Kalbotyros pastabos

ACTA LINGUISTICA LITHUANICA XLIV (2001), 89–103 Notes on Linguistics* SIMAS KARALIŪNAS Institute of the Lithuanian Language, Vilnius The article discusses questions of etymology in connection with (1) Lith. dial. berklai ‘foot-bridge’, (2) Lith. gintaras, Latv. dzītars, (3) Lith. dial. krontas and krōnas ‘snow-drift’, (4) Latv. viņš, fem. viņa ‘he, she’, (5) Lith. reflexes of Baltic *kikir- ‘pea’ in hydronyms, (6) Old Prussian rapa ‘angel’. 16. LITH. DIAL. berklai ‘FOOTBRIDGE’ In the northeastern corner of Lithuania (Salakas), the word berklai (acc. plur. bérklus) is used, whose meaning is given as ‘footbridge (usually made of poles)’ (LKŽ₁ 626; ²LKŽ₁ 766). Considering the historical sound correspondences, the original form can be reconstructed as *bēr-tlo-. Structurally and functionally—as a kopnomen instrumenti with the functional meaning ‘that by which people are carried (transported), an instrument for transporting people’—the word is immediately connected with ārklas (plur. arklai) < *ārH-tlo- (with the suffix -tl- formed by analogy), whose etymological equivalent is Gk. ἄροτρον, Cretan ἄρατρον ‘plough’ (< *arH-tro-). By this analogy, the etymological equivalent of berklai would be Gk. φέρετρον (and φέρτρον) ‘bier’, < *bher(H)-tro-), a derivative of φέρω ‘to carry’ (cf. Frisk 1970: 1003, 1005; Chantraine 1968: 1191). After all, biers are also made of poles. An Old Indic bharītra-, a hapax legomenon in the Rigveda, probably meaning ‘hand’ or ‘finger’ (that is, *‘that with which one carries, a carrying implement, tool’), is also considered an old derivative with the suffix -tro- (not -tlo-, as is sometimes conjectured) (cf. Mayrhofer 1963: 478; 1996: 251). But a more precise equivalent can also be found for berklai—namely, Lat. fērculum (and ferculum) ‘bier; tray’ < *bher(H)-tlo-m (Ernout, Meillet 1959: 226; somewhat differently Walde, Hofmann 1938: 484), a derivative of ferō ‘to carry’. Consequently, berklai must also be a derivative of the root *bher- ‘to carry’, which means that the Baltic verb bērti bēria bėrė (and dial. bérti bēria bėrė, long associated with the Greek and Latin verbs mentioned above (cf. Fraenkel 1962: 40), also had this kind of meaning. The structure of the root and the acute accent in berklai show that its original form was *bēr-tlo-, but this form raises considerable questions: did its long root vowel arise through qualitative ablaut by lengthening the root vowel e, or was the lengthening caused by the loss of a laryngeal? In the latest “Indo-European Verb Dictionary,” the root under discussion is reconstructed as an aniṭ root, that is, without a laryngeal; Lithuanian bérti is also tentatively assigned to it (cf. Rix, ed. 1998: 61–62). M. Mayrhofer, too, is inclined to accept such a root, but he also notes that the existence of a set variant of this root cannot be entirely ruled out of consideration (Mayrho- For the beginning, see Questions of Lithuanian Linguistics 17, 207–209; 29, 65–78; 31, 125–129. 90 | SIMAS KARALIŪNAS fer 1996: 249). Kad s. ind. bharitra- rodytų buvus *bherH-, jo nuomone, nėra tikra, nes bharitra- gali būti ad hoc darinys pagal caritra- koją’ (Mayrhofer 1996: 251). Mūsų atveju svarbu tai, kad šios šaknies vediniai slavų kalbose taip pat turi akutą: serbų-kroatų brème, slovėnų bréme, čekų břímě, rusų бе́ремя ‘našta’ < sl. *brémen- veikiausiai rodo buvus pailgintojo laipsnio šaknies variantą *bher- (Mayrhofer 1996: 249). Tad ir berklai pradinės formos *bēr-tlo- ilgasis šaknies balsis taip pat veikiausiai yra kiekybinės balsių kaitos dalykas. Todėl jo darybos pagrindu laikytina būtojo laiko forma bėrė (bėrė cirkumfleksinė priegaidė, matyt, yra antrinė kaip metatonijos rezultatas). The suffix -tlo-, which is now tending to be reconstructed for Proto-Slavic as well, is a means of forming names of instruments characteristic of Baltic and Latin (Seldeslachts-Swiggers 1995: 23–37). Conclusion: berklai ‘a footbridge (usually made of poles)’ will have preserved the original meaning ‘to carry’ of the verb bérti (bėra, bėrė). Although Lithuanian berklai (< *bēr-tlo-) and Latin ferculum (and fericulum) ‘a carrying frame; tray’ < *bher(H)-tlo-m are most likely products of independent development, their exact correspondence being noteworthy in terms of Baltic and Latin isoglosses. 17. LITH. gintaras In Prussia, especially in Samland, amber is called by the purely German word Bernstein. The rare German word Kentner ‘amber’ is derived from Medieval Latin gentārum; the latter points to Lithuanian gintaras, Latvian dzītars and zītars. Which of these words, however, do not give the impression of being genuinely Baltic words; compare Hungarian gyanta ‘resin’, dialectal gyantar ‘amber’. Perhaps the Lirian language should be considered their source? If the Aestii—that is, undoubtedly, ‘das Bernsteinvolk’—really were Balts, why did they not have a Baltic word for their most important trade commodity? E. Sittig asks (Sittig 1934–1935: 20). J. Endzelīns answered E. Sittig’s question with another question: How, then, do we know that the Aestii did not have a Baltic name for amber? If glesum really was a Germanic word, initially foreign to the Balts, the Aestii—if they were Balts—could have become acquainted with it through contact with their Germanic neighbours or heard it from southern traders, who, on their way to the land of the Aestii, might have picked up this word from the Germans and then asked the Aestii for ‘glēsum’. As a consequence, in communicating with foreigners the Aestii might themselves have begun using the word glėsum, despite possibly having their own name for amber. Sixteenth-century literature mentions the Prussian name for amber gentars, J. Endzelīns adds (Endzelīns 1943: 5–6). As we see, the discussion between E. Sittig and J. Endzelīns about the ethnic affiliation of the Aestii described by Tacitus—whether they were Germans (E. Sittig) or Balts (J. Endzelīns)—raised a very important question: Did the ancestors of the Prussians, Lithuanians, and Latvians have their own Baltic name for amber? Here it is also necessary to recall the fact that J. J. Mikkola, writing about the origin of gintāras, expressed agreement with the opinion of J. Gerulis, a well-known Balticist, regarding the origin of this word. J. Gerulis’s answer was as follows: in its formation, gintaras is said to be isolated; there seems to be no way to connect it etymologically with the Baltic languages; in short, gintaras, dzintars, dzītars cannot, it is said, be considered Baltic words (Mikkola 1938: 35). Nothing is said about its origin in E. Fraenkel’s “Lithuanian Etymological Dictionary” either (Fraenkel 1962: 152). As can be seen, the question of whether the ancestors of the Prussians, Lithuanians, and Latvians had their own Baltic name for amber is raised with good reason. KALBOTYROS PASTABOS | 91 J. J. Mikkola argued that Lithuanian gintaras came from an unspecified eastern country (compare Chuvash janDar ‘glass vessel; glass’, Cheremis jumdar ‘bottle’); compare Mikkola (1938: 33–37). B. Larin argued that the Baltic name for amber, together with Hungarian gyanta, originated among the indigenous inhabitants of the eastern Baltic region in prehistoric times, the amber gatherers, the ‘Pomeranians’. Lithuanian giñtaras, gintãras, gentãras, it is said, cannot be explained either morphologically or semantically within the Lithuanian language. Hungarian gyanta could not have arisen from gyantár, since final r is not lost in Hungarian. Both the morphological alternation -a: -ár and the characteristic meaning ‘amber; pine resin’ are said to have come from the language from which the Hungarians borrowed this name. The Hungarians served as intermediaries in the word’s migration into the Baltic languages (Larin 1956: 149–162). O. Trubačev’s explanation of the origin of giñtaras is fantastic and therefore unacceptable: the word is supposed to have been borrowed from Russian янтарь, ентарь, which, via Chuvash janDar ‘glass vessel; glass’, janDal ‘precious beads; amber’, derives from Old Indian yántar- ‘restraining’ (which supposedly reflects the magnetism of amber) (Trubačev 1980: 7–18). Attempts have also been made to explain the word on the basis of Baltic-language material. E. Blesse morphologically segmented the name for amber as *gint-aras, connected it with Latvian guntina ‘little fire’, assigned it the original meaning ‘fire’, sought to substantiate the connection between the meanings ‘fire’ and ‘amber’, and traced both words, as well as Latvian dialectal guns ‘fire’, to the Indo-European root *gʰen- ‘to cut, strike; drive’, which is represented by Latvian dzīt (dzenu, dzinu) ‘to chase, drive’, Lithuanian ginti (gēna, ginė); compare Blesse (1958: 200, note 1). This scholar’s efforts to substantiate the connection between the meanings ‘fire’ and ‘amber’ are noteworthy, since they accord with typological data. Latvian guntina is a diminutive of the dialectal (chiefly Upper Latvian dialects’) guns ‘fire’, and diminutives tend to add -t-, e.g., uguntina, uguntinš alongside uguns ‘fire’ ME IV 294; therefore, a root *gunt- need not be present here. Historically, this dialectal word cannot be separated from Latvian uguns and Lithuanian ugnis. Together with their etymological cognates, Old Church Slavonic огнь, Latin ignis (< *egnis) and Old Indian agniḥ, they developed from the Proto-Indo-European paradigm *Hgnei-/*Hᵒgnei-: *Hogni-: Hegni- (Karaliūnas 1990b: 177–178). It is interesting that P. Anttila also connects ginti (gēna, ginė) with giñtaras, but regards its primary meaning as ‘resinous icicle, drip’ (Anttila 1997: 112, 114). Priesagos -tar- buvimas baltų kalbose yra toks neaiškus (Skardžius 1943: 303–304), kad V. Mažiulis buvo priverstas griebtis priesagos -tr- (girdi, vėliau paveiktos priesagos -ter-), kad paaiškintų gintaro pavadinimo baltų kalbose susidarymą – iš veiksmažodžio ginti (gīna, gynė) šaknies: girdi, senų senovėje pagrindinė gintaro funkcija buvusi magiškoji – ap(si)gynimo nuo tam tikrų blogybių (Mažiulis 1988: 352). Nesunku čia matyti įtaigą hipotezių, graikų kalbos gintaro pavadinimą ἤλεκτρον (šaknis *eh₁lek-) aiškinančių veiksmažodžiu ἀλέγω ‘ginti, atremti, atmušti’ (ide. *h₂lek-s-) (Toporov 1979: 215), tačiau neturinčių tvirtesnio pamato. Antai L. Deroy ir R. Halleux, turėdami galvoje, kad gintaro gabaliukai ligi atsirandant monetoms buvo vartojami skaičiavimui mainuose, jo pavadinimą susiejo su gr. ἀλέγω ‘rūpintis kuo, galvoti apie ką, kreipti dėmesį’ (Deroy, Halleux 1974: 37–52). First, it is worth looking at how Hungarian linguists themselves explain their language’s name for amber. In the most recent Hungarian etymological dictionary, gyanta ‘amber’ (with the variants gyenta and gyonta) (for more, see Voigt 1970: 575–578), attested in historical sources as early as the beginning of the sixteenth century (gyentha 1500, güwnthar 1525, giantar 1539, gentharos 1542), is described as a migrating word of Baltic origin (Wanderwort), like Medieval Latin gentarium and German Kentner (now obsolete). In Hungarian it also means ‘resin’ and ‘glaze’, but these meanings are considered to have developed 92 | SIMAS KARALIŪNAS … from the primary meaning ‘amber’. The final -r of the word was lost in a closed syllable (Benkō, ed., 1993: 492). We have no reason to doubt these claims by Hungarian linguists, especially since the word under discussion is attested in their language relatively late. On the other hand, it is possible that, as V. Voigt conjectures, this migratory word was borrowed into Hungarian in the first millennium A.D. and adapted accordingly (Voigt 1970: 574–580). The name for amber is also attested late in Russian. It is first encountered in 1551, енping?, appears most often in seventeenth-century writings, and from then on its form is “hypernormalized” by scribes and becomes янтарь, just as, for example, ендова (a borrowing from Lithuanian indauja) becomes яндова́. The presence of consonant n shows that енping? entered East Slavic dialects no earlier than the tenth century; otherwise it would have been *ямарь (Larin 1956: 149–151). This is an indisputable borrowing from Lithuanian, and the form used by the Prussian Lithuanians, jentāras, gentāras, is generally considered its original (Vasmer 1978: 558; Sławski 1952: 501; Kiparsky 1973: 69; Laučiūtė 1982: 26). From Russian, this name for amber also spread to other Slavic languages (Vasmer 1978: 558). According to the data in K. Būga’s Aistiškų Studijų (34) and the academic Lietuvių kalbos žodyno (LKŽIII 322), gintaras ‘fossilized resin, washed ashore by the sea or dug from the ground (glaesum)’ is most widely used in Samogitian dialects, in their western margins (Kretinga, Skuodas, Plungė, and Šilalė districts). It is also familiar in the northeastern dünininkai dialects (DūnŽ 486). The plural form and the meaning ‘(amber) beads’ are also known in the eastern dialects (Salamiestis, Kupiškis district; Troškūnai, Anykščiai district). That it was once more widely used in the eastern dialects is evident from its record in the third edition of K. Sirvydas’s dictionary (gintaras, gen. sing. gintaro, adjective gintarinis) (Pakalka, et al., 1979: 123), as well as in his Punktuose sakymų (I, 183). The gintāris of the dialects around Druskininkai (DrskŽ 101) is thought to have come from the standard language. The accentual variant gintāras is found in F. Kurschat’s dictionary (K2 211). K. Būga testifies that gintāras was also used in everyday speech, for example, in the environs of Tilsit (Aistiški Studijai, 34). Among Prussian Lithuanians, it seems to be attested earliest (early eighteenth century) in J. Brodowski’s dictionary. The rare form gentāras occurs in the dictionaries of P. Ruhig and K. Milk (LKŽIII 237), but F. Kurschat says he had never heard it. The mid-seventeenth-century manuscript dictionary Lexicon Lithuanicum gives the forms gentāras and gintorius (Drotvinas, ed., 1987: 100, 180). Moreover, the form jentāras, also attested in the dictionaries of Prussian Lithuanians (Ruhig, Milk, Krause, Nesselmann) (LKŽIII 340), likewise has e in its root. Thus, the accentuation of the suffix -ar- is well attested and beyond doubt. The fact that the sound j- occurs here instead of g- should apparently be explained by language contact. In Middle Low German, spoken by the German inhabitants of Prussia, the consonant g before front vowels was pronounced like j (cf. jēn = gehen ‘to go’, jīcht = Gicht ‘gout’). Iš įvairių vokiškų dokumentų sus ekama prūsus vartojus gentars (Nesselman 1873: 46; Toporov 1979: 211–212). Iš Karaliaučiaus senamiesčio parapijos klebono Pollianderio laiško, lotyniškai rašyto į Leipcigą 1535 m. draugui mokytojui Kasparui Börneriui, sužinome, kad Semboje, kur gyvena sūduviai, „yra daugiau kaip 32 kaimai, kurių gyventojai jau nuo senų laikų visu pajūriu gaudo gintarą, kurį jie savo protėvių kalba vadina gentārum“ (Būga 1961: 151). Forma gentārum, suprantama, yra lotynizuota. Kiek vėliau, 1551 m., tai patvirtina ir A. Aurifaberis savo veikale Succini historia pasakydamas: „Sūduviai gintarą savo kalba vadina Gentarn“ („Die Sudawen heissen den Börnstein in jrer sprach Gentarn“) (Būga 1961: 150–151; Toporov 1979: 211–212). Šitokią gintaro pavadinimo formą teisingai yra paaiškinęs K. Būga: „Sūdai gintarą bus tarę gentaran (nom. KALBOTYROS PASTABOS 93 sing. n.). The word’s vocalism has e from i. Thus, gentārum should be read in Prussian as gintaran” (Būga 1961: 150). V. Toporov, agreeing with K. Būga, considers Prussian gentars to be a later formation, or even an incorrect reconstruction based on East Prussian German Gentar; he also does not rule out the possibility that gentārum in Latin sources is the result of folk etymology, whereby amber was understood as a producer of fire and its name was adapted to Latin genitor ‘begetter’ (Toporov 1979: 212). V. Mažiulis also conjectures (“probably”) that Prussian had gintars (Mažiulis 1988: 352). If we assume that Prussian and Sudovian (Yotvingian) had gintar(a)s (with i in the root), it is difficult to explain the Lithuanian dialect forms gentāras ir jentāras. Since these forms are attested in Prussian Lithuanian dictionaries, one may suppose that they entered the dialects of the region’s Lithuanians from Prussian dialects. In Prussian place names in the Order’s sources, as well as in Prussian-language monuments—the Elbing Vocabulary and the catechisms—the letter e is often written in place of etymological i. From this, the reasonable conclusion has been drawn (Gerullis 1922: 215; Trautmann 1925: 124; Endzelīns 1943: 25) that in Prussian dialects, at least in some of them, i may have been pronounced widely and openly, that is, as ę, as K. Būga put it (for further details, s.v. Prussian keckirs ‘pea’, see below). The original form *gintara- (with root vowel i) is supported by the Latvian equivalent dzītars, known from folksongs (ME₁ 559, EH₁ 362; Gāters 1979: 283–294), as well as dzītitars in folksongs and colloquial speech (e.g., Pūre, Tukums District), with variants dzintārs (unstressed) in the Curonian Spit (Schmid 1983: 261–262; 1989: 34, 36), dzinteris (unstressed) in folksongs, and dzintārs in Ungurmuiža (ME₁ 552, EH₁ 359). The word dzintārs and its variants are considered to have come from Curonian and to have been Latvianized because the diphthong in was preserved (Endzelīns 1913–1914: 98 [=Endzelīns 1974: 506]; 1940: 254 [=Endzelīns 1980: 561]; ME₁ 552; Būga 1961: 179–180); it is well attested in the place names Sintere 1253, Sinter 1291, Zintere 1305, Zynthere 1338 in Curonian territory, and in the river names Gintarà, a tributary of the Šaltuona, Raseiniai; Гинтора, 16th century (Būga 1961: 168, 175; Kiparsky 1939: 199), Gintaras, a river in Raseiniai; Giñtaris, a river in Šeduva, in Lithuanian territory. The form zītars is widely used in contemporary Latvian, with the variants zīters, zīteris², ziters ME₁ 737, EH₁ 810; it has word-initial z-, sporadically simplified from dz-, cf. zęguze ‘cuckoo’: dzęguze ‘same’ (cf. Lithuanian gegužė); zeldināt ‘to sting’: dzelidnāt ‘same’; zelt ‘to sting’: dzelt (cf. Lithuanian gėlti). A review of the evidence shows that the name for amber *gintara- is an old Baltic word, dating back to the thirteenth century. K. Būga rightly isolated the suffix -aras in the word giñtaras and explained the difference in accent between Lithuanian giñtaras, Latvian dzītars (circumflex vs. acute) by metatony (Būga 1959: 390). The original form of the suffix seems to have been -ar- (< Indo-European *-or-), since, first, Baltic -er- later developed from it (apparently by analogy with the front vowel in the ending -is, later extended to the position before the ending -as), and, second, it was lengthened to -ār-, apparently on the model of gāugaras ‘mountain top, knoll’: gaugorà ‘a raised place in a field or meadow…’ LKŽ₃₃ 167; kaũkaras ‘hillock, hill, mound’: kaũkoras ‘same’, kaukoris ‘mound, pile’ LKŽ₇ 420, 423, 424. As W. Schmid has already observed (Schmid 1993: 426), the lengthening of the suffix vowel may also have been influenced (cf. gintorius) by semantically similar words with the Slavic suffix -orius, used in folksongs, such as: Does the lamp shine? Does the fire blaze? LB 102 (from a collection of Lithuanian folksongs published by A. Leskien and K. Brugmann in 1882). The name for amber in the Baltic languages has also been examined in a historical-comparative perspective, and its derivation and semantics persuasively explained, by the distinguished German Indo-Europeanist and Balticist Wolfgang Schmid (Schmid 1993: 425–430). He begins his study with Lithuanian giñtaras, 94 | SIMAS KARALIŪNAS and thus also of its cognates in other Baltic languages into the morphemes gint- and -ara-. That the root *gent- with the zero-grade ablaut variant *gint- (< *gṇt-) is possible in the Baltic languages, though unattested, and that a substrate hypothesis is therefore unnecessary to explain the origin of this word, is evident from the following examples (Schmid 1993: 428): *kent-/*kint-: Lithuanian Viškinas, Prussian Wissekint; the element kint- is related by root ablaut to the verb kęsti (kenčia) ‘to endure’; *ment-/*mint-: mįsti (minčia, mintė) ‘to set a riddle, to guess’, mintis ‘riddle’, mintė ‘memory, mind’ are cognate with pa-męsti (-menčia, -mentė) (unstressed) ‘to advise’ LKŽVIII 84, 232, 245, 284; *pent-/*pint-: iš-péntėti (-ėja) ‘to crumble away, to rot; to dry out, become parched, harden…’: pintis ‘a kind of tree growth, a fungus growing on decaying wood’ LKŽIX 798, 1061; *rent-/*rint-: rintis (also rintis, rintis) ‘notch, incision; a mark around a cow’s horn indicating how many calves it has had…’: ręsti (renčia, rentė) ‘to make notches, cuts…’ LKŽXI 477, 671. W. Schmid further draws on R. Much’s claim that gintãras may be cognate with Old Norse kynda, Middle High German künden, German dialect (Bavarian) kenten ‘to kindle’ (Germanic *kundjan, *kandjan) (Much 1937: 405; 1967: 512). W. Schmid considers these Germanic verbs and the Baltic word for amber to be related from ancient times (urverwandt), deriving them from the root *gni- ‘to burn’. The Finnish Germanic loanword kynttilä ‘candle’ also belongs to this root. Thus, the motivation behind the name for amber was not its protective function (against illness), but rather its property of catching fire quickly. The formation of the name for amber in German is likewise based on this motivation: Bernstein ‘amber’ was borrowed in the early period of New High German from Middle Low German bern(e)stein, barnstēn, bornstēn ‘burning stone’ (attested from the 13th century), whose first element is the verb bernen ‘to burn’, itself considered to have arisen through metathesis of r from High German brennen ‘same’ (Kluge 1989: 76). However, W. Schmid is far from satisfied with this explanation, since it is only a root etymology and does not explain the formation of the Baltic word. With respect to derivation, W. Schmid further compares Lithuanian gintãras with Old Norse kyndill masc. ‘lantern; candle; light’ (from which the Finnish word mentioned above was borrowed) and supports the regularity of the formation (Baltic -r-/-ar-: Germanic -l-) by analogy with the following correspondences: Lithuanian mentūrė ‘tool for stirring porridge, mashing potatoes, etc.; tool for churning butter in a churn’: Old Norse mǫndull masc. ‘mill spindle of a hand mill’; Lithuanian vént eris ‘fishing net’: Old Norse vǫndull masc. ‘bundle of hay’; Lithuanian vingurys ‘bend (of a road, path, stream, etc.)’: German Winkel ‘angle’; Lithuanian kaugurys ‘low rise’ or kaukaras ‘hill, hillock’, or kaukaras ‘same’: German Hügel ‘hill, hillock’; Lithuanian nōterė ‘nettle’: German Nessel ‘same’; Lithuanian ungurys: Latin anguilla ‘same’ (Schmid 1993: 429). From this, W. Schmid concludes, it is clear that the long-associated Lithuanian gintãras and Old Norse kynda ‘to kindle’ amount to more than just a root etymology. The formation of gintãras is typically Baltic, although the verb has disappeared; the root survives in North Germanic languages. W. Schmid therefore asks where this word originated. The main amber-finding locations were the Sambia Peninsula and the Curonian Spit (Juodkrantė), places that were certainly not inhabited by Lithuanians and Latvians in earlier times. That these amber sites were most readily reached across a frozen sea or lagoon is shown by Latvian folk songs, which often mention dzintarzemē. Just as the word dzintars spread from the Courland coast to Vidzeme, so, presumably, gintaras reached the main Lithuanian settlement area from the Sambia coast. This means that the Baltic coast was the homeland of the early Baltic language (“eine Bodenständigkeit des Frühbaltischen”) and that its division into dialects is a later development. The early Baltic word gintãras belongs to the list of Baltic–Germanic correspondences (Schmid 1993: 430). KALBOTYROS PASTABOS 95 Baigiant dar galima pažymėti, kad gintaro pavadinimo variantui gentāras lietuvių kalboje randame homonimą – tai A. Juškos užfiksuotas gẽntaras, -ė ‘tinginys, slinkis’ LKŽIII 237 ir, reikia pasakyti, neaiškios darybos bei kilmės žodis. Galimas daiktas, su priesaga -ara- (dėl jos žr. Skardžius 1943: 302) jis yra pasidarytas iš veiksmažodžio gẽsti (-sta, gẽnto) ‘giminiuotis, draugėn dėtis, bičiuliautis’. Jų darybos ir antoniminių reikšmių santykis būtų toks pat kaip, pvz., gaĩmaras ‘gaišlys, delsūonis, vilkintojas; nevalgaus žmogus, nudvėsėlis’ šalia gaimìnti (-ina) ‘gimdyti, veisti; auginti’ LKŽIII 31–32; bandžiūlis ‘draugas, bičiulis; sukčius…’ LKŽI 643–644, bandžius ‘draugas, bičiulis; kerdžius…’ ar tiesiog bandà ‘būrys galvijų, kaimenė’ vedinio (kaip didžiūlis ‘didelis žmogus, didžiuoklis’: didis). Taip išanalizuoti ir paaiškinti gẽntaras, -ė ‘tinginys, slinkis’ bei gaĩmaras ‘gaišlys, delsūonis…’ patvirtina priesagos -ara- buvimą baltų kalbose. 18. LITH. DIALECT krontas ‘SNOWDRIFT’ In the vicinity of Varėna and in Rodūnia, Varanavas District (Belarus), according to LKŽVI 682, the noun krontas (unstressed) ‘a large pile of snow, a snowdrift’ is used. What is unusual about this word is that before tautosyllabic n there is a vowel o. According to the rules of Lithuanian phonotactics, only the vowel a can occur in this position. Incidentally, in Dubičiai, Varėna District, and in Nočia, Varanavas District (Belarus), we find krõnas used with the same meaning (LKŽVI 682). It would seem as though a consonant krõnast had recently been inserted into the form. The Latvian dialects in the northwest of central Latvia (Vidzeme) (Valmiera, Valka, Mazsalaca, Vecsalaca) have krān(t)s [krānc, krānts] (gen. sing. -s) ‘stove’, with the consonant t clearly inserted [MEII 266, 268 (s. v. II krāsns), EHI 645]. J. Endzelīns considers this word to have arisen through dissimilation from krāsns (Lithuanian krósnis), with the consonant s still retained when a vowel follows n, for example in the loc. sing. krāsni (MEII 266). However, the Lithuanian krontas clearly contradicts this explanation, suggesting that a consonant t was nevertheless added to the words Lithuanian krõnas, Latvian krāns ‘stove’ in Vecsalaca (MEII 266), and krāns ‘same’ in Ipiķi, Pļaviņas, Valka (EHI 645), although the reasons for its appearance remain unclear. The long root vowel ā here is etymological, since Latvian krāsns, Lithuanian krósnis, and also Latvian krāns and Lithuanian krõnas are derivatives with the suffixes -sn- and -n- from a verbal root represented by Latvian krāt (krāju) ‘to accumulate, load, gather’ (as, for example, Lithuanian plónas ‘earthen floor, base, flooring’, Latvian plāns ‘clay floor, flooring’—from plóti [-ója], Latvian plāt [plāju] ‘to lay, to layer’). On this occasion it is essential to emphasize that there are also further parallels linking the southernmost Lithuanian dialects with the Latvian linguistic area, partly with its northern dialects. For instance, in Lazūnai, Vija district (Belarus), instead of the common Lithuanian oak, the form úožuolas (Petrauskas, Vidugiris 1985: 275) is used; its initial diphthong corresponds to Latv. uozuõls, but because of the consonant ž it is considered to be of Lithuanian origin. In Zietela, Diatlovo district (Belarus), instead of bruvìs and žuvìs, the forms brīvas ‘eyebrow’ and zivis (zivē), živis (žive) are used—žiuvė is also said (Vidugiris 1998: 93, 791, 811)—and they have -iv-, which derives from -uv-, a characteristic phonetic change in Latvian dialects (cf. Latv. zivs, zive). Moreover, speakers of the Zietela dialect also pronounce them with the consonant z-. K. Būga also observed that just as in Ašmena County (now Belarus) there is the river Gàuja (a right tributary of the Nemunas) with its left tributary Vìja, so too in Latvian territory Gauja (which flows into the Gulf of Riga) has a left tributary Vija (Būga 1961: 630). 96 | SIMAS KARALIŪNAS 19. LAT. he, she ‘HE, SHE’ In standard Latvian and in most dialects—in all Leivu dialects and in the western part of the Augšzeme dialects—the third-person personal pronouns are, as is well known, viņš, feminine viņa ‘he, she’. This pronoun is regularly inflected as an io-stem (Endzelīns 1951: 517), that is, viņš, gen. sing. viņa… like ceļš, ceļa… (Lith. kelias, kelio…). But its older meaning is ‘that one, the other (German jener)’, well preserved in such fixed expressions consisting of a form of the pronoun viņš used attributively (!), and a noun in the locative with a spatial or temporal meaning (cf. viņu pusē ‘on the other side’, viņā saulē ‘in that world’) (Endzelīns 1951: 517; ME₍IV₎ 601); less often, the genitive, dative or accusative (cf. viņu gadu ‘last year’) (Bergmane e.a., 1959: 521). In these constructions, viņš has the feature ‘non-proximal (distal)’, like the pronoun tas. Since there are cases in which viņš is always replaced by tas, the use of viņš and tas stands in a relation of complementary distribution, and they are therefore considered positional variants (Rosinas 1988: 51). He, she are also used when referring to animate beings, whereas for inanimate objects, the third-person pronouns generally used are that, that. This feature of usage must be borne in mind when seeking to uncover the origins of the pronoun viņš: attributive usage also brings out further semantic features of the pronoun, preserved in fixed expressions. For example, viņu māja is equivalent in meaning to viņsēta and means ‘another, neighboring yard; another, neighboring homestead’; the plural form viņsētas means ‘other, neighboring homesteads; inhabitants of another, neighboring yard, more distant neighbors’ (ME₍IV₎ 601; Endzelīns 1951: 517). It is easy to see that here the pronoun viņš not only functions attributively but also has the meaning ‘other’. The word group viņu ļaudis contracts to viņlaudis, and in some local dialects even to viļaudis (nom. plur.), meaning ‘people from another, neighboring homestead, people from another yard, neighbors’ (ME₍IV₎ 601). There are even expressions in which the pronoun viņi directly denotes neighbors, e.g. viņu rija = kaimiņu rija ‘the neighbors’ homestead’ (literally ‘the neighbors’ threshing barn’); viņu tēvs, māte = kaimiņu tēvs, māte ‘the neighbors’ father, mother’ (ME₍IV₎ 601); viņuos (gen. plur.) guovs ‘the neighbors’ cow’ (EH₍II₎ 786; Endzelīns 1951: 517). Thus, both the attributive use of the pronoun viņš, fem. viņa, which presupposes its apparently older meaning ‘other (i.e. neighbor)’, and its inflection as a regular io-stem representative suggest that viņš, fem. viņa was an independent lexical unit that only recently entered the pronominal system¹. As has long been noted (Endzelīns 1951: 517), the demonstrative pronoun viņš, fem. viņa ‘other’ took over the function of the personal pronoun and replaced Latvian jis, ji in the same way that in Eastern Lithuanian dialects the demonstrative pronoun anàs, anà, becoming ānas, anà (plur. āņys, ānos), took the place of jis, ji. In the Pilda dialect of the eastern Upper Latvian dialects, the pronoun veņis, fem. veņei is used (EH₍II₎ 771). It has several distinctive features. First, some forms of this pronoun have long vowels in their endings. Second, the palatalized consonant -ņ- (-ń-) in all forms indicates that it reflects *-nj-. Third, the pronoun under discussion differs from viņš, fem. viņa in its root vocalism—it has e instead of i (Rosinas 1988: 211). Assuming that long vowels in endings could have been preserved only in monosyllabic forms, A. Rosinas considers the pronoun in question to consist of two components: 1 By origin it has traditionally been associated with Prussian winna adv. ‘outdoors, outside’, wins ‘air (Luft)’: Endzelin (1922: 381; 1951: 517–518); Stang (1966: 238–239); Rosinas (1988: 211); Karulis (1992: 534–535). But no one has provided an argumented explanation of the yawning semantic gap between ‘outside; air’ and the function of third-person pronouns (or is such an explanation impossible?). KALBOTYROS PASTABOS | 97 *vinE- (where E is some vocalic element) and the third-person pronoun *(j)is, *(j)ī. He derives the root vowel e from i. This is because in some Upper Latvian dialects there is a rule whereby i becomes e before j, ŋ (Rosinas 1988: 211). The features of the aforementioned Pilda pronoun venīs, fem. venēi should not be very surprising, given that, first, this pronoun with this root vocalism also occurs in other Upper Latvian dialects: veŋdien ‘the other day’ in Zvirgzdine, veŋ in the expression ek veŋ ‘look! (sieh da!)’ in Sausnēja and Dzelzava (ME₁ 538). Second, peripheral dialects, such as the dialects just mentioned, have often preserved more archaisms. For example, in the eastern Upper Latvian dialects (Aknīste, Susēja, Zvirgzdine, Kārsava, Pilda, Nirza, Dignāja), forms jer, jera ‘is’ with e in the root are used instead of or alongside ir ‘is’². It could therefore hardly be coincidental that in the Pilda and Zvirgzdine dialects both the demonstrative pronoun venīs, veŋ-, and the third-person forms of the verb “to be” are distinguished by root vocalism e. Moreover, the occurrence of e vocalism in these forms can scarcely be explained by the aforementioned tendency for vowel i to change to e before j, ŋ, since in the forms jer, jera the vowel e occurs before a velarized consonant and could not have arisen phonetically from i. In view of these circumstances, it is preferable to assume that the root vocalism e in the forms venīs, veŋ- and jer, jera is ancient and is related by ablaut to the root vocalism of the corresponding forms viņš, fem. viņa and ir, ira ‘is’. When examining how venīs, fem. venēi came about, little importance should be attached to the fact that some forms of this pronoun in the Pilda dialect have uncontracted endings characteristic of monosyllabic pronouns. Dialectologists have observed that shortening long vowels and diphthongs at the end of words is not very characteristic of Upper Latvian dialects (Rudzīte 1964: 293). Moreover, in the pronominal system the word-final vocalism, and even the position of stress, in disyllabic forms can be based on the inflection and stress of monosyllabic forms. This is clearly illustrated, for example, by the parallelism of katras (dial. kataràs), katrō, katrám… – kās, kō, kám… in Lithuanian. Similarly, the inflection of fem. venēi may be based on the endings of jei ‘she’. This principle of analogical formation, together with the Upper Latvian dialects’ tendency not to shorten long vowels and diphthongs in endings, would also explain the preservation of long vowels and diphthongs in some forms of venīs, fem. venēi. In other words, just as the parallelism of, for example, patī, pac̃ios, pac̃iā… – jī, jōs, jāi… does not indicate that patī is composed of *pat and jī, it is likewise hardly possible to infer from the absolute coincidence of the endings of venēi – jei ‘she’ that venēi is a compound form, with *(j)ī, *jās… as its second component. Moreover, if venīs is derived from the compound *vin- + *(j)is, its long ending can scarcely be explained—the vowel in the pronouns Latvian jis, Lithuanian jis < *(j)is is short. It seems, therefore, that here we nevertheless have feminine forms *venī/*venjās alongside masculine *venja-/*venija- (nom. sing. *venijas > *venijs (cf. loc. sing. vēnijā) > *venīs with generalized -ń- < *-nj- from *venja-), just as there would be feminine *patī/*patjās (> patī, pac̃iōs) alongside the i-stem masculine *patis (> pats). With all this in mind, it seems possible to reconstruct for Proto-Latgalian dialects a pronoun with alternating root *ven-/*vin- (< *un-), meaning ‘other (i.e. neighbor)’ and functioning with two stems: (1) a normal-grade root-vocalism (i)io-stem *uen(i)io-, fem. *uenī-/*ueniā- and (2) a zero-grade root-vocalism io-stem *uiṇio-, fem. *uiṇiā- (< *uṇ-io-, fem. *uṇ-iā-). 2 Ancītis (1935: 186); Endzelīns (1951: 719; 1980: 274) (balsio e atsiradimą aiškinti kontaminacija su rus. есть ar jo tiesioginė įtaka nėra pagrįstas dalykas). 98 | SIMAS KARALIŪNAS From a historical-comparative perspective, this Baltic pronoun is very likely related to Germanic nouns such as Old Icelandic vinr (modern Icelandic vinur), Norwegian ven, Swedish vän, Old English and Old Frisian wine, Old Saxon and Old High German wini ‘friend’ (< Germanic *weni-). Also belonging here from other languages are Old Irish coi-bnius ‘kinship’ and Old Breton co-guenou ‘local, native’. The development of Indo-European *uen-(i)jo-/*uṇ-io- ‘friend’ into Baltic *uen(i)ja-/*uinia- ‘neighbour, another (cf. Old Church Slavonic kaimynas)’ is apparently based on the idea of reciprocity, characteristic of the Indo-European system of exchange and the exchange of gifts and social services, and also carried the connotation of the other (stranger, enemy); cf. the Ancient Slavic drugb druga ‘one another’ (: drugb ‘friend’) and Hittite araš aran ‘one another’ (: araš ‘friend’), which have the same semantic structure. 20. IS PR. keckirs ‘PEA’ A LOANWORD? The Elbing (compiled, it is believed, in the late thirteenth or early fourteenth century) and S. Grunau (dating from 1517–1526) glossaries of German–Prussian contain, respectively, keckers (cf. also lituc-kekers ‘lentil’) and keckirs ‘pea (Erweis, Erbse)’ (Mažiulis 1981: 24, 52). Opinio communis holds that this is a borrowing from German (German Kicher, Middle High German kicher, from Latin cicer, neuter ‘pea’)³. In the northeastern corner of Lithuania, on the basis of the cluster of hydronyms, it is possible to reconstruct a situation of linguistic-ethnic contacts whose components may have been languages of the East Baltic and West Baltic types. West of Zarasai, near the Latvian border, there is a group of lakes with unusual names: Čičirys (acc. sing. Čičiri) (and Čičiris) in Imbradas, Čičiraitis in Imbradas, and the river Čičirė, which flows into Lake Čičiris, in Antazavė. A little farther west, in the vicinity of Antazavė, is a lake called Zirnajys (acc. sing. Žirnajį) (and Zirnejas). In its place, to the south in the East Aukštaitian area, occurs the river Žirnaja (acc. sing. Žirnąją) (rising in Lake Žirnajai) in Pabaiskas; Lake Žirnajai (acc. plur. Žirnajus) in Pabaiskas, Ukmergė District; and Lake Žirnajys in Tauragnai, Utena District. Comparison with the lake name Žirnis in Šeduva makes it possible to identify the suffix -aj- in these names (Vanagas 1970: 80) and to establish that their etymon is the appellative žirnis (Vanagas 1970: 404); because of the consonant ž, these names must be considered Lithuanian in origin. Since Zirnajys and Zirnejas are justifiably regarded as remnants of Selonian, both their suffix -aj- and their etymon žirnis are apparently relics of Selonian rather than Latvian, since Selonian had the consonants s and z in place of Lithuanian š and ž (on Selonian, see Būga 1961: 274–282; Kuškis 1967: 9–20; Karaliūnas 1972: 17–19; Rudzīte 1980: 159–163; Rudzīte 1996: 45–50; Kabelka 1982: 81–85; Breidaks 1999: 27–28). Also considered remnants of Selonian are Čičirys (and Čičiris) and Čičirė, whose original form is reconstructed as *čičir- < Baltic *kikir- (Būga 1959: 22, 108; 1961: 275, 733; Vanagas 1981: 7; Kabelka 1982: 84). Interestingly, Čičirys, Čičiris, and Čičirė, on the one hand, and Zirnajys and Zirnejas, on the other, are geographically adjacent; this makes it plausible that the first group of names was also formed from the lexeme for ‘pea’, namely Selonian *čičir- < Baltic *kikir- ‘pea’. In other words, 3 Būga (1959: 219; 1961: 661, 681); Endzelīns (1943: 191); Schmalstieg (1976: 264); Sabaliauskas (1957: 351); Toporov (1980: 302–304); Mažiulis (1993: 153–154). R. Trautmannas plg. keckers, keckirs sulugyino su lot. cicer ir arm. sisem ‘žirnis’, tačiau neatmetė jų skolinimosi iš lenkų kalbos galimybės (lenk. cieciorka), Trautmann (1910: 355). KALBOTYROS PASTABOS 99 Čičirys, Čičiris, and Čičirė can be treated as the result of translating Zirnajys and Zirnejus. Thus, there is sufficient basis to reconstruct in Selonian, alongside the lexeme *kikir- ‘pea’, the lexeme *kikir- ‘pea’, whose etymological counterpart is Prussian keckers, keckirs ‘pea’. In fact, the quality of the root vowel in these Prussian words is unclear (whether the vowel is e or i), but it is easy to see that cases of fluctuation between i and e occur in Prussian; cf., for example, the place names Bliwoth and Blewothyn, Myluken and Meluken (Gerullis 1922: 215). In particular cases, one even observes the raising of vowel i to e; cf. Prussian lankenan deinan ‘a day being celebrated’ in the I Catechism: länkinan deinan ‘the same’ in the III Catechism; Prussian leygenton ‘would judge’ in the II Catechism: liginton ‘the same’ in the III Catechism; Prussian kamenis: Lithuanian kaminas; Prussian warene ‘copper (cauldron)’: Lithuanian varinė ‘copper basin’ (from K. Sirvydas’ Punktai sakymų). Thus, Selonian must have had the lexeme *kikir- ‘pea’, characteristic of West Baltic languages (Karaliūnas 1990a: 86–87). This should not be surprising, since there are other lexical isoglosses linking the northeastern corner of Lithuania (Dusetos, Zarasai) with the West Baltic linguistic area: Lake Vėnčavas in Dusetos and the Yotvingian Wensowa forest, recorded in 1562 in Olecko County, Masuria (Karaliūnas 1991: 73–77). V. Toporov has reconstructed an analogous contact situation in the territory of Moscow, by the Yauza River, one of whose tributaries is called Čėčera (Чечера). V. Toporov associates this river name with Čičirys (and Čičris), Čičiraitis, and Čičirė, and considers it to be of Baltic origin. Since the Čėčera flows through a Moscow district known since antiquity as Гопоховое поле (‘Pea Field’: Russian горох ‘pea’), V. Toporov convincingly regards this name as a translation of Baltic *kikir-/*kiker- ‘pea’ (Toporov 1982: 3–61). With his customary thoroughness, V. Toporov presents archaeological, linguistic, and historical evidence showing that Baltic tribes lived in the Upper Volga and Oka basins, maintaining their culture and, in some places, their language until the twelfth century; the existence of a period of Baltic-Russian bilingualism cannot be entirely ruled out. Conclusion: Baltic *kikir- ‘pea’ was widespread across a large area of northeastern Europe, from the Baltic Sea in the west to the upper Volga and Oka in the east; therefore, Prussian keckers, keckirs ‘pea’ cannot be a borrowing from German. Prussian keckers, keckirs ‘pea’, Selonian *čičir-<Baltic *kikir- ‘pea’, Latin cicer, -is ‘the same’, and Armenian sisern ‘the same’ (< Indo-European *kiker-) form a Baltic, Italic, and Armenian isolex. 21. WHAT DID PR. rapa MEAN? In the Old Prussian–German Elbing Vocabulary, the word dewis ‘god’ is followed by rapa ‘Engel’ (Mažiulis 1981: 14). It is attested that, like Lithuanians and Latvians, the Prussians used loanwords for this important concept in religious and spiritual life: S. Grunau gives angol ‘angel’ in his vocabulary (a word probably borrowed through Polish), while engels (borrowed from German, cf. German Engel) is found in the third catechism (Mažiulis 1981: 49, 165, 168. For the historical and etymological context of all these forms, see Endzelīns 1943: 141, 167, 238; Mažiulis 1988: 77–78, 265; 1997: 12; Toporov 1975: 87; 1979: 41–42). In light of these facts, rapa appears to be a rather unexpected and obscure word for ‘angel’; it is therefore difficult to understand why the Prussians would have needed to coin a new word, and one suspects that something may be amiss. F. Nesselmann has mentioned that W. Pierson had already noticed that the gloss Engel – rapa had been misread: instead of reading German Eugel ‘toad’, it had been read as Engel ‘angel’; he connected Prussian rapa ‘toad’ with Lithuanian raupėžė, repečkā, rapūže, rūpužė, and so 100 | SIMAS KARALIŪNAS pat su lat. ràpt ‘ropoti’ (Nesselmann 1873: 146). Įdomu, kad prūsų kalbos tekstuose raidės u ir n rašant taip pat kartais painiojamos (Endzelīns 1935: 103 [=Endzelīns 1980: 233]). „Vokiečių kalbos atlaso“ leksikai skirto tomo (Band IV, Gießen, 1955) žemėlapyje Nr. 4 yra suminėtos tokios rupūžės pavadinimo tarminės formos: Oigel, Oig(e), Oik(e), Ogel, Ockel⁴. Ypač svarbu, kad forma Oigel vartojama Westfalijoje, iš kur į prūsų kraštą buvo atsikėlę nemaža vokiečių kolonistų. Todėl H. Schallis visiškai teisus sakydamas, kad Elbingo žodynėlio glosoje Engel – rapa yra pr. ràpa ‘rupūžė’, tikriau sakant, ‘ropojantis gyvūnas’, ir jį siedamas su latvių kalbos veiksmažodžiais rāpuot, ràpt(ies) ‘ropoti, réplioti, šliaužti’, rāpus ‘ropomis, réplom’ (Schall 1966: 28). Minėto atlaso rupūžių pavadinimų arealus žymintis žemėlapis, be to, nurodo, kad šiaurės rytų Prūsijos kampe, į rytus nuo Kuršių marių, vokiečių tarmėse buvo vartojamas ir žodis Rapetschke su variantu Repetschke. H. Schallis mano, kad čia, skalvių plote, pr. ràpa išsilaikęs ligi pat naujųjų laikų (Schall 1966: 28–29). Nors ir nėra pakankamai aiškių fonetinių kriterijų atskirti prūsiškos kilmės vokiečių kalbos skolinius nuo jų lietuviškos kilmės, pastarosios dvi formos vis dėlto veikiau laikytinos į vokiečių kalbos tarmes Rytų Prūsijoje atėjusiomis iš lietuvių kalbos šnektų (Bielfeldt 1970: 48). Akademinio Lietuvių kalbos žodyno ir Lietuvių kalbos atlaso duomenimis, rapėčka LKŽₓ₁ 177, repėčka, repėčka, repėčke, repėčė ‘rupūžė’ LKŽₓ₁ 460 vartojami pačiame pietų vakarų lietuvių kalbos ploto pakraštyje (Lietuvių kalbos atlasas, I, 1977: 153, žemėl. Nr. 81), kuris ligi pat II pasaulinio karo ribojosi su vokiečių kalbos arealu. Jų darybos pamatas yra būdvardžiai repėckas, repėčkas, repėčkas ‘einantis visomis keturiomis, keturpėsčias’ LKŽₓ₁ 460, plg. dar prieveiksmius repėčka, repėčka, repėčka, repėčka ‘visomis keturiomis, ropomis’. In Lithuanian dialects, forms of the name for ‘toad’ with root vocalism o (<ā) are also found: dialectal ropūžė in Pelesa; ropūžė in A. Kuršaitis’s dictionary, ‘toad (Bufo)’ LKŽₓ₁ 841; and ropūžė in the Lazūnai dialect (Petrauskas, Vidugiris 1985: 225), alongside rapušė, rapužė with short root a in the Zietela dialect (Vidugiris 1998: 530). Their derivational base is apparently the verb ropóti (-ója) ‘to move, crawl (usually of insects and reptiles)’, cf. also the adverb ropà ‘on all fours, crawling’. The presence of forms with root vocalism o (<ā) in Lithuanian dialects, and their concentration toward the southern periphery, in the historical Yotvingian and Prussian area, makes the reconstruction of the word rāpa in Prussian entirely possible. V. Mažiulis does note, however, that the ā-stem form rapa (nom. sing. fem.) in the Elbing Vocabulary would have been written *rapo (with final ā changing to o, as in, for example, galwo, mergo), and therefore he emends rapa to *rapan (Mažiulis 1997: 12). Nevertheless, J. Endzelīns’s logical explanation should not be discounted: rapa, if it is a genuine word, may have come, along with manga ‘prostitute’ and merga ‘girl, young woman’ in S. Grunau’s vocabulary, from those Prussian dialects in which ā was not rounded after labial and velar consonants (Endzelīns 1943: 62). Thus there may also have been Prussian rāpa (<*rāpā). But if we consider the alternative, it seems more likely that Prussian ràpa had the form *ràpaz-, and so on, and therefore had the suffix *-az- (Lithuanian -až-). Suffixes -ėž- (/*-až-)/-ėž- can be distinguished in Lithuanian names for ‘toad’ such as rėpežė, repėžė, rėpežis ‘toad’ LKŽₓ₁ 466: rėpénti (-ėna) ‘to crawl’ (cf. also repetóti [rėpetoti, repetoti] ‘to crawl; to walk on all fours…’); Lithuanian bebėžė (and bėbėžė) ‘toad; louse’ LKŽ₁ 705: bėbė ‘louse (in children’s speech)’ 4 Šis šaltinis Lietuvoje neprieinamas. Cituojama pagal Schall (1966). KALBOTYROS PASTABOS 101 kr(i)upėža ‘a toad, a little toad (mild curse)’ LKŽVI 651, 716: kr(i)ūpis, kr(i)ūpė. There are other suffixes too, for example -ek-/-ėk- in the words repėka, repėkė, rėpekė LKŽXI 463; especially common are the suffixes -už-/-ūž-: rāpuža, rāpužė, rapūžė (used as a term of abuse) LKŽXI 180, alongside rapóti (-ója) ‘to walk with difficulty, crawl; to creep’; repūžė (rėpužė) LKŽXI 475: repénti (-éna) ‘to crawl’; kūkužė (and kukužė) ‘toad (a milder word of abuse)’ LKŽVI 818: kukūtė ‘an amphibian somewhat resembling a toad’. Therefore, Latvian place names Rapa, Rapas (Dambe 1972: 58) should now be explained differently—that is, they should not be connected with Prussian rapa. They seem to be indistinguishable from the Latvian surnames Rapa, Raps (< *Rapas). The Lithuanian surnames Rapas and Rapis (unstressed), found around Rokiškis, are also of Latvian origin (cf. Vanagas, ed., 1989: 573–574). Their derivational base may be the Latvian verb iz-rapēt ‘to rot, decay, become porous’ MEI 789, cf. Lithuanian rāpti (rãmpa, rāpo) ‘to be unwell’ LKŽXI 178. Conclusion: the gloss Engel – rapa in the Old Prussian–German Elbing Vocabulary was probably misread: instead of German Eugel ‘toad’, it was read as Engel ‘angel’. LITERATŪRA ANČITIS, K. 1935: Par Aknīstes izlok­sni. Filologu biedrības raksti 15, 161–196. ANTTILA, R. [АНТТИЛА, Р.] 1997: Недостающие звенья в лексических цепочках. Балто-славянские исследования 1988–1996, Москва, 109–127. BENKŐ, L., Hrsg., 1993: Etymologisches Wörterbuch des Ungarischen I, Budapest: Akadémiai Kiadó. BERGMANE, A. e. a. 1959: Mūsdienu latviešu literārās valodas gramatika. I. Fonētika un morfoloģija, Rīgā: Latvijas PSR Zinātņu Akadēmijas izdevniecība. BIELFELDT, H. 1970: Die baltischen Lehnwörter und Reliktwörter im Deutschen. RÜKE-DRAVINA, V., ed., Donum Balticum. To Professor Christian S. Stang on the Occasion of his Seventieth Birthday 15 March 1970, Stockholm: Almqvist & Wiksell, 44–56. BLESSE, E. 1958: Zum lett. uguns ‘Feuer’. KZ 75 (3/4), 191–206. BREIDAKS, A. 1999: Ievads baltu valodniecībā, 2. daļa, Daugavpils. BŪGA, K. 1959, 1961: Rinktiniai raštai 2–3, Vilnius: Valstybinė politinės ir mokslinės literatūros leidykla. CHANTRAINE, P. 1968: Dictionnaire étymologique de la langue grecque, Paris: Klincksieck. DAMBE, V. [ДАМБЕ, В.] 1972: Соответствия в топонимии Латвийской ССР с древнепрусским языком. Baltistica I priedas, 55–62. DEROY, L., HALLEUX, R. 1974: A propos du grec ἤλεκτρον ‘ambre’ et ‘or blanc’. Glotta 52, 37–52. DROTVINAS, V., red. 1987: Lexicon Lithuanicum. Rankraštinis XVII a. vokiečių–lietuvių kalbų žodynas, Vilnius: Mokslas. ENDZELĪNS J. 1913–1914/1974: Zu den kurischen Bestandteilen des Lettischen. Indogermanische Forschungen 33, 96–104 [=id., Darbu izlase II, 504–512]. ENDZELIN, J. 1922: Lettische Grammatik, Rīga: Kommissionsverlag A. Gulbis. ENDZELĪNS, J. 1935/1980: Prūšu tekstu grafika. Filologu biedrības raksti 15, 86–103 [=id., Darbu izlase III (2), Rīga, 1980, 211–233]. ENDZELĪNS, J. 1940/1980: [Recenzija] Die Kurenfrage von V. Kiparsky (Annales academiae scientiarum Fennicae B XLII). Helsinki, 1939. 474. Filologu biedrības raksti XX, 248–258 [=id., Darbu izlase III (2), Rīga, 1980: 553–565]. ENDZELĪNS, J. 1943: Senprūšu valoda, Rīga: Universitātes apgāds. ENDZELĪNS, J. 1951: Latviešu valodas gramatika, Rīga: Latvijas Valsts izdevniecība. EH – Endzelīns J. un Hauzenberga E. 1946: Papildinājumi un labojumi K. Mīlenbacha Latviešu valodas vārdnīcai, II, Rīga: Kultūras fonds. 102 SIMAS KARALIŪNAS ERNOUT, A., MEILLET, A. 1959: Dictionnaire étymologique de la langue latine, 4ᵐᵉ éd., Paris: Klinck-sieck. FRAENKEL, E. 1962: Litauisches etymologisches Wörterbuch I, Heidelberg: Winter, Göttingen: Vandenhoeck & Ruprecht. FRISK, H. 1970: Griechisches etymologisches Wörterbuch II, Heidelberg: Winter. GĀTERS, A. 1979: Dzintarzeme (Bernsteinland) in den Dainas und die lettische Bezeichnung für Bernstein. Humanitas Religiosa. Festschrift für Haralds Biezais, Stockholm: Almqvist & Wiksell, 283–294. GERULLIS, G. 1922: Die altpreußischen Ortsnamen, Berlin und Leipzig: Vereinigung Wissenschaftlicher Verleger. KABELKA, J. 1982: Baltų filologijos įvadas, Vilnius: Mokslas. KARALIŪNAS, S. 1972: Sėlių kalba. Mokslas ir gyvenimas 1, 17–19. KARALIŪNAS, S. 1990a: Etymological Notes on Lithuanian Hydronyms. Beiträge zur Namenforschung. Neue Folge 25(1), 85–88. KARALIŪNAS, S. 1990b: Notes on Latvian Etymology. Metuzāle-Kangere, B. & Rinholm, H., eds., Symposium Balticum. A Festschrift to honour Professor Velta Rūķe-Draviņa, Hamburg: Helmut Buske, 177–180. KARALIŪNAS, S. 1991: Kalbotyros pastabos. Lietuvių kalbotyros klausimai 29, 65–79. KARULIS, K. 1992: Latviešu etimoloģijas vārdnīca, II, Rīga: Avots. KIPARSKY, V. 1939: Die Kurenfrage, Helsinki (Annales academiae scientiarum Fennicae B XLII). KIPARSKY [КИПАРСКИЙ], V. 1973: О балтизмах русского литературного языка. Baltistica 9 (1), 67–70. KLUGE, F. 1989: Etymologisches Wörterbuch der deutschen Sprache. 22. Auflage völlig neu bearbeitet von Elmar Seebold, Berlin–New York: De Gruyter. KUŠKIS, J. 1967: Dažas sēļu valodas vokālisma īpatnības pēc mūsdienu latviešu valodas sēlisko izlokšņu materiāliem. Latvijas Valsts universitātes Zinātniskie raksti, 60. sējums, 9. A laidieņs, Rīga, 9–20. LARIN, B. A. 1956: О слове янтарь. Sokol̦s, E., ed. Rakstu krājums. Veltījums akadēmiķim profesoram Dr. Jānim Endzelīnam viņa 85 dzīves un 65 darba gadu atcerei, Rīga: Latvijas PSR Zinātņu Akadēmijas izdevniecība, 149–162. LAUČIŪTĖ, J. [ЛАУЧЮТЕ, Ю. А.] 1982: Словарь балтизмов в славянских языках, Ленинград: Наука. Lietuvių kalbos atlasas I. Leksika, Vilnius: Mokslas, 1977. LKŽ = Lietuvių kalbos žodynas I, red. J. Balčikonis e.a., Vilnius, 1941–. 2LKŽ = Lietuvių kalbos žodynas I, antras leidimas, red. J. Kruopas, Vilnius, 1968. MAYRHOFER, M. 1963: Kurzgefaßtes etymologisches Wörterbuch des Altindischen, II, Heidelberg: Winter. MAYRHOFER, M. 1996: Etymologisches Wörterbuch des Altindoarischen, II, Heidelberg: Winter. MAŽIULIS, V. 1988, 1993, 1997: Prūsų kalbos etimologijos žodynas I, Vilnius: Mokslas; II, Vilnius: Mokslo ir enciklopedijų leidykla; IV, Vilnius: Mokslo ir enciklopedijų leidybos institutas. MAŽIULIS, V. 1981: Prūsų kalbos paminklai, II, Vilnius: Mokslas. ME = K. Mülenbacha Latviešu valodas vārdnīca. Rediģējis, papildinājis, nobeidzis J. Endzelīns. IV, Rīga: Kultūras fonds, 1929–1932. MIKKOLA, J. J. 1938: Einiges über den eurasischen Bernsteinhandel. Senatne un māksla 1, Rīga, 33–37. MUCH, R. 1937: Die Germania des Tacitus, Heidelberg: Winter. MUCH, R. 1967: Die „Germania“ des Tacitus. Dritte, beträchtlich erweiterte Auflage, unter Mitarbeit von Herbert Jankuhn herausgegeben von Wolfgang Lange, Heidelberg: Winter. NESSELMANN, G. H. F. 1873: Thesaurus linguae prussicae. Der preussische Vocabelvorrath, Berlin: Ferdinand Dümmler. PAKALKA, K. e. a., 1979: Pirmasis lietuvių kalbos žodynas, Vilnius: Mokslas. PETRAUSKAS, J., VIDUGIRIS, A. 1985: Lazūnų tarmės žodynas, Vilnius: Mokslas. KALBOTYROS PASTABOS | 103 Rix, H., Hrsg. 1998: Lexikon der indogermanischen Verben. Die Wurzeln und ihre Primärstammbildungen, Wiesbaden: Ludwig Reichert Verlag. Rosinas, A. 1988: Baltų kalbų įvardžiai, Vilnius: Mokslas. Rudzīte, M. 1964: Latviešu dialektoloģija, Rīga: Latvijas Valsts izdevniecība. Rudzīte, M. 1996: Latviešu valodas sēliskās izloksnes. Latvijas Zinātņu Akadēmijas Vēstis, A daļa, 50(4/5), 45–50. Rudzīte, M. [Рудзите, М. К.] 1980: К вопросу о селах на правобережье Даугавы. Цимерманис, С., ред., Этнографические и лингвистические аспекты этнической истории балтских народов, Рига: Зинатне, 159–163. Sabaliauskas, A. 1957: Dėl žirnio pavadinimo kilmės. Literatūra ir kalba 2, 346–355. Schall, H. 1966: Baltische Gewässernamen im Flußsystem „Obere Havel“. Baltistica 2(1), 7–42. Schmalstieg, W. R. 1976: Studies in Old Prussian, University Park and London: Pennsylvania State University Press. Schmid, W. P. 1983: Zum baltischen Dialekt auf der Kurischen Nehrung. Indogermanische Forschungen 88, 257–268. Schmid, W. P. 1989: Nahrungskurisch. Sprachhistorische und instrumentalphonetische Studien zu einem aussterbenden Dialekt, Wiesbaden: Franz Steiner Verlag. Schmid, W. P. 1993: Bemerkungen zu litauisch gintaras ‘Bernstein’. Grinda, K. R., Wetzel, C.-D., Hrsg., Anglo-Saxonica. Beiträge zur Vor- und Frühgeschichte der englischen Sprache und zur altenglischen Literatur. Festschrift für Hans Schabram zum 65. Geburtstag, München: Fink, 425–430. Seldeslachts, H., Swiggers, P. 1995: Remarks on Baltic, Slavic and Latin tool names. Linguistica Baltica 4, 23–33. Sittig, E. 1934–1935: Die Aestier. Studi Baltici 4, 18–20. Skardžius, P. 1943: Lietuvių kalbos žodžių daryba, Vilnius: Lietuvių kalbos institutas. Sławski, F. 1952: Słownik etymologiczny języka polskiego I, Kraków: Towarzystwo Miłośników Języka Polskiego. Stang, Chr. S. 1966: Vergleichende Grammatik der baltischen Sprachen, Oslo–Bergen–Tromsö: Universitetsforlaget. Toporov, V. N. 1975, 1979, 1980: Прусский язык. Словарь (А–Д), (Е–Н), (I–К), Москва: Наука. Toporov, V. N. 1982: Древняя Москва в балтийской перспективе. Балто-славянские исследования 1981, Москва, 3–61. Trautmann, R. 1910: Die altpreußischen Sprachdenkmäler, Göttingen: Vandenhoeck & Ruprecht. Trautmann, R. 1925: Die altpreußischen Personennamen, Göttingen: Vandenhoeck & Ruprecht. Trubačev, O. N. 1980: Из балто-славянских этимологий. Этимология 1978, Москва, 7–18. Vanagas, A. 1970: Lietuvos TSR hidronimų daryba, Vilnius: Mintis. Vanagas, A. 1981: Lietuvių hidronimų etimologinis žodynas, Vilnius: Mokslas. Vanagas, A., red., 1989: Lietuvių pavardžių žodynas II, Vilnius: Mokslas. Vasmer [Фасмер], M. 1978: Этимологический словарь русского языка IV, Москва: Прогресс. Vidugiris, A. 1998: Zietelos šnektos žodynas, Vilnius: Mokslo ir enciklopedijų leidybos institutas. Voigt, V. 1970: Hungaro-Baltica. Rüke-Dravina, V., ed., Donum Balticum. To Professor Christian S. Stang on the Occasion of his Seventieth Birthday 15 March 1970, Stockholm: Almqvist & Wiksell, 574–580. Walde, A., Hofmann, J. B. 1938: Lateinisches etymologisches Wörterbuch I, Heidelberg: Winter. Simas Karaliūnas Lietuvių kalbos institutas Antakalnio g. 6, 2055 Vilnius, Lietuva Gauta 2001 06 18