probably of foreign
origin, although it is not certain whether this word was borrowed from Latin ravicella (f.) 'a conifer' or not.
11 This similarity may be a linguistic confirmation of Witczak's hypothesis of the Bessan origin of the Albanian population .
Some remarks on the Albanian vocabulary .133
011 the other hand, Albanian has not kept a large number of names vJ. the deciduous trees, borrowing them from Ancient Greek in the Archaic period, from Latin at the time of Roman domination in the Balkans or from Slavs after their great migration. The following arboreal terms can be quoted here:
(h) ' ash': A]b. Tosk. frasher, Gheg. frashen < Lat. fraxinus f.
'ash', cf. Rourn. frdsin (see Sect. 4a);
(i) ' willow': Alb. shelq, -u (m.) < Lat. salix, -icis f. 'willow', cf.
Roum. scilcie f. 'willow', also 'a willow t wig ';
(j) 'poplar': Alb. verre (f.) 'silver poplar, Populus alba' < PGk.
*awiger( y)o s, cf. Greek at,E.poc; or atyElpoc; (f.) 'black poplar, Po pulus nigra';
(k) 'poplar': Alb. plep, -i (m.) < Vulgar Lat. *plopus < Lat.
popultts 'poplar', cf. RoU1Il. plop;
(1) 'poplar': Alb. topole, tupli < Slavic *topolb 'poplar' (Svane 1992, 128);
(l) ' willow': Alb. Gheg. mqite 'Salix alba' < South Slavic *rak yta
(Sva.ne 1992, 127- 128);
(m) 'fig-t ree': Alb. fik, -tt (m.) < Lat. ficus, -us or -f (f.) 'fig-t ree', cf. Aroum . hie (m.) 'id .' , hica (f.) 'fig, grape';
(n) 'bay-tree': Alb. lar, -i (m.) < Lat . laurus, -zor -us (f.) ' bay
tree\ cf. Roum. ldur (m.);
(o) 'chestnut ': Alb. geshten.ie , -a (f.) < Lat. castanea 'chestnut (tree and nut)', cf. Roum. castan (m .) ' chestnut -tree' , castana (f .) 'chestnut';
(p) 'oak': Alb. rre 'English oak, Quercus robur' < Lat. ri5bur
'id .';
(r) 'elder': Alb. shtog < Vulgar Latin sabucus < Lat. sambucus
'id.', cf. Roum. soc (see Sect. 5d).
This list shows tha.t Albanian has preserved native tenns for conifers typical of a mountain landscape and lost the terms for hydrophilous trees that grow on lowland plains along rivers, marshes and lakes such as ' willow' or ' poplar'. In this situation it is to be supposed that the forefathers of the Albanians lived in areas that were comparatively badly watered and where conifers were dominant, i.e. in typically mountainous areas.
REFERENCES
ADAMS, D. Q. 1985. "The Inda-European word for ' apple' again,,, Indogerm a nische Forschungen 90, 79- 82.
BEDNARCZUK, L. 1976. ''Indoeur opejskie j zyki Balkan6w", in L. Bednarczuk,
ed., J zyki indoeuropejskie, t. 1, Warszawa: PWN, 471- 513.
BENDER, H. H. 1922. The home of the Inclo-Europeans, Princeton.
BURROW, T H. 1976 . Sanskrit, Moskva. (In Russian; primarily published in English as The Sanskrit language, London 1959).
DANKA, I. R. 1983. Stanowisko j<;.zyk6w anatolijskich w rodzinie indoeuropejskiej
i ich wzajemne zwiqzki, Lodi: Wyd. Uniwersytetu Lodzkiego.
DJAHUKIAN, G. V. 1967. Ocerki po istorii dopis'mennogo perioda armjanskogo jazyka, Erevan.
FRIEDRICH, P. 1970. Proto-In da- European trees. The arboreal system of a p1·e
historic people, Chicago.
GEORGIEV, V. I. 1981. Introduction to the history of the lndo-European lan guages, Sofia.
GUSMAN!, R. 1964. Lydisches Worterbttch, Heidelberg: Winter .
HAMP, E. P. 1957. "Albanian and Messapic", in Studies presented to Joshua Whatmough on his sixtieth birthday, The Hague: Mouton, 73-89.
- · 1984. "Armenian dalar 'green', fJa>.t::p6<; 'moist',,, Die Sprache 30, 156-159.
HULD, M. E. 1978. "Albanian i thelle and its ea.stern relations,,, Zeitschrift Ji.fr
vergleichende S'prachforschung 92, 297-299.
- · 1981. « Alban ian corrigenda to Friedrich's Proto-Indo-European trees,,,
Zeitschrift fur vergleichende Sprachforschung 95, 302- 308.
-· 1984. Basic Albanian etymologies, Los Angeles: Slavica Publishers. LAMBERTERIE, CH. DE. 1983. "Une bagatelle etymologique ( A rtew a-
nunk' / opt::1ietv·r1),,, Revue des etudes armeniennes 17, 21-22.
MANN, S. E . 1984-1987. An Inda-European comparative dictionary, Hamburg : Buske.
MULLER, G. 1984. "Notes on four Old Armenian etymologies,,, Arbejds Papirer
(4) udsendt af Institut for Lingvistik K¢benhavns Universitet, 99-101. NEROZNAK, V. P. 1978. Paleobalkanskije jazyki, Moskva.
POKORNY, J. 1959. Indogermanisches etymologisches Worterbuch, Bern- Miinchen: Francke .
REDEI, K. 1986. Zu den indogermanisch-1.walischen Sprachkontakten, Wien.
RUDNICKI, M. 1956. "Wartosc nazw drzewa bukowego, lososia i rdzenia lendh dla wyznaczenia prakolebki (praojczyzny) indoeuropejskiej i slowianskiej", Biuletyn Polskiego Towarzystwa J zykoznawczego ]5, 127- 137.
SVANE, G. 1992. Slavische Lehnworter im Albanischen, Aarhus: University Press. WALDE, A., POKORNY, J. 1927-1932. Vergleichendes Worterbuch der indoger
manischen Sprachen, Bde. 1- 3 , Berlin- Leipzig.
WITCZAK, K. T. 1992. "Inda-European *:J2kpewa 'elder, Sambucus nigra L.' and its Baltic descendants", Linguistica Baltica l, 201- 211.
Some remarks on the Albanian vocabulary 135
WITCZAK, K. T. 1993. "T he triple representation of prothetic vowels in an cient ( u:eek" , in Studies in Greek Linguistics . Proceedings of th e 14th Annual Meeting of the Department of Linguistics, Faculty of Philosoph y, Aristotle Uni versit y of T hessaloniki , 27-29 April 1993, Thessaloniki 1993, 487- 497.
University of Lodz
Chair of Classical Philology ul. Wolczanska 90
P L- 90- 522 Lodz
Igna cy Ryszard Danka and
Krzysztof 'l'omasz Witczak
FACVLTAS PHILOLOGICA
LE INDOEVROPAEA CRACOVIENSIA
IOANNIS SAFAREWICZ
MEMORIAE DICATA
Some remarks on the Albanian vocabulary
in the Palaeo-Balkanic and Indo-European view
IGNACY RYSZARD D ANK A
AND
KRZYSZTOF TOMASZ WITCZAK
JAN SAFAREWICZ (1 904 - 1992), professor of Indo European linguistics at the Jagellonian University (from 1937 onward) and honorary doctor of the universities of Vilnius and Cracow, was the most outstanding Polish scholar working in the domain of Latin and Greek linguis tics, which he studied from the historical and comparative point of view. He left behind a few hundred publications, including several monographs and text-books for students of Classical philology. In addition to the Classical languages, he also investigated Lithuanian and the Slavonic languages.
This volume contains the papers read at a conference held at Cracow in 1993 to...commemorate the first anniversary of the scholar's death. Outstanding Polish linguists as well as specialists in the domain of Inda-European scholarship from different European countries are among the contributors. The papers deal with a wide variety of subjects, covering almost the whole Inda-European language group, from Hittite to Slavonic and Lithuanian. Some authors tackle general and methodological issues. To conclude with, the volume contains a few biographical contributions on Jan Safarewicz as well as a full bibliography of his scholarly output.
1. THE lNDO-EUROPEAN DIALECTOLOGY: LEXICAL EVIDENCE FOR THE PALAEO DALKANIC LANGUAGE LEAGUE
We believe that the primary Indo-European dialects can be di vided into five language groups1 :
(a) northern, i.e. Germanic-Balto-Slavic;
1 T he division of the Indo-European languages suggested here generally follows the classification presented by Danka (1983, 94-114). An innovation is separating the eastern (Indo-Iranian) group from the central group, which is based on a close cognation of t he Indic, Da rdic, I<afir and Iranian languages (expressed in striking phonetic, morphological, lexical, mythological-religious and other peculiarities,
\
Analect a Indoeuropaea Cracoviensia
I. Safare wicz memoriae dicata Ed. W. Smoczynski
Cracoviae: Univcrsitas 1995, pp. 123- 135.
IGNAC Y R.YSZARD DANKA, I<R ZYSZ T OF TOMASZ WITCZAK
(b) western, i.e. Italo-Celtic;
(c) southern, i.e . Anat olian;
(d) eastern, i.e. Indo-Ir anian;
(e) central, i.e. Palaeo-Ba.lkanic 2 .
This last group included languages existing on the Balkan Peninsula (e.g. Thracian, Dacian, Illyrian, or Macedonian) or in areas not far away from there (e.g. Messapic in southern Italy; Phrygian in Anatolia; Armenian in trans-Caucasian regions) in ancient times. From among the Palaeo-Balkanic languages only three a.re still used: Albanian, Armenian and Greek. The others have become irrevocably extinct, leaving, apart from glosses and onomastics, some inscriptional texts ( esp. Phrygian and Messa.pie, to a smaller extent Thracian and Mysian) or no written traces at all (e.g. 11acedonian , Illyrian , Epirotic, or Paeonian).
The central (or Palaeo-Balkanic)group was relatively closely connected wit h the eastern (or Inda-Iranian) one, but it has none of the specific innovations of the eastern group such as MONOVOCALISM (i.e. IE *e * a *o > Indo-Iranian *a ), PER MANE NT SATEM QUALITY, COMMON ETHNIC TERMINOLOGY (* A rya-), SPECIFIC RELIGIOUS VOCABULARY (e.g.
rta- ' cosmic order', the gods Asura-, Mitra-, Nasatiya-, Indra-) and
other features (Burrow 1976, 17-22). On the other hand, the Palaeo - Balkanic group is divided from the eastern one by phonetic archaisms (e.g. preserving the so-called 'prothetic vowels' that have completely disappeared in the eastern, northern and western groups, and have only partly survived in the archaic southern group, see vVitczak 1993) and by lexical correspondences such as3 :
(A) *andheyeti ' blossoms, blooms': Alb. endem 'I blossom, bloom'
II Gk. av{)iw 'id.' II Arm. andem 'I culti vat e';
see Burrow 1976, 7- 36) on the one hand, and on distinct similariti es within the Palaeo-Balkanic group suggesting the existence of a. primeval language league (resembling the present Balkanic league) on the other, cf. Bednarczuk 1986, 510- 511.
2 For the terms "Palaeo-Balkanic languages" and "Pa.la e o -Balkanic philology", see Neroznak 1978, 7- 20.
3 Below we list a number of isoglosses that appear exclusively in the Paleo
-Balkanic area and that are attested at least in three Paleo languages.
Some remra ks on the Albanian vocabulary 125
(B) *argel ya (f.) 'hut or underground dwelling': Alb. ragcil (f .) ' hut' II Rum. argea (f.) 'underground dwelling' < Dacian II Maced. ap,i>.>.a ' vapour -bath' I Kimmer . a p, D,) ,a i (f. pl.) 'underground
dwellings ' (cf. Neroznak 1978, 190);
(C) * autos 'the same:' Alb. vete 'id.' II NP hr . AYTOE, OP hr .
autaj (<lat . sg . fem.) I Gk. av r 6<;, Cret. a f r o<; II Mess. atavetes 'in the same year', cf. Gk. avr o£r i <; ad v. ' id.';
(D) *d2 gherdos/*d2 ghfd-s (f.) '(wild) pear-tree': Alb. dardhe (f.) ' pear-t ree' II Gk. axcpOo<; (f.) and axpa<; ( - a oo<; ) 'wild pear-tree, Pyrus amygdaliformis' II Maced. a1 i p5a (f.) 'common pear-tree,
Pyrus comn1unis' (cf. Huld 1984, 48);
(E) *;J2i fl-s / ;J2i §i- (f.) 'goat': Alb. edh 'kid, young goat' (alternatively from Latin haedus 'id .', cf. Rum. ied) II Gk. att, al,6<; (f.) 'goat' II Arm. aye (i-stem) 'she-goat ' (Huld 1984, 48);
(F) * biik]>o s 'vulva': Alb. bythe 'bottom, arse' II Gk. (Hes.)
f]vrro<; 'vulva' II Arm. puc'. 'id .' (Mann 1984-87, 58);
(G) * clesm6 s (m.) 'fat ': Alb. dhjame 'fat, lard' II Gk. Att. 817µ6<;
m. 'fat' II Arm . tam-uk' 'moisten' (Neroznak 1978, 182; Huld 1984,
60);
(H) *drepano- 'sickle': Alb. Tosk. draper, Gheg. drapen 'sickle'
II Gk . 5pbravov (n.) and 8p£1r6.vr; (f.) 'id.' II Arm. artewan1mk'
(Lambert erie 1983, 21- 22);
(I) *dhdl yeti 'blooms, sprouts' and *dhaler6s adj . 'springing forth, fresh': Alb. clal 'I bloom, go out' (Aor. clola) II Gk. 196.>..>..w 'I bloom,
sprout' (Aor. I k1J17>..a, perf. ri1J11>..a, Dor. Aeol. ri-i?a.>..a), ad j.
19a>..E.p6<; 'blooming, fresh, moist' II Arm. dalar adj . 'green, fresh'
(Hamp 1984; Muller 1984, 100);
(J) *dh rftbhos (m.) ' tomb': Gk. r6'cpo<; (m.) 'id.' II R oum. climb,
do.mb ' h Hlo ck. hill, rising round' < Da.cian II Arm . damban 'tomb' II
Gk. "Pelasgiau" TVf.l/30<; ( m . ) 'tomb, grave' (cf. Pokorny 1959, 248; Mann 1984, 87, 193);
(K) *d 1 dusna (f.) 'pain, ache, grief': Alb. dhune £. 'pah1, ache, grief, suffering, shame , disgrace, affront, violence' I Gk. Aeol. tovva (f.), At t . Ion. oo{;vr; (f .) 'gri ef' II Arm. erkan 'birth-pains; toil, ]abour
in childbirth' (Mann 1984-87, 170);
lGNACY RYSZARD DANKA, KRZYS Z'l'OF T OMAS Z WITCZAK
(L) *a1 rino- ( m . andn.) 'cloud, mist': Alb. Gheg. re (OGheg ren)
'cloud' II Gk. (Hes.) £plvov (n.) 'cloud' II illyr. pivoc; (m .) 'mist' II
1V[ess. p!.vo11 (n.) 'cloud' (Hamp 1957, 80; Neroznak 1978, 163);
(M) *kak6s adj. ' bad': Alb. keq adj. 'bad, evil' II Gk. 1,,0,1,,6c; adj. 'bad' II NPhr. 1,,0,K,ovv (n.) 'harm ' (see Huld 1984, 79- 80);
(N) *k6wilos adj. 'hollow, empty': Alb. thelle 'deep, dark' (as to colours) II Gk. 1,,oZ>.oc; a dj. 'hollow', Myc. ko-wi-ro II Arm. soyl ' cave'
(Huld 1978, 297-299; 1984, 118);
(0) *oler6- (adj.) 'impure, turbid ', (n.) 'dirt, mud, excrements': Alb. lere 'dirt (on the body)' II Gk. o>.£p6c; adj. 'impure, turbid' II
Maced. a>.ipov (n.) 'excrements' (Neroznak 1978, 180);
(P) *0(1.t )-k JJe: '(and) not': Alb. as ' not ' II Gk. ovr£ 'and not' II
NPhr. v - 1,,£? II Arm. oc' ' not';
(R) *a3k'1"-ya 1 (dual) 'two eyes': Alb. sy 'eye(s)' II Gk. ocHJ£ (du.) 'two eyes' II Arm. ac1c' 'e yes';
(S) *;)3ner yos (m.) 'dream': AJb. Task. ender, Gheg. ader 'id.'
II Gk. Att. Ion. OV£lpoc;, Aeol. ovolpoc;, Dor. avo.lpoc; II Arm. anur.f
(o-stem);
(T) *skorodom (n.) 'garlic, Allium sativum': Alb. hurdhe (f.) and
hudher, hudhre (f.) 'id.' II Gk. <71,,6p(o)6ov (n.) II Arm. xstor, NArm.
dial. sxtor 'garlic' (Djahukian 1967, 220).
2. NOTES ON THE SEMANTIC CHANGES IN Tiffi lNDO-EUROPEAN ARBOREAL TERMINOLOGY
This paper aims at discussing PaJaeo-Balkanic and Albanian arbo real terminology with special attention paid to semantic innovations in this :fiel d. Already at the very beginning of Indo-European studies as a scientific discipline, it was noticed that semantic innovations, usually motivated by observable extralinguistic factors, are of particular importance for comparative studies. It has been known for a long time that such words as Greek <.p171 6c; (f.) 'oak' and Latin fagus (f.)
'beech', although stemming from the uniform original Inda-European
etymon *bhag6s (feminine o-stem), differ considerably in their mean ing. Usually it is explained by the fact that beech is very rare in Greece and thus the original name has been transferred to another
Some remarks on the Albanian vocabulary 127
tree, narnely to oak, which 1s quite common there (Walde and Poko rny 1927, 128). An explanation for a similar correspondence between Latin fraxinus (f.) 'ash' and Sanskrit bhu1ja (m.) 'a kind of birch', Ossetic barz 'birch', Pol. brzoza 'id.', etc. (all from IE. *bhrH§- or bherH§ -, orig. 'birch') is that birch is hardly to be found on the Ital ian Peninsula and thus the Romans could have easily transferred its original name to ash, which also has light bark (see Friedrich 1970, 26- 31, esp. 29).
The triple semantic change ' a conifer' > ' a kind of willow' is well documented in the Slavic material, namely:
(1) Russ. dial. bred 'wicker', bred 'Salix caprea', bredina 'wil low, Salix', Ukr . bredina 'Salix caprea', Slovin. bfaza (f.) 'fruit-tree'
< Proto-Slavic *bred5, *bredin a, * bredja (f.) < IE. *bhroid( h)os,
*bhroid{h)ya (f.) 'pine', cf. Alb. bredh 'pine', also 'spruce' and 'larch', Roum. brad 'pine, spruce' (both from IE. *bhroid[h}os ); Lat vjan priede 'pine', dial. 'spruce' (from Baltic *braidja f. with initial p- by a cross with the synonymous Baltic word *peuse f. 'pine'), also pdeds f. 'jd.';
(2) Russ. iva 'willow', Pol. iwa, SC. 'iwa, et c. < PSI. *jiva f. 'a kind of willow' < IE . *twos (f.) 'yew', cf. Gaulish ivos, Fr. if, Welsh yw m. (sg. ywen), Bret. ivin m. (sg. ivinenn), Corn. ew, OCor n. (h)ivin 'yew'; OPruss. iuwis 'yew'; ONord . yr m. 'yew, yew-wood bow\ OE. iv, eov, E. yew, OHG. iwa, G. Eibe, MDu. iwe, Du. ijf 'yew';
(3) Cz. rokyta, dial. also rakyta, Pol. rokita, Russ., Blg., SC. rakita, etc.< PSI. *orkyta (f.) 'a kind of willow, brittle willow'< IE. *arkutos (f.) 'juniper', cf. Gk. o:.pK,cv'l3o<; f. 'juniper or cedar, Iuniperus'; Eteo Cretan &p,croc; 'id.'; Latvian ercis, also ecis ' juniper' .
Three observed semantic changes in Common Slavic are convincingly explained by the fact that the early Slavs, when moving from mountainous areas to plains rich in water, transferred the names of conifers (so numerous in the mountains) to the newly found kinds of willows.
In the Palaeo-Balkanic area, the arboreal terminology has also undergone characteristic semantic changes, in which Albanian participates to a smaller or a greater extent. Therefore, it is enough to examine, firstly, in which pan-Balkanic changes Albanian participates and in which it does not, secondly, in which cases it retains the original n1eaning and in which it does not, and thirdly, which tree terms it borrows and which it does not, to gather often invaluable information on the prehistory of Albanian in the Palaeo-Balkanic and ludo-European contexts.
3. 'BEECH' IN lNDO-EUROPEAN AND PALAEO-BALKANIC
Within the Palaeo-Balkanic languages, one can see two characteristic semantic changes concerning the arboreal terminology:
(a) The old appellative for 'beech' (IE. *bhag6 s f.) is used in the Palaeo-Balkanic area as a na111e for 'oak', cf. Greek <p'Tf,6c;, Dor. cpii 1 6c; (f.) 'Valonia oak, Quercus aegilops', Albanian bunge (f.) 'a kind of oak, Quercus esculentis' (from * bhiigna 4, see Huld 1981, 306; Adams 1985, 82), Phrygian Ba,a.Zoc; 'oak-god' (epitheton of the sky
-god, cf. Greek (Zcvc;) ?Pa,wva'ioc;, Umbrian (Iuppater) G,abovius'oak-god '5) .
The reason for this semantic change is not quite clear. Usu.ally its appearance is explained by scarcity or lack of beech. However, such an explanation is hard to accept as beech can be found almost all over the Balkans (except for the eastern and southernmost parts 6) in fair abundance, and the semantic change, as we can see, has a
4 It should be added that the Albanian word for ' Querc us esculentis', bunge, attests the presence of the velar stop *g in the Indo-European etymon * bhag6s (f.) 'beech'.
5The Indo-European sky-god *Dyiws is closely connected with the oak tree,
which used to be his sacred tree. Already in the Inda-European epoch, there appeared the epitheton *Perk'lun6s, literally 'oak-god>, which was derived from IE.
* perk'!Jv.s f. 'oak' (cf. Lat. quercus f. 'id .' ) and especially accepted in the northern
group. The divine epitheton * Perk'*iin6s (cf. ONord . Fj9rgynn, Lith. Perkilnas, OP ru s. Percunis and Slavic *Peruna/* Peryna with an irregular sound change under the influence of the verb * perq 'I strike') has supplanted the original name
of the sky-god (IE. *Dyiws or * Dyiws p.12 tir) in thenorth. There is no source
justification for the assertion that there were two different Inda-European gods having those names (i.e. * Dyiws and * Perk'lun6s) as the ranges of appearance of the two gods are disjointed and mutually complementary, which is a proof that
*Perk'Jl,un6s is only a local (i.e. northern) hypostasis of *Dyiws.
6See the map 16 in Friedrich (1970, 113).
Some remarks on the Albanian vocabulary 129
Pan Balkanic range. Probably we have to do with a non-conditioned semantic divergence that has occurred within the Palaeo-Balkanic group and has been generally accepted 7 .
(b) The semantic change 'beech, Fagus' > 'oak, Quercus' has caused an "empty place" (a free semanteme) in the microsemantic system of the particular Palaeo-Balkanjc languages. To fill it in, two independent semantic changes have taken place in the Palaeo Balkanic area:
(1) The meaning of 'beech' has been gjven to the old appellative form of 'ash, Fraxinus' (IE. *osk6s and *oskewa f.) . This process has appeared in the three basic Palaeo-Balkan languages:
Alb. ah (f.) 'beech' (< IE. *osk6s f. 'ash-tree', cf. Okel. askr m. 'ash, ashwood spear', OE. asc, etc. < Germanic *askaz m., Skt. ak$d
m. 'the tree Elaeocarpus ganitrus' (from *osk6s with the metathesis of -sk- > -ks- );
Gk. ofia and o(u17 (f.) 'beech' (from IE. *osJ..ewa f. 'asht
Arm. usi 'a deciduous tree', perhaps 'beech' (Djahukian 1967, 255), Mod. Arm. hosi 'a particula.r bush' (from IE. *oskew<i f. ' ash'). All other ludo-European stocks have retained the original mean ing, e.g. Lat. ornus 'ash', Slavic *( j)aseno 'ash' (both from PIE.
*H3osH1 enos ), see Pokorny 1959, 782.
(2) To denote 'beech' in other Palaeo-Balkanic languages, the old Indo-European etyrnon for 'elder', b(o)u§h6s (cf. Slavic *buzo beside
*bozo 'Sambucus nigra') has been used. This process can be proved for 1vi ysian (cf. µv(J6c; ' beech'8) and perhaps for some Armenian dialects (cf. Mod. Arm. piik(i in the Karabakh dialect).
7 Walde-Pokorny's explanation, cited in the proper text , was rightly abandoned by Rudnicki (1956, 128). A basis for I.he Palaeo-Balkanic shift from 'Fagus' to ' Quercus (esculent.is)' "".ould have been the edibility of both Aesculus acorns and beechnuts. They may be rendered into soap and, because they are sweet and tasty, into oils for cooking, as is in fact done in southern France. A similar explanation was suggested by Bender (1922, 30). According to him, the Greek shift took place in two stages, first from the beech to the <'native sweet chestnut that is still so characteristic of northwestern Greece", and then to the oak. "In any case, all three trees, beech, oak, and sweet chestnut , bear edible nuts, and belongs to the same family, Fagacaeae" (Bender 1922, 30). Compare also Friedrich (1970, 109).
8 See Georgiev (1981, 125). For the Mysian (or Lydian) provenance of µva6<:;
'beech', see Gusmani (1964, 276).
l GN AC Y R YS ZARD D ANK A, KRZYSZTOF TOMASZ WITCZAK
Both semantic processes discussed in points 3b (1- 2) can be convincingly applied to the period of existence of the Palaeo -Balkanic linguistic league.
4 . ' A SH-TREE' lN THE BALKANIC LANGUAGES
Further semantic changes took place independent ly from one an other, and particular Palaeo-Balkanic peoples introduced specific and divergent innovations to complement the semanteme ' ash':
(a) Albanian borrowed the termfrasher (Tosk.) / frashen (Gheg.) ' ash' from Latin fraxinus (f.) 'id.'. It is impossible to find out what term was used to describe a given tree jn the pre-Roman period;
(b) Greek adapted an old Indo-European word for ' Ostrya carpinofolia', thus Gk. µc:,>..i a (f . ) ' a sh ' continues IE. *melisa or
*melisya (f.) ' Ostrya carpinofolia' , cf. Alb. rnelleze, mellinje (f.) 'id.', Inda-Iranian *rnarisa - (the word attested in Fenno-Ugric *maris 'lime-tree': Mord. M. mar s, maras 'elm, ash'; Vot. dial. b§r jz, berji,
ber ji,- pu [pu ' tree , wood'L ber ji , berii , berzi 'lime-t ree' ; Zir. beris
'id.')9•
(c) To describe ' a.sh', Armenian used the original term for 'elder tree, Sambucus nigra L.': Arm. hac'i (gen. pl. hac'eac') 'ash' < IE.
*a2 kpewa or * d 2 k]>ewos (f . and m.) 'elder ', cf. Gk. 0:11,ria. (f.) and
cr.11,rio<; (m.) 'Sambucus nigra', Dacian <7i /3a., seva 'id .'. For further cognates, see Witczak (1992).
5 . ' E LDER-TREE' IN THE BALKANJC LANGU AGES
In the Palaeo-Balkan area, both primaeval Inda-European appellatives for 'elder-tree, Sambucus nigra', IE. * d 2kpewa and *bough6 s
(f.), have survived, but in most cases they have undergone a semantic change.
(a) The original meaning of the firnt appellative is preserved in Greek and Dacian (see above, Sect. 4c). In Armenian this appellative
9 An alternative etymology of the Fenno-Ugric words is suggested by Redei (1986, 55), who assumes a borrowing from Middle Iranian * barza- ' birch'.
Some remarks on the Albanian vocabulary 131
means 'ash-tree', whereas in Albanian and in Bessan it has undergone a deeper semantic divergence: 'Sambucus nigra' > 'Ilex aquifolium'
> 'Tussilago farfara', cf. Bessan ao-a 'colt's foot, Tussilago farfara ' (from *ao-ia and IE. * 2kpewii ), Alb. ashe, ashja ' Ilex aquifolium' beside 'Tussilago farfara.'. The semantic change is fully motivated by the similarity of sharp-pointed leaves of all these plants.
(b) In the Palaeo-Balkanic area, the alternative word for 'elder
-tree', IE. *b( o)ugh6s (f.), has been transferred to 'beech', cf. Mysian
µvo-6<; 'beech' (see Sect. 3b).
(c) Dacian has introduced an alternative word for ' Sambucus ni gra', oAµa (f.)10 , which, however, still has the original Indo-European etymon of 'elm-tree' (IE. */m os (f.), cf. Lat . ulm1.ts f. 'id.', 0Ir. lem
< Celtic *limos, Slavic * jblmo) .
(d) The native form of the Albanian appelJat ive for 'elder-tree' is not known. In the contemporary literary language, similarly as in dialects, the only term to appear is shtog 'Sambucus nigra', borrowed from Vulgar Latin *sab'ltcus < Lat. sambiicus, cf. Roum. soc ' elder
-tree'.
,6 THE ALBANIAN ARBOREAL TERMINOLOGY AS AN EVIDENCE FOR THE ORIGIN
OF THE ALBANIANS
The observations described in points 3- 5 are fairly important for determining the connexons between Albanian and ancient languages of the Balkans. It is worth noticing that Albanian and Mysian have dHferent (although native and primaeval) terms for 'beech' (cf. Alb. ah 'beech', orig. 'ash-tree' vs. Mys. µvo-6<; 'beech', orig. 'elder-tree'),
which should exclude the possibility of the Mysian origin of Albanian . Moreover, there is a great semantic and phonetic divergence bet,.,ween Dacian and Albanian as far as the ludo-European term *d2 kj:>ewii
(f.) 'elder' is concerned, which speaks against the Dacian hypothesis (cf. Dae. o-i/3a, seva 'Sambucus nigra' vs. Alb. ashe 'Ilex aquifolium; Tussilago farfara'). On the other hand, one should note the similarity
10 Neroznak (1978, 61) supposes that the Dacian word 5>..µcr is a borrowing from Latin ulmus 'elm'. Such an interpretation, however, has no semantic, phonological, or morphological justification.
IGNACY RYSZARD DANKA, KRZYSZTOF TOMASZ WITCZAK
between Alb. ashe and Bessan &a& covering both semantics (thesame im1ovation) and phonetics11 .
Albanian has kept Indo-European terms applying to particular conifers such as:
(a.) ' yew': Alb. Gheg. bershe 'stinking juniper ' (< IE.
*ebhur-isyo-, cf. Ofr. iber m. 'yew', Gaulo-Romance *eburos 'id.', MHG. eber-boum and German Eber-esche ' mountain-ash', all from IE. *ebhuros 'yew', see Pokorny 1959, 334);
(b) ' pine': Alb. bredh (<* bhroid{h}o-, cf. Roum. brad 'pine, spruce', Lat v. prieds f. and priede f. 'pine\ Russ . bred, dial. bred, bredina ' willow, Salix', Slovinian braza 'fruit-tree' (see Sect. 2);
(c) 'ftr ': Alb. hale f. 'black pine, Pinus nigra' (from IE. *adhla or
*adhlis f. 'fir, Abies', cf. OPr us. addle 'fir, spruce', .Jatvingian egle 'fir-tree' , Ljth. egle, dial. agle, Lat v. egle; OPol. jedl, jedla, Pol. .iodla 'fir', Cz. jedle, Polabian jadla, Russ. elb, dial. elina; 0 Ir . aidlen, MCorn. adlen 'Abies', Bret . aedlen, edlen 'fir';
(d) 'spruce' : Alb. pishe (< *pit-so-, cf. Gk. 'lfiTvc; f. 'pine, spruce\ Skt . pztu-dciru ( n.) 'a kind of pine, or resinous tree', see Huld 1983, 303). An alternative etymology connects the Albanian word with Latin picea f. 'spruce';
(e) 'juniper': Alb. dellinje, dellenje, delli (< IE . *(s)ghelin yo- 'ju niper', cf. Gk. ( Hes.) <JXi>..ivo<; 'wild cypress or juniper';
(f) 'larch': Alb. bre, breni 'spruce, fir' (< IE . * brin Hu- 'larch' , cf. Romance *brinu- 'larch', attested in Italian dialects: brenge, brendze (Valle dell' Orca) , brenva (Val'soana), brengola. (Canavese); Pol. dial. brzyn, brzym (Silesia), Cz. dial. brim 'larch', Sln. brin 'juniper' or 'larch, Larix meleze', SC. dial. brina (f.) ' juniper' or 'spruce, Picea ' , Sln. brina : brinja (f.) 'a conifer', usually 'juniper', dial. 'spruce' and also ' pine' .
(g) Only Alb. rrobu.ll ' Pinus leucodermus' is probably of foreign
origin, although it is not certain whether this word was borrowed from Latin ravicella (f.) 'a conifer' or not.
11 This similarity may be a linguistic confirmation of Witczak's hypothesis of the Bessan origin of the Albanian population .
Some remarks on the Albanian vocabulary .133
011 the other hand, Albanian has not kept a large number of names vJ. the deciduous trees, borrowing them from Ancient Greek in the Archaic period, from Latin at the time of Roman domination in the Balkans or from Slavs after their great migration. The following arboreal terms can be quoted here:
(h) ' ash': A]b. Tosk. frasher, Gheg. frashen < Lat. fraxinus f.
'ash', cf. Rourn. frdsin (see Sect. 4a);
(i) ' willow': Alb. shelq, -u (m.) < Lat. salix, -icis f. 'willow', cf.
Roum. scilcie f. 'willow', also 'a willow t wig ';
(j) 'poplar': Alb. verre (f.) 'silver poplar, Populus alba' < PGk.
*awiger( y)o s, cf. Greek at,E.poc; or atyElpoc; (f.) 'black poplar, Po pulus nigra';
(k) 'poplar': Alb. plep, -i (m.) < Vulgar Lat. *plopus < Lat.
popultts 'poplar', cf. RoU1Il. plop;
(1) 'poplar': Alb. topole, tupli < Slavic *topolb 'poplar' (Svane 1992, 128);
(l) ' willow': Alb. Gheg. mqite 'Salix alba' < South Slavic *rak yta
(Sva.ne 1992, 127- 128);
(m) 'fig-t ree': Alb. fik, -tt (m.) < Lat. ficus, -us or -f (f.) 'fig-t ree', cf. Aroum . hie (m.) 'id .' , hica (f.) 'fig, grape';
(n) 'bay-tree': Alb. lar, -i (m.) < Lat . laurus, -zor -us (f.) ' bay
tree\ cf. Roum. ldur (m.);
(o) 'chestnut ': Alb. geshten.ie , -a (f.) < Lat. castanea 'chestnut (tree and nut)', cf. Roum. castan (m .) ' chestnut -tree' , castana (f .) 'chestnut';
(p) 'oak': Alb. rre 'English oak, Quercus robur' < Lat. ri5bur
'id .';
(r) 'elder': Alb. shtog < Vulgar Latin sabucus < Lat. sambucus
'id.', cf. Roum. soc (see Sect. 5d).
This list shows tha.t Albanian has preserved native tenns for conifers typical of a mountain landscape and lost the terms for hydrophilous trees that grow on lowland plains along rivers, marshes and lakes such as ' willow' or ' poplar'. In this situation it is to be supposed that the forefathers of the Albanians lived in areas that were comparatively badly watered and where conifers were dominant, i.e. in typically mountainous areas.
REFERENCES
ADAMS, D. Q. 1985. "The Inda-European word for ' apple' again,,, Indogerm a nische Forschungen 90, 79- 82.
BEDNARCZUK, L. 1976. ''Indoeur opejskie j zyki Balkan6w", in L. Bednarczuk,
ed., J zyki indoeuropejskie, t. 1, Warszawa: PWN, 471- 513.
BENDER, H. H. 1922. The home of the Inclo-Europeans, Princeton.
BURROW, T H. 1976 . Sanskrit, Moskva. (In Russian; primarily published in English as The Sanskrit language, London 1959).
DANKA, I. R. 1983. Stanowisko j<;.zyk6w anatolijskich w rodzinie indoeuropejskiej
i ich wzajemne zwiqzki, Lodi: Wyd. Uniwersytetu Lodzkiego.
DJAHUKIAN, G. V. 1967. Ocerki po istorii dopis'mennogo perioda armjanskogo jazyka, Erevan.
FRIEDRICH, P. 1970. Proto-In da- European trees. The arboreal system of a p1·e
historic people, Chicago.
GEORGIEV, V. I. 1981. Introduction to the history of the lndo-European lan guages, Sofia.
GUSMAN!, R. 1964. Lydisches Worterbttch, Heidelberg: Winter .
HAMP, E. P. 1957. "Albanian and Messapic", in Studies presented to Joshua Whatmough on his sixtieth birthday, The Hague: Mouton, 73-89.
- · 1984. "Armenian dalar 'green', fJa>.t::p6<; 'moist',,, Die Sprache 30, 156-159.
HULD, M. E. 1978. "Albanian i thelle and its ea.stern relations,,, Zeitschrift Ji.fr
vergleichende S'prachforschung 92, 297-299.
- · 1981. « Alban ian corrigenda to Friedrich's Proto-Indo-European trees,,,
Zeitschrift fur vergleichende Sprachforschung 95, 302- 308.
-· 1984. Basic Albanian etymologies, Los Angeles: Slavica Publishers. LAMBERTERIE, CH. DE. 1983. "Une bagatelle etymologique ( A rtew a-
nunk' / opt::1ietv·r1),,, Revue des etudes armeniennes 17, 21-22.
MANN, S. E . 1984-1987. An Inda-European comparative dictionary, Hamburg : Buske.
MULLER, G. 1984. "Notes on four Old Armenian etymologies,,, Arbejds Papirer
(4) udsendt af Institut for Lingvistik K¢benhavns Universitet, 99-101. NEROZNAK, V. P. 1978. Paleobalkanskije jazyki, Moskva.
POKORNY, J. 1959. Indogermanisches etymologisches Worterbuch, Bern- Miinchen: Francke .
REDEI, K. 1986. Zu den indogermanisch-1.walischen Sprachkontakten, Wien.
RUDNICKI, M. 1956. "Wartosc nazw drzewa bukowego, lososia i rdzenia lendh dla wyznaczenia prakolebki (praojczyzny) indoeuropejskiej i slowianskiej", Biuletyn Polskiego Towarzystwa J zykoznawczego ]5, 127- 137.
SVANE, G. 1992. Slavische Lehnworter im Albanischen, Aarhus: University Press. WALDE, A., POKORNY, J. 1927-1932. Vergleichendes Worterbuch der indoger
manischen Sprachen, Bde. 1- 3 , Berlin- Leipzig.
WITCZAK, K. T. 1992. "Inda-European *:J2kpewa 'elder, Sambucus nigra L.' and its Baltic descendants", Linguistica Baltica l, 201- 211.
Some remarks on the Albanian vocabulary 135
WITCZAK, K. T. 1993. "T he triple representation of prothetic vowels in an cient ( u:eek" , in Studies in Greek Linguistics . Proceedings of th e 14th Annual Meeting of the Department of Linguistics, Faculty of Philosoph y, Aristotle Uni versit y of T hessaloniki , 27-29 April 1993, Thessaloniki 1993, 487- 497.
University of Lodz
Chair of Classical Philology ul. Wolczanska 90
P L- 90- 522 Lodz
Igna cy Ryszard Danka and
Krzysztof 'l'omasz Witczak