Etymologische Bemerkungen zu lit. „kãmanos“, „kamíenas“, „kumẽlė“

ACTA LINGUISTICA LITHUANICA XLIX (2003), 5 – 31 Etymological remarks on Lithuanian kāmanos, kamiėnas, kumėlė BERND GLIWA Sargeliai New material is provided for an etymology of Lith. kāmanos ‘bridle’, kamiėnas ‘stem, butt-end (of a tree, timber)’, kumėlė ‘mare’, kamantinis ‘riding horse; hop’ etc. Special attention is given to the practical, ethnological side of the problem, neglected by earlier etymologists. The semantic derivation of these words in relation to distinct IE roots *(s)kem- ‘press, hinder, squeeze’, *(s)kem- ‘bend, curve, bow’, is discussed. The correspondence of Balt. k-, Germ. h- and Sl. ch-, probably from IE *(s)k- is outlined. Kāmanos is seen as a deverbal derivation from Balt. *kam- ‘domesticate’ < IE *(s)kom- ‘press, hinder’ with an initial meaning *‘domestication; tool used for domestication’. It is argued that Lith. žaboti ‘put a bit between the horse’s teeth, curb, check’ is related to žābtas ‘mouth...’, rather than to žābas ‘twig...’. Etymologies of kāmanos ‘bridle’ based on an alleged meaning *‘equipped with wooden bits’ are rejected due to practical requirements and archaeological data. Based on the same verb Balt. *kam- ‘domesticate’, kumėlė ‘mare’ could have originally meant *‘domesticated horse’. Kamiėnas ‘stem, butt-end’ could refer to the concave curve the stem’s surface shows in axial direction near the root < IE *(s)kem- ‘bend, curve, bow’. 1. Būga compares Lithuanian kāmanos ‘harness’ with Latvian kamanas ‘sleigh’ and explains this as follows: “In ancient times the Aukštaitians called a felled tree trunk together with its roots a kāmana [...] A trunk with roots, like two horns (hence: ragės arba rogės ‘sleigh’)” (Būga 1959: 295). The technical aspect is simple yet ingenious where the sleigh is concerned: the trunk provides a straight runner, and the root produces a durable curve. Suitable trees are easy to find, and one need not, like the fox in the folktale, implore a tree to assume a suitable shape: “Bend, bend, little tree in the whole forest, bend, bend... And it bent. The fox then made herself a little sleigh.” (Slančiauskas 1974: 65). This is merely to point out that kumpti ‘to become bent’ is suitable for describing the shape of a runner, as is lėnkė ‘sleigh runner’ (for further examples, see Morkūnas 1977: 101). Būga’s further arguments are less convincing: “That the ancient kamanų had nothing to do with leather and iron is shown by the saying kamanoms reikia žaboty’, i.e. to put a žabą: žabą ‘twig’ into the horse’s mouth...” (Būga 1959: 295). It is certainly possible and usual for derivations or verbs to be formed on the basis of a noun to describe actions performed with the concrete object. In an individual case, however, this must always be examined, especially where there are semantic difficulties, and possible homophony must be questioned. Lithuanian žābas 6 | B E R N D G L I W A ‘branch; handle; switch; shrub, thicket; *bit’; figuratively: ‘anything that is dry’ (LKŽₓₓ 15f.) is the basis for, for example, žābaras ‘thin branch, stick; handle; shrub; switch; splinter’ (LKŽₓₓ 11f.), as well as žābyti ‘to strike, beat’ (LKŽₓₓ 18), ž̇abūoti ‘to cover with branches; to beat with a switch; to chop up branches’. It is questionable, however, whether žābas ‘branch; handle; switch...’ is also the basis for žābōti ‘to put a bit into a horse’s mouth; to tame, control; to put a rope or straw into the mouth of a tame bovine; to drive a wheel against something; to spread a net; to gape’ (LKŽₓₓ 19f.) and žābinti ‘to enclose, fasten’, ¹ž̇abūoti ‘to harness a horse’, žābāngai ‘trap; bit on a bridle; intrigues’ (LKŽₓₓ 9). Words that certainly denote a bit (possibly alongside other meanings) are, at best, derivatives of žābōti and not of žābas: žābōkliai, žābōklai, žābōklis, žābōklės, among others (LKŽₓₓ 19), as well as žābotinė ‘a kind of rein’ (LKŽₓₓ 23). If these referred to a particular kind of branch, perhaps one used for a specific purpose, one would expect a designation formed directly from žābas. Moreover, žāslai ‘bit; muzzle’ (LKŽₓₓ 208) is a widespread term for a bit. It can be compared with žāndas ‘jawbone, jaw, cheek’ (Fraenkel 1965: 1289, cf. Pokorny 1994: 381f.). Furthermore, since one also finds Lithuanian žabtà ‘muzzle, mouth, edge thereof; mouth’, žābtai ‘ditto; a tool for squeezing something together; part of the harness to which the reins are attached’, ¹žābtas ‘muzzle, mouth; cheek, face; food’, žābùčiai ‘a gag for a piglet being weaned from sow’s milk’, žābùkas ‘bit, harness’, žābùkliai ‘bit’, žābùklai ‘ditto; trap’, žābulai ‘mouth’, ²žābtas ‘glutton; pushy person’, ³žābtas ‘greedy, pushy in speech’, žābùs ‘gluttonous’, žābtùs ‘sharp-tongued; talkative; spirited’ (LKŽₓₓ 23ff.), žāplė ‘gums’, žāplos ‘bit on a bridle’, žāpsėti ‘to peck greedily, gobble, figuratively: to eat’ (hardly onomatopoeic) (LKŽₓₓ 162), and with a nasal: žamba ‘muzzle’, žambai ‘tonsils, adenitis colli’² (LKŽₓₓ 151), it is very plausible to posit a semasiologically analogous formation here, as is considered certain for žāslai ‘bit; muzzle’. In that case, it has nothing to do with any branches supposedly serving as a bit, but rather with the fact that the bit is put into the mouth and connected at the sides of the mouth with the rest of the harness and the reins. This requires either the assumption of a Baltic verb *žab- ‘to gape, open wide (e.g. of the mouth)’, on the basis of which Lithuanian žābtas *‘something gaping, opened wide’ would be a deverbal *-to- formation close to a passive participle. I would also connect žābāngai with this, contrary to Fraenkel (1965: 1282f.), who relates it to žābas³, with the analogous meaning *‘something gaping open’, since whether it is a snare, a net or a clapping- 1 The meaning *‘bit on a bridle’ is reconstructed and comes from the linguistic, that is, metalinguistic, writings of Būga and Salys; it is therefore not a linguistic fact in the strict sense. 2 Siehe auch Pokorny (1994: 369, 382), dessen Frage „ob gémbh- ‘beißen’ als nasalierte Form zu [gép(h)-, gébh- ‘Kiefer, Mund; essen, fressen’] gehört?“ (1994: 382) hier ein bejahendes Indiz erhält. 3 How one might construct a trap out of branches is beyond my practical understanding, unless what is meant here is a pitfall covered with branches, which would more likely be named after the pit or the falling in (cf. also Kluge 1999: 247) than after the branches. Lit. kāmanos, kamienas, kumelė | 7 In the former case, this is initially open. Or one may consider 3žabti ‘to press firmly’ and reconstruct *žab- ‘to press forcefully, bite’ > žābtas ‘*something that bites, presses’ > ‘snout’, whereby a similar development could also lead to ‘trap’. Specifically, however, the *-to formation argues against such a development for žābtas ‘snout...’. At this point I would like to discuss two Prussian words, sbeclis – veder (E 539) and slango – gebys (E 452), for which a new etymology has recently been proposed, namely derivation as a borrowing from German (Smoczyński 2000: 198–201). The criticism of the previously established comparison with Lithuanian žabóklai ‘bit’ (cf. Mažiulis 1997: 78) is based, first, on semantic considerations: if one accepts the above-criticized development from žābas ‘branch, stick’ > žabóklai ‘bit’, a development from this to sbeclis ‘spring in a lock’ remains difficult to explain. Secondly, the word-initial consonant cluster sb- (with German sound value), phonologically corresponding to žb-, is said to have no parallels in Baltic and is therefore interpreted as sp- (cf. Smoczyński 2000: 198). I would dispute this argument, since it is not justified to cite sbeclis first as an isolated instance of sb-, if shortly afterward slango is also presented as a scribal error for *sbango (Smoczyński 2000: 201). Thus, there are two instances of sb-, which support one another. In my view, the lack of parallels in other Baltic languages is explained by a specifically Prussian feature: the loss of the (unstressed) vowel of the first syllable, as, for example, in the place name Parengel (1298), > later Prengel (Smoczyński 2000: 24, 43). As for the semantic development, the foregoing considerations show that, morphologically, Prussian slango ‘bit’ corresponds to Lithuanian žabángai ‘trap’, and Prussian sbeclis ‘spring in a lock’ to Lithuanian *žabēklis (Mažiulis 1997: 78), with an almost diametrically opposed distribution of meanings. If we set Lithuanian žābas ‘branch, stick’ aside and compare semasiologically with Lithuanian spęsti ‘to stretch; to set a trap’, spąstai, spąslai ‘trap’, spendėlė ‘spring in a lock’ < and Indo-European *(s)pen-d- ‘stretch, pull, spin’ (Fraenkel 1965: 862, 865), we see that such a development is indeed quite common, and that the above *žab- ‘to gape, open wide (e.g. of the mouth)’ might perhaps be better posited as ‘(to) stretch’. Thus, the traditional comparison with the Lithuanian cognates should be maintained, but strictly distinguished from žābas ‘branch, stick’. Lithuanian ¹žabúoti ‘to bridle a horse’ and žabóti ‘to put a bit in a horse’s mouth; tame, control; put a rope or straw in the mouth of a tame ox; collide with something on a bicycle; spread a net;’ belong to the family of žabùs ‘greedy’, ¹žābtas ‘mouth...’, whereas ²žabúoti ‘to cover with branches; strike with a switch; chop up branches’ is indeed a formation from žābas ‘branch; handle; switch’. I would also point out that the figurative meaning of žabóti ‘tame, control’, žabóklai ‘2. prohibition, obstacle’ need not necessarily derive from ‘putting a bit on a horse’, but may equally come from žābyti ‘to strike, hit’, ²žabúoti ‘to strike with a switch’, that is, from the use of a switch as a herding tool, with which animals are driven and, when necessary, struck. If there is such a derivation from žābas ‘branch; handle; switch...’, it should be explained in this way. 8 | B E R N D G L I W A 2. If we follow the idea that the runners of the sled were made from an excavated tree together with one of its main roots, we see that the curve in the runner is exactly where the root bent away from the trunk. It is therefore quite interesting to consider whether Lithuanian kamienas ‘thick end of the trunk’ refers precisely to this point on the trunk, since the emerging roots cause the trunk to thicken, disrupting there the otherwise approximately conical shape of the trunk and producing a concave curvature of the trunk surface in the direction of growth. Specifically, could it be a formation based on a Baltic verb *kam- ‘to bend’? Simas Karaliūnas has demonstrated the semantic proximity of ‘to bend, turn’ and ‘thick end of a trunk, trunk’ with these pairs: žuolis ‘thick tree, trunk, railway sleeper’: žvilti ‘to incline, bend’, kūkė ‘piece of wood, log, piece of trunk...’: kūkti ‘to bend, bend down’, giūoga, guōgas ‘club, cudgel with a thick end’: giogti ‘to incline’, Latvian gudžēt ‘to huddle together and sit’ (Karaliūnas 1987: 12–13). Evidence that Lithuanian kamienas originally meant not the whole trunk but only its thick end is also provided by entries in the 17th-century manuscript Lexicon Lithuanicum: kamiēna – Stamend des Baums and liemū – Stam eines Baums (Drotvinas 1987: 357), although the editor notes in the index: ‘kamiena, medžio liemuo: kamiēna’ (Drotvinas 1987: 509). Since he also writes, ‘Kai abejojama dėl atstatymo tikrumo, rašomas klaustukas, o tokių žodžių reikšmė aiškinama pagal vokiečių kalbos žodžių reikšmę’ (Drotvinas 1987: 481), we can see that the meaning here was not doubted—prematurely. Further examples are kamėna ‘end of the trunk; thick end of a felled tree’ (Nesselmann, LKŽv 175), kamblys ‘trunk, handle; thick end of the trunk’ (LKŽv 174). A reviewer of an earlier version of this article had objected that Latvian kamanas ‘sled’ should first be compared with Latvian kamans ‘thick end of a beam’, as well as with Czech kmen, Polish kien, etc. Fraenkel adds Russian kom ‘lump’ to this traditional view (1962: 212). This cannot be accepted as it stands: first, sled runners (usually pluralia tantum) are curved, not thickened; second, the thick end of a beam is identical to the thick end of a felled trunk, since sawn timber has not existed for very long, and beams would not have a thick end anyway—what was said about kamiēnas therefore applies here. While, for example, kāmpa ‘growth on wood, burl...’ (LKŽv: 179) may still be regarded as tuber-shaped, such an interpretation for a trunk or stalk is far-fetched. Since these are not lump-shaped and have not been bunched together, one would have to assume a semantic transfer, perhaps beginning with ‘block’ or ‘stump’, though these are not particularly lump-like either. My counterproposal: the meaning begins with Baltic *kam- ‘to bend’ and refers to the point on a trunk or stalk where it merges into the root, or where branches fork off. Generalized, pars pro toto, this can be used to name the whole trunk or stalk. That leaves the block and stump. A stump is obtained when the trunk is felled, precisely at the point where it has a concave curve. Thus they have Lit. kāmanos, kamienas, kumėlė | 9 thus both, the trunk and the stump, are parts at the bend, and both have the right to bear this name. That the term later becomes lexicalized and restricted to either the trunk or the stump is to be expected and does not contradict this. As for the block of wood, a few practical remarks may be in order here: a tree is felled for timber or firewood. Crooked, branchy sections and forks are unsuitable as timber. Branches are also difficult to chop and thus, like root balls, unsuitable for firewood (except in an open campfire or for making pitch). Yet such branchy sections are exceptionally suitable as chopping blocks, butcher’s blocks, etc., precisely because they cannot be split and the wood is particularly dense and durable there. Here, too, the bend of diverging branches or roots may therefore be the basis, and the aforementioned objection as such is unfounded—or even entirely consistent with what is presented here—insofar as all the terms can be traced to the Indo-European root *kom- ‘to bend’. Quite apart from that, the comparison of stone with Czech stem ‘stalk, trunk’, Polish stump ‘log, stump, pole’ is disputed. ESSJ reconstructs Slavic *kmenъ, from which Czech kmen ‘handle, sex, kin’⁴ derives, < and Indo-Germanic *kn-men with the suffix -men- (ESSJₓ₃ₙ 196), and a related form with formant -n- *kъnъ/kъnь (ESSJₓ₃ₙ 196, 205), to which Polish stump ‘log, stump, pole’ is assigned (ESSJₓ₃ₙ 205), and which in turn is compared with Lithuanian body ‘body’, both semantically ‘body’ – ‘log’ and formally with *to swell (ESSJₓ₃ₙ 205), possibly also with Old Russian кинятися, кынятися ‘to gather in one place’, киняхуся ‘to buzz, swarm (especially of bees)’ (ESSJₓ₃ₙ 264). According to ESSJ, Lithuanian stone probably belongs instead with Slavic *lump (ESSJₓ₃ₙ 196) > and Polish dialectal lump ‘clump, ball’ etc. (ESSJₓ 179f.). Fraenkel disputes the connection between body and Polish stump on account of the Lithuanian ū, and assigns stump to bridle, which in turn is compared with ball, to roll etc. and Slavic *kъmy (gen. *kъmene, n-stem), Russian lump ‘clump’ (1962: 214). The semantic development presented is convincing in neither case. The idea that kamiènas, kamièna ‘trunk, thick end of the trunk’ was originally adjectival, with the ending -iena- < *-eina- (Skardžius 1996: 287f.), and thus represents a semantically attributive noun (Ambrazas 2000: 152), does not contradict the view put forward; semantically, however, a connection with Baltic *kam- ‘to bend’ is preferable to one with *kam- ‘to form a lump’. Kamiènas is thus the part of the trunk that has a bend. 3. Kāmanos, kamanà, kāmana denotes ‘leather bridle with a bit’ (LKŽᵥ 167–169), although the separate attestation [arklys] nuog kamanų ir žąslių tuoj nurimsta (LKŽᵥ 169) qualifies this statement, since if kāmanos unequivocally included the bit, such a statement would be unnecessary⁵. The wording common in songs, iš šilkų kasų kamanėlės (e.g. ø LKŽᵥ 169), is probably to be understood figuratively. 4 The terms used in the context of kinship are secondary transfers of vegetative (pre-)images, as, for example, in Modern High German Stamm, Stammbaum, seine Wurzeln irgendwo haben, Spross, Geschlecht. 5 One does not, after all, say *I’m driving a car and four wheels, since the car necessarily includes the wheels. 10 | Bernd Gliwa and refers to žirgas ‘steed, riding horse’, insofar as it symbolizes the young man (e.g. Sruoga 1957: 307–312); nevertheless, the term kāmanos also subsumes the reins: kamanužės rankužėlei jojau (Juška 1955: II 160ff., no. 564). A bridle without a bit is called, for example, apinast̃ris, as one learns from the statement kamanos su žabokliais, o apinastris be žaboklių – jis tik art (LKŽₓₓ 19), or also apīnasris (LKŽ₁ 216), apýgalvis (LKŽ₁ 205, 1063); it is noted that, for a good horse—good here meaning well-trained—this also replaces the kāmanos. On the other hand: gero arklio ir už apinastrio nenuturėsi (LKŽ₁ 216), where geras arklys probably means ‘strong horse’ here. A bit can be attached to this strap system, and at precisely the point where the bit is connected to the straps and the reins also take hold, a cheekpiece, Lithuanian laužtukai, can be attached on each side (for archaeological material and a reconstruction illustrating this, see, for example: Volkaitė-Kulikauskienė 1999: 316–327, Kulakov 2002: 215f.). When the reins are pulled, it exerts additional pressure on the horse’s cheek, thereby forcing it to change direction. Such a breaking bridle is called either plėstinės or laužtinės: plėbtinnes – Brechzaum (LEX 19a, in: Drotvinas 1987: 102), kamanos plėbtinės / zyfly / ydilay – monßuk końfki (SD₃ 179, in: Lyberis 1979: 277), laužtinės, laužtiniai, laužtinis, laužtuvės ‘bridle with cheekpieces’ Stov žirgeli... laužtinėliais pažaboti (song text, KlvD 236, cited in LKŽᵥᵢᵢ 207), laužtukai ‘cheekpiece on a bit’ (LKŽᵥᵢᵢ 207f.). In both cases the etymology is straightforward and points to the use of force: láužti ‘1. to break; 6. to change direction; 10. to pierce, hurt; 11. to force...’ (LKŽᵥᵢᵢ 201f.), plėšti ‘1. to tear off; 2. to pull forcefully to the side...’ (LKŽₓ 179ff.). The development of Lithuanian brizgilai, brizgilas, brizgilis, brizgilos ‘bit; bridle with a bit; ornament on a bridle; fittings on a sled; mouth’ (LKŽ₁ 1063f.), Old Prussian brisgelan ‘bridle’, perhaps related to bružduklis ‘bridle (straps only)...’ (LKŽ₁ 1099) (Fraenkel 1962: 60f., Mažiulis 1988: 157f., 161f. and literature cited therein), is comparatively unclear. Semantically, one might posit either an original meaning ‘to dangle, rattle’ (Toporov 1975: 252f.), cf. brazgiliai ‘bells on a horse’s neck strap’ and the sound produced by contact between the teeth and the bit, or ‘something detached and suitable for use as a bit’ (Mažiulis 1988: 158). A possible connection with brazlai ‘iron fittings on bridle and harness’, brazgėti ‘to crack’ and the like is unclear, as is one with branktas ‘swingle-tree; draught pole; (generally, a log used for a specific purpose)’ (LKŽ₁ 1002f.), which displays a striking number of meanings in common with bružduklis. After this brief survey of the parts and names of the bridle, one can return to kāmanos with some practical knowledge. V. Mažiulis reconstructs East Baltic *kamana- ‘something that has certain lumps, unevennesses’ < from Proto-Baltic-Slavic *kama- ‘forming a lump, ball’ with the suffix *-na (Mažiulis 1993: 102). It is unclear, however, where the lump-shaped feature is supposed to be found. Mažiulis explains, following Būga’s idea cited at the outset: „*‘a bridle-bit resembling in some way a (small) primitive sled runner’ = *‘wooden bit’“ (Mažiulis 1993: 102). Fraenkel, too, seems to regard the bit as a lump- or rod-shaped object motivating the name, and compares Latvian kamans ‘thick end of a beam’, kamanas ‘sled’, Czech kmen ‘stalk, trunk’, Polish kien ‘block, stump,’ Lit. kāmanos, kamienas, kumėlė 11 Stange’, russ. kom ‘Klumpen’, lett. kams ‘Klumpen, große Masse, Kloß’ (1962: 212). Wie man den Ausführungen weiter oben entnehmen kann, ist ein Zaumzeug ohne Gebiss sehr wohl möglich, nicht aber ein Gebiss ohne Zaumzeug. Mit einiger Sicherheit ist auch die technische Entwicklung in dieser Reihenfolge vonstatten gegangen und es ist sinnvoll, anzunehmen, dass es eine Bezeichnung des Zaumzeugs ohne Gebiss gegeben hat. Ob diese Bezeichnung erhalten ist, ist natürlich überhaupt nicht sicher. Wenn sie erhalten ist, dann kommen kāmanos und brizgilai in die engere Wahl, evtl. auch apinasris. Es ging bei dem Zaumzeug zunächst nicht um die konkrete Form6, sondern zunächst um eine Funktion, nämlich das Pferd zu zähmen und zu beherrschen. Diese funktionelle Sicht spiegelt sich etwa in arklys < *arklja ‘Pferd, das einen Hakenpflug zieht’ dar (Skardžius 1996: 62; Mažiulis 1993: 103). Das Zaumzeug kāmanos ist nun das wichtigste Mittel, um das Pferd zu zähmen und zur Arbeit verwenden zu können. Für die Arbeit, die erst nach dem Zähmen einsetzen kann, kommen dann noch weitere Geschirre hinzu, etwa Kummet mit Gurten, um Pflug oder Wagen anzuspannen. Mažiulis’ view: noun bridle < adjective uneven *‘having certain irregularities, clumps’ < noun Baltic-Slavic *clump- ‘lump, agglomeration’ < verb Baltic-Slavic *press- ‘press together, bunch’ (1993: 102). I agree with the proposed *press- ‘press together, bunch’, but see no reason to rule out a direct deverbal formation. It is certainly correct that a considerable number of adjectives are formed with the suffix -ana-, for example dárganas ‘unpleasant, rainy weather’: dárga ‘the same’ (Skardžius 1996: 226ff. Otrębski 1965: 174). This does not, however, justify claiming exclusivity, since alongside these there are also genuinely deverbal formations with -ana: a sheaf ‘an armful of hay or the like’, remainder ‘remainder’, interest ‘harvest; interest’, hemp shive ‘shive (male hemp)’, eye discharge ‘eye pus’ (Skardžius 1996: 227ff.), witch ‘witch’ (Gliwa 2003: 1–8), gift ‘gift; giving’ (LKŽ II 616), as well as formations with -ena: fighting, fighting ‘beating’, scolding ‘scolding’, gift ‘gift; giving’ (LKŽ II 616, Ambrazas 1993: 59f., 94f.,). Such directly deverbal formations are absent, however, with -anas, just as ablauting -ena or -sena, with an additional *-s- which seems to derive from a (Proto-)Future marker (Mažiulis 1988: 152f., Ambrazas 1993: 61 and Lit.), in forming abstracts such as gait; procession ‘gait; procession’, generally has no counterparts in -enas or -senas. Related ablaut forms in the zero grade may perhaps include fragmentarily attested verbal forms (active half-participles) in -ina-: ‘hurrying’, ‘leading’, whose formation, for syntactic reasons, must clearly be pre-individual-language (cf. Gliwa 2003a). The matter is presented somewhat differently from Mažiulis by a reviewer of this article, according to whom it was a wooden bit that was inserted into the horse’s mouth. 6 If one takes the process of its manufacture into account, one should return to meanings such as ‘bind, knot, wind around’. 7 It would also be interesting to ask how the desacralization of the horse, once important in religious cult (cf. Puhvel 2001: 293; Gimbutienė 1996: 263), is compatible with domestication. It was precisely for this reason that a new name was also needed, alongside the inherited ašva- ‘horse’. 12 | BERND GLIWA horse’s mouth, since the issue here is no longer primarily the shape but the material—that is, a derivation from an unattested Baltic *kama- ‘trunk, handle’ > kāmanos *‘what is wooden’ > ‘bridle with a wooden bit’. First, the overwhelmingly predominant use as a plural tantum does not suggest that the word denoted a single 8 piece of wood that would have been inserted into the horse’s mouth as a bit. Moreover, it is a drastic claim that the bit was made of wood. Bridles found by archaeologists have either metal bits (e.g. Volkaitė-Kulikauskienė 1999: 316–327, Kulakov 2002: 215f.) or ones made of horn or bone (finds dating to around 4500 BC, Gimbutienė 1996: 259; first millennium BC, Luchtanas 1992: 70), not to mention ethnographic evidence. I know of no reference to wooden mouthpieces in the archaeological literature. This proves little, however, since the objection that wood would probably not have survived is justified here. Furthermore, the earliest horse burials found in Lithuanian territory, the main source of bridle finds, date only to around the 2nd–4th centuries AD (Volkauskaitė-Kulikauskienė 1999: 310, Varnas 1998: 291–294). So let us approach the question from the practical side: could there have been wooden bits? Is it possible to make and use them? It is best to put this question to practically minded horse owners. They are very sceptical about whether oak could be used for this purpose, but immediately emphasize: ‘arklys gi sveikatos turi’ (S. Sigitavičius 2002, pers. comm.). The possible thickness of such a piece of wood would be limited by the fact that the horse must still be able to breathe, its tongue must not be immobilized, and its mouth should not be kept open. The mouthpiece could perhaps be removed for eating, as some horse owners do with the two-part iron mouthpiece (which allows freedom of movement for the tongue!), especially in the past, when horses were more heavily worked and more important than today. In extreme situations, one could therefore use typical one-use-and-discard wooden bits. Sigitavičius’s suggestion that people already knew and used metal when they bridled horses (2002, pers. comm.) is entirely consistent with what learned linguists (metal among the Indo-Europeans by around 4000 BC, Euler 2002: 47f.) and archaeologists (subject to the uncertainty involved in assigning finds to a language group) say on the matter, and deprives the discussion of almost any basis. For if one could make relatively durable metal mouthpieces, wooden pieces in this function would perhaps have been an inferior substitute and could hardly have motivated the naming. Bone is also far better suited, and was certainly available—J. Šeškauskas, however, attributes to bone a durability of only a few weeks and recalls that, as a young farmer (until 1950), he already used very thin, worn mouthpieces (2003, pers. comm.), which shows the limited service life even of iron mouthpieces. If one emphasizes the function of the bridle and considers the Baltic root *kam- ‘press together, bunch up’ as the basis of a possibly deverbal formation, 8 Carving wooden chain links is possible, but extremely laborious and they are very short-lived. Lit. kāmanos, kamienas, kumėlė | 13 one may infer a figurative meaning from the related9 kamantinė́ti ‘question closely, interrogate; pester with questions’ (LKŽV, 170) kamúoti ‘toil, torment; serve...’ (LKŽV, 191; likewise Fraenkel 1962: 214), a meaning close to the process of taming and controlling an animal, even though the concrete meaning ‘tame, domesticate’ < *‘accustom to the house’ (Euler 2002: 46; Pokorny 1994: 198–200) is absent. Lithuanian terms for taming, however, are motivated differently: on the one hand, Lithuanian (pri-)jaukìnti ‘accustom (something, someone)’, jaukėti, ²jaukti ‘(become) accustomed to something’: jaukas ‘bait’ (LKŽIV 300ff.), (pri-)jùnkinti, jùnkyti, jùnkti ‘accustom’: junklas ‘bait’ (LKŽIV 385ff.) < *‘accustom by bait, food’; on the other hand, they also refer to the other half of the “carrot and stick” recipe, namely trám(d)yti ‘comfort, soothe; tame, restrain, subdue; prohibit; destroy...’10 (LKŽXVI 556ff.): tremìti ‘knock down, strike, banish...’ (Fraenkel 1965: 1117); tvárdyti ‘calm, control, restrain’ (LKŽXVII 217): tvérti ‘grasp, seize; enclose...’ (Fraenkel 1965: 1152f.). It therefore seems possible to designate violent taming or control—and that is precisely what a bridle is needed for (cf. laužti, plėšti!)—with Baltic *kam- ‘press together, bunch up’, which involves applying force, especially pressure < Indo-European *kem- ‘press together, press, hinder’ (Pokorny 1994: 555f.; Rix et al. 2001: 329). As for word formation, primary Baltic formations in -ana seem predominantly to denote abstract nouns (nomina actionis), from which a semantic shift to concrete nouns (nomina acti) or to instrument nouns took place: dovanà ‘giving’ > ‘gift’, alongside dõvis ‘giving; ration; dowry’, dõtas, dotìlas ‘gift, dowry’ (LKŽII 615ff.). Ambrazas takes the same view of -ena: kãsena ‘digging’ > ‘tool for carving vessels’, mėtena ‘blow’ > ‘whip, switch; noose’ (Ambrazas 1993: 194). There are also primary concrete nouns such as liekana ‘remainder’ < *‘what was left behind’: lìkti ‘leave behind’, trãskana ‘eye discharge’ < *‘what was squeezed out’: trėkšti ‘splash, squeeze out’. The view that kāmanos was initially an abstract noun meaning *‘taming’, which subsequently shifted in meaning to *‘what is needed for taming, bridle’, is therefore semantically plausible and morphologically not an isolated case11. A brief excursion into phraseology and stable motifs in folklore provides further clues about kāmanos. A number of folktales contain the saying pard(u)odamas aržilą, neparduok su kamanom (LKŽV: 169). It concerns a sorcerer’s apprentice who, in order to earn money, turns into a stallion that his father is supposed to sell. He has to be sold without a bridle, since only then can he turn back into himself and escape (or, conversely, escape and turn back into himself). 9 Rix et al. reconstruct *kem- ‘press, squeeze’ and *kemh₂- ‘become tired, toil’, to which kamúoti could also belong, with a centum reflex? 10 The meaning ‘comfort, soothe’ is probably to be understood figuratively, influenced by raminti ‘calm, alleviate pain’ kur tramas, ten ramas (LKŽXVI 555) 11 This interpretation can also be applied quite elegantly, for example, to ragana *‘rising again; apparition, appearance’ > ‘what has risen again, what appears’ > ‘ghost’; I had not considered this before (on the etymology of ragana: Gliwa 2003: 1–8). 14 | BERND GLIWA (Basanavičius 1998: 208, No. 104). Corresponding versions exist with a rope and a boar or ox; sometimes the father forgets to remove the rope or bridle, and then the apprentice sorcerer cannot escape. However, it ends well, for someone eventually removes the bridle, allowing the transformation back to take place (Kerbelytė 1999: 158ff.). In anderen Märchentypen (Kerbelytė 1999: 196, 210, 313, 337) und auch in Sagen gibt es das Motiv, dass eine Hexe durch Überwerfen des Zaumzeugs einen Bursche in ein Pferd verwandelt, bzw. umgekehrt der Bursche die Hexe oder Hexentochter in eine Stute: jis greitai paėmė ir užmetė apynasrį. Ta pavirto į kumelę. (Vėlius 1979: 213), šeimininke kamanas ant jo galvos užmauna, paverčia jį arkliu, užsėda ir joja ant Šatrijos kalno (1979: 226; LKŽII 88). Since an analogous motif also occurs, for example, in Indian folktales, though here with oxen and a leading rope, where the transformation back can take place only when the rope tears out of the hole in the nasal septum (Hertel 1998: 138), this motif can hardly be a Baltic innovation. The point, then, is that a bridle can be used to exercise absolute power over human and animal alike (the concrete realization is subject to shifts in the props; cf. Lüthi 1996: 84). In considering what may today seem magical or merely folkloric, however, we must not forget that the domestication of livestock was religiously significant: after all, the deceased ancestors of the Indo-Europeans were represented, among other things, as cattle (cf. Vėlius 1987: 163ff.). The transition from an animal conceived as a representative of the ancestors to livestock (economization) goes hand in hand with desacralization. If one regards the animal as equal to a human, simply in a different form, then the bridle, as a means to an end, turns this human in animal form into livestock that can now be owned and controlled. Finally, there is a saying: Jei randi arklio kaukolę ir tą pažabok – taip sakydavo (J. Šeškauskas 2002, pers. comm.), which can make sense only in a religious context. The meaning ‘crooked’ as a starting point is also supported by the meaning of Latvian kamana ‘1. sleigh, 2. crooked woman’ (Mühlenbach 1927: 148), the latter of which can be understood both as a substantivized adjective and as a deverbal noun like ‘trāskana eye discharge’ < *‘what was squeezed out’: trēkšti ‘to spray, squeeze out’. 4. The origin of German Hamen ‘horse collar’, possibly identical to Hamen ‘net’, is considered unclear; a comparison with Lithuanian kāmanos, Greek αημός ‘muzzle, woven lid of a ballot urn, fish trap, mud bandage, etc.’ or hemmen has nevertheless been considered (Kluge 1999: 352). Kummet <, by contrast, is borrowed from Polish chomąto / Upper Sorbian chomot ‘same’, “perhaps corresponding to Hamen” (Kluge 1999: 493). According to Fraenkel, Lithuanian kamanītai ‘horse collar’ likewise comes from Polish chomąt(o) ‘same’. I would question this. On the one hand, kamanītai can be etymologized in the Baltic context; on the other hand, its meaning is hardly restricted to ‘horse collar’. The following meanings are given: ¹kamaňtas ‘horse collar, wooden bow on harness; harness, trace harness; breast strap on trace harness; lazybones, scoundrel’, ²kamantas ‘interrogating, harassing’, ³kamántas ‘piece, plot of land’ (LKŽV 169f.). In terms of word formation, provided they are not borrowed, compare ramāntas ‘support, crutch’. Lit. kāmanos, kamienas, kumėlė 15 (Ulvydas 1965: 393), gramanìtas ‘clod of earth’, lamanìtas ‘piece’; place names: Kūkantas ‘hill in the Skaudvilė area’, Šalantas ‘right tributary of the Minija’, among others (Skardžius 1996: 374). It is, however, a “very old formation not only for participles, but also [...] especially for nouns” (Skardžius 1996: 376). In a detailed study of hydronyms containing the element -nt-, Skardžius concludes that at least some of these names are formations confined to individual languages, which use, on the one hand, the inherited suffix -ant-, as in Kražāñtė ‘river through Kražiai’: Kražiai ‘town in Kelme district’ (Skardžius 1998: 894), and, on the other hand, insert -t- into endings of the type -ana-, -ena-, as in Kieméntas ‘lake in the Giedraičiai area’ (formerly also called Kiemeitis): kiemena ‘custom, characteristic of a farmstead’ (Skardžius 1998: 894). These two categories are denominal in nature (Skardžius 1998: 918–927). A third, deverbal, type of formation is comparatively rare: Bėgantỹs ‘tributary of the Dubysa’, Tėkančioji ‘tributary of the Nevėžis’; these are nearly identical to the active present participle, and in many cases the formation can no longer be determined (Skardžius 1998: 939). As regards common nouns, gramanìtas ‘clod, large piece’: grāmatas ‘heap, group’: grāmas ‘cud’ and lamanìtas ‘clod, piece’, lamātas ‘piece’, lamākas ‘piece, plot of land’, lāmas ‘same’ (Skardžius 1998: 922f.) seem to belong to the denominal formations, whereas ramanìtas ‘crutch, piece’, remēntas, ramenìtas ‘crutch’, alongside rāmstis ‘support, crutch, railing’: rēnti ‘support, hold, build...’, (LKŽₓᵢ 118ff., 428ff.), suggest a formation closer to the participle, with the basic meaning *‘that which supports’. Several possible explanations are thus available for kamantas. For ¹kamántas ‘piece, plot of land’, ²kāmas ‘clod, lump; porridge’ is a plausible base for a denominal formation. ²Kamantas ‘interrogating, harassing’ and kamantà ‘one who interrogates’ (LKŽᵥ 169) are related to kamantinėti ‘to interrogate, pester with questions’, kamantìnti ‘same; torment’; more precisely, they form the basis of the latter verbs and cannot be derived from them. I therefore take ²kamantas ‘interrogating, harassing’ and kamantà ‘one who interrogates’ as deverbal formations (~ active present participle), from a base *kam- ‘tame, subjugate to one’s will, torment’, whence also kamiuoti ‘torment, serve’, which in turn reflects a semantic shift from the Baltic form *kam- ‘press, squeeze; restrain’ already discussed (cf. Pokorny 1994: 555).¹² The meaning characteristic of an active present participle is particularly clearly preserved in kamantà ‘one who interrogates’. ¹kamanìtas ‘horse collar, wooden bow on harness; trace harness; breast strap on trace harness; lazybones’ can be analyzed in the same way: a meaning *‘that which tames’, concretely used as an instrumental noun, has an exact counterpart in ramanìtas *‘that which supports’, in terms of word formation. Alternatively, one could posit a development kamana ‘bridle’ (<* ‘that with which one tames’) > kamanìtas ‘harness, etc.’, in line with the developmental paths proposed by Skardžius. In this case it would presumably be a language-specific process, which may have been influenced by Polish chomąto and is therefore inseparable from borrowing. The first alternative certainly suggests an older formation, especially since, for Slavic *chomotъ/chomotъ, it has been proposed that -otъ is 12 Cf. Lithuanian slogutìs ‘nightmare, incubus’: slégti ‘press, squeeze’ and ‘torment’. 16 | Bernd Gliwa the old suffix of the present active participle is involved (cf. ESSJVIII 70), or whether the Slavic word may have been borrowed from Gothic *hamands (Pokorny 1994: 555, ESSJVIII 70 and literature cited therein). ESSJ points to the diversity of meanings; since terms for a bunch, bundle, and fishing net occur here alongside the word for ‘collar’, it can hardly be a loanword. A formation based on Indo-European *(s)kom- ‘press together, squeeze’ is proposed and compared with Lithuanian kāmanos and German Hamen (ESSJVIII 70). The derived term kamantinis occurs as a riddle word: Atlēkē šaltabarzdis, pasiëmē kikibarzdži. Pasiimčiau kamantinį, pasivyčiau šaltabarzdį, atsimėčiau kikibarzdį, in which wolf, sheep, and horse are encoded (LKŽV 170). One step further gives kamantikai, again in a riddle: Atbēgo vogìbasai, pagavo kikirmikį, nuvijo kamantininkai ‘wolf, sheep, (mounted) shepherds’ (LKŽV 170). Kamantinis is an adjectival formation with -inis on the base kamant-as13; such adjectives express properties, affiliation, origin, and so forth (Ulvy das 1965: 568ff.). Given the attested and reconstructed meanings of kamantas ‘harness; breast strap...’ *‘*that which tames’, an underlying meaning *‘bridled horse’ is plausible for kamantinis ‘horse’. Proponents of borrowing Lithuanian kamāntas ‘collar’ might object that a meaning *‘horse wearing a collar’ would have to be reconstructed (with the other meanings then being transferred or generalized?). Against this, besides the word’s range of meanings, it should be noted that a riding horse is not harnessed with a collar. One might still suppose that ‘horse’ is to be understood as a generalization from ‘draft horse’—but it is then puzzling that the word refers exclusively to a riding horse. Kamantinis ‘riding horse’ must also underlie the riddle word kamantinkai ‘mounted shepherds’. Mažiulis regards the development that led to Old Russian комонь ‘warhorse’ as formally similar: Slavic noun *kamana ‘wooden bit’ > adjective *kamanja- ‘fitted with a bit’ > noun *kamanja- ‘steed (horse requiring a bit to control it)’ > Old Russian комонь ‘warhorse’ (Mažiulis 1993: 102f.). Objections concern the semantic development. The proposal involving a wooden bit should be replaced by a more abstract notion, *‘what one uses to tame a horse’, or more concretely, *‘bridle’; moreover, it is highly unlikely that Old Russian комонь ‘warhorse’ developed from *‘horse requiring a bit to control it’, since a warhorse in particular must be trained well enough to respond to pressure from the rider’s legs or to commands, because in battle the rider generally has no hand free to handle the reins and their accessories14. Since, however, *kamanja- ‘steed (tamed and trained horse)’ can readily be posited on formal grounds, this presents no serious difficulty. From an analogous Prussian adjective *kamanja- 13 Analogous are ramanītis ‘crutch’ and ramantinis ‘thick club’ (LKŽXI 118), likewise substantivized. 14 This applies both to the medieval knight (armed with shield and sword or a two-handed battle sword) and to early Indo-European mounted warriors (bow and arrow). The latter are often regarded as warlike nomads who defeated their enemies and rapidly expanded thanks to their mounted warriors (Gimbutienė 1996: 256ff.; Renfrew 2000: 40–48, among others, argues against this). Lit. kāmanos, kamienas, kumėlė 17 may also have developed > noun kamanis > diminutive noun kamanītis > Old Prussian (GrC) kampnit ‘horse’, (GrG) camnet ‘ditto’ (Mažiulis 1993: 105). Toporov, however, analyzes Old Prussian camnet ‘horse’ as *kam(a)n-ēt-(a)s > camnet, which would correspond to the Lithuanian participle *kaman-ét-as; the fact that Lithuanian kamanėti ‘move, shift’ is attested only with a different meaning does not undermine the interpretation. Toporov also agrees with the comparison to Old Russian комонь ‘warhorse’ (1980: 194). 5. Since the zero grade Indo-European *kṃ- can lead not only to Baltic *kim- but also to *kum-, it is worth examining whether Lithuanian kumėlė ‘mare’ and kumelỹs ‘stallion, foal’ might also belong here. In relation to a Baltic verb *kam- ‘tame’, kumėlė would be a zero-grade suffixal formation. Ambrazas notes in this connection: “The derived suffix *-elo-/-olo- may be quite old. It was probably already used in Proto-Indo-European to form deverbal adjectives, which later became nouns and acquired meanings referring to actions, their results, or agents” (Ambrazas 1993: 71), but does not address whether kumėlė is such a formation. The view that kumėlė had the meaning *‘tamed mare’ (the result of the action) would be functionally motivated and finds a semasiological parallel in jáutis ‘ox, male bovine’ < *‘animal yoked to a yoke’ (Fraenkel 1962: 191). A comparison with New Persian dām ‘tame animal’, Old Indic dām-: dāmyati ‘be tame; tame’, damana ‘taming, subduing; horse tamer; taming; punishing’, damya ‘young bull to be tamed’, damin ‘tamed; taming; self-controlled’, Greek δαμάλης ‘subduing, taming; young bull’, ‘young cow’, δάμαλις ‘calf’, δαμάλη ‘young cow; girl’ (Pokorny 1994: 199, Kaegi 1990: 160, Mylius 2001: 201), ἀδαμάντινος ‘steel-like, firm, hard’ < ἀδάμας ‘unconquerable’ (Kaegi 1990: 9), δαμάζω ‘active and middle: tame, break in, train; (a girl) marry off; defeat, subdue; stupefy; weaken; kill in battle; passive: be tamed, be brought under control’ (Kaegi 1990: 160). Sabaliauskas (1968: 156 u. Lit.) erörtert, ob es sich bei kumėlė um eine primäre Bedeutung *‘kastriertes Pferd’ handeln könnte, ähnlich wie sl. *komolъ(jь) ‘hornlos; Stumpf, Stummel’ (ESSJₓ 174f.)¹⁵. Schließlich favorisiert er den Vergleich mit комонь, bedauert aber, dass „jų giminystė įrodyti kol kas neįmanoma” (Sabaliauskas 1968: 156). Kazlauskas criticizes Sabaliauskas’s study for failing to take into account, in the discussion, other meanings of kumelỹs and related words (1970: 132). First to be mentioned here are kumėlė ‘1. mare; 2. a curse word; 3. bridge on which the strings of a violin and other instruments rest¹⁶; 4. double bass; 5. eye in the millstone into which the handle is fitted¹⁷; 6. handle bar on a millstone; 7. defect in a rope winch or in the warp threads (in the form of a loop); 8. crosspiece on the 15 A separate onomasiological study would need to show that the very variety of Lithuanian terms for ‘without horns; stump, stub’ is based on a concept ‘to be crooked’, alongside the less common ‘to be smooth’, so that this comparison with the Slavic terms is not out of place here. 16 This part of the instruments is designated in many languages by terms corresponding to ‘horse’, ‘mare’, and ‘donkey’ (G. Kirdienė, 2002, pers. comm.). 17 The eye is also called šuniukas, ausis (LKŽₓᵥ 365). 18 | BERND GLIWA dragnet; 9. a tool for bending wheel rims; 10. rack; 11. an edible mushroom with a thick stem and thick cap (LKŽVI 867); further meanings add kumeliūkas ‘1. foal; 2. a game; 3. a long toothed lever for pressing down the fastening battens over the straw layer of a thatched roof18; 4. a bend in a mown strip; 5. egg yolk’ and kumelỹs: ‘1. horse, stallion; 6. a board for beating smooth a thatched roof; 9. mouth of a dragnet; 10. float on a net; 11. irregularities on a rope or whip; 12. bulges formed on a scythe when it is hammered19; 13. backbone’ (LKŽVI 868ff.) are added. These meanings should be sorted out somewhat. What follows should be understood as an attempt to do so. Its use as a swearword, especially for designating stocky women, coincides with the use of German Stute ‘mare’ or Kuh ‘cow’; the double bass also belongs here, on account of its shape. The violin bridge and the rack are semasiologically related to the gymnast’s Pferd ‘horse’ or the Sägebock ‘sawhorse’. Up to this point, then, these are all meaning transfers originating in the established concept of the horse. The roofing and rim-bending tools, however, appear to be independent formations, which I would relate to a zero-grade base of Baltic *kum- ‘press, squeeze’; kumeliūkas ‘egg yolk’, which can plausibly be compared with kãmas ‘lump, ball; mush’ (LKŽV, 172), and the thick mushroom probably belong here as well. I relate kumelỹs ‘irregularities on a rope; bulges on a scythe; backbone20’ and kumėlė ‘eyelet in or on a millstone where the handle is fitted; handle bar on a millstone; fault in a rope winch’ to a base *kum- ‘bend, turn’. For lack of subject knowledge, I will refrain from assigning the fishing terms; I am also unfamiliar with the course of the game. There are thus several identical, homophonous but largely independent formations, which underscore the aptness of such a formation: it produces both deverbal concrete nouns (the result of an action) and designations of tools (instrumental nouns). Fraenkel relates kumėlė “most probably to Sanskrit kumārā- ‘(newborn) child, boy, youth, son’” (Fraenkel 1962: 309), which in turn is derived from < *ku-mel-, with the prefix *ku-21 added to the root *-mel- ‘soft, tender, young’ (ESSJX 178). This etymology has at least one clear shortcoming, namely the assumption of a prefix *ku- otherwise unattested in Baltic. As an argument that kumėlė cannot be a suffixal derivation from *kum-, it is pointed out that kumė was merely a back-formation22 (Fraenkel 1962: 309) and has nothing in common with Russian комонь. 18 Besides kumeliõkas, kumelỹs, this tool is also called erzīlas ‘stallion’, probably by a semantic transfer (LKŽII 1152). 19 It is also called paplampa, klumbė (LKŽVI 869). 20 Semasiologically, compare Lithuanian kãklas, probably a reduplicated formation from Indo-European *kʷel- ‘turn’ (Fraenkel 1962: 205; Kluge 1999: 351). 21 Sanskrit ku-, however, normally conveys a negative meaning: kūmanas ‘out of tune’, kumitra- ‘bad, false friend’: mitrá- ‘friend...’ (Mylius 2001 112, 114, 370). 22 The assumption of a back-formation is very questionable. At first glance, it does seem possible that kūmė was shortened to satisfy rhythmic (metrical) requirements, since in the preceding stanzas one finds disyllabic terms here: Aš turū gãidi, reiks ma vīstos...Aš turū qnti, reiks ma Lit. kāmanos, kamiènas, kùmėlė 19 has (ESSJx 178). Even if the attested kumė is a back-formation, this says very little and by no means rules out an original *kum-el-. Just as one finds no ežilas, eržalas, erželas, eržilis ‘stallion’ (LKŽII 1151f.) *eržas/eržis with the same meaning without having to doubt that it was formed with an -l- suffix, the existence of Lithuanian kumė is in no way a necessary precondition for an analogous formation23 of kumėlė with an -l- suffix. If, then, one accepts the idea that kumėlė originally meant ‘tamed mare’, then the primary meaning of kumelys would be ‘foal’, a relational formation denoting the offspring of a mare. M. Gimbutienė argues that the original motives for horse breeding on the steppe were the use of milk and meat; only later did riding and work become important, more important than the former uses (Gimbutienė 1996: 259). She also points out that the predominance of stallion bones among the finds already indicates a later stage of breeding, whereas in its initial phase mare bones are chiefly to be expected (1996: 259). In the Baltic region, too, the fact that almost all the horse bones found had been chopped up indicates that horses were still being raised primarily for meat in the first millennium BC (Luchtanas 1992: 70). The importance of mare’s milk among the Balts is attested by Wulfstan (c. 890 AD): kings and nobles drink mare’s milk, while the poor and slaves drink mead (Vėlius 1996: 168). 6. Like the riding horse kamantinis, a lake is also called Kamantinis (near Rimšė, Ignalina district). The preferred interpretation derives the name from a personal name24, but a comparison with kamàntas ‘plot of land’ is also considered (Vanagas 1981: 144; Skardžius 1998: 893f.). Skardžius notes that there may be a connection with other hydronyms containing the root kam-, possibly with a base kãmas ‘lump, clod, ball’ etc. (1998: 893). Other bodies of water include Kamaìnė (river), Kãmė (river), Kamojà (river), Kamanỹs (lake), Kamatis (river), Kãmona (river), among others (Vanagas 1981: 144). These are understood, on the one hand, as derivatives of terms for the horse (Otrębski 1961: 262f., cited in Vanagas 1981: 144), and, on the other, as reflecting a conjectured and plausible meaning of kamas, namely *‘moss mound or clump of grass’; compare kėmsa ‘moss mound, moss-covered stump; tuft of grass’, kìminas ‘peat moss, Sphagnum spp.’, kìmonai ‘the same’ (LKŽV 802f.; Vanagas 1981: 144). It should be objected, however, that such moss hummocks form in standing water, especially in marshy areas and raised bogs, but not in or beside flowing water. Clumps of grass are also hardly found along kùru...Aš turiu žirgą, rejks ma kùmės.... But since the pattern is broken shortly afterward... Aš turiù žãmbi, rejks ma akėčios... Aš turiù piėmeni, rejks ma gùlovos... (Juška 1955: I 300ff., no. 150.), this can hardly be assumed to be the cause. Moreover, it is quite common in genuine Lithuanian folk songs for each stanza to have a somewhat different melody, so that the number of syllables is easily adjusted in this way (D. Šeškauskaitė, 2003, pers. comm.). 23 Juška is also said to have created a back-formation erėlis ‘eagle’ ãras, ėras (Fraenkel 1962: 122). According to Skardžius, this does not originate with Juška, since the form occurs in songs (1996: 21). Since poetic German Aar (a relic) is also found, from which Adler < derives from Middle High German adel-are (Kluge 1999: 15), there is no reason to deny that a corresponding relic exists in the song texts’ ãras, ėras and to regard it as a back-formation. 24 This is not preserved, but its existence can be reconstructed from the personal name Kamantauškas (Vanagas 1981: 144). 20 | BERND GLIWA water, but rather on meadows. It is certainly reasonable to derive helonyms such as Kemsinaite (pool) < Kemsinė (pasture) (Bilkis 1999: 57), Kemsėlis (pool) < kėmsas ‘moss mound’25 (Bilkis 1999: 65) in this way; for flowing water, however, this interpretation is questionable, especially when these are primary names such as Kãmė, which can hardly be traced to adjacent pastures, as can, for example, Kemsinaite (pool). I would therefore point to Kumė (river), Kumpė (lake), Kumpalė (river), among others (Vanagas 1981: 171), which Vanagas connects with Lithuanian kũmpti ‘to bend, become crooked’ and Latvian kùmt ‘to become hunchbacked, crooked’ (1981: 171). Semantic reference to the meanders of rivers is extremely common26, as in Vingis (river): vìngis ‘arc, bend’ (Vanagas 1981: 385), Kulė (river): kulỹs ‘bay, promontory, bend, curve; corner, edge, piece’ (Vanagas 1981: 170), Ringė (river): ringė ‘curve’ (Vanagas 1981: 278), Lùnginas (river): lungúoti ‘to sway, meander...’ (Vanagas 1981: 171), among others (see also Vanagas 1988: 8). The fact that Kamantinis is a lake may perhaps be explained as the result of transferring the corresponding name from an inflow or outflow to the lake; however, I have no data supporting such an interpretation in this particular case. Namen wie Kumėlupis (Vanagas 1981: 171) könnten sich durchaus auf Stuten beziehen, müssen sie aber nicht. Einerseits war auf die Bedeutungsvielfalt von kumėlė und kumelys verwiesen worden, der teilweise ein *kum- ‘drehen, krümmen’ zugrunde zu liegen scheint, andererseits hat man eine der Wortbildung nach exakte Entsprechung in Ving-el-upis (F) (Vanagas 1981: 385). Auch hier wäre also ein Bezug auf die Mäander des Flusses als Benennungsmotivation denkbar mit einem erst sekundären Anschluss an kumėlė. Mit einem -r-haltigen Suffix27 könnten zu dieser Wurzel auch Kamarià, Kamãrė (F), Kamarà (F) u.a. (Vanagas 1981: 144) gehören, womit die wenig überzeugende Anlehnung an das Lehnwort kamarà ‘Kammer...’ hinfällig wäre. Dass Kamárkos upėlis (F) auf Kamárka (Wiese) zurückgeht (Vanagas 1981: 145) ist denkbar, mindestens genauso wahrscheinlich ist der umgekehrte Gang: ein Appellativum für Wiese lañkas nannte ursprünglich die Wiese, die sich in einer Flussbiegung befindet (Fraenkel 1962: 339), ist also ursprünglich nach dem sich 25 LKŽ, und darauf verweisend auch Bilkis (1999: 65), geben für kėmsas ‘samanomis ar žole apžėlęs kelmas ar žemės kauburėlis, kašas’ (LKŽ, 563) an, und gehen damit am Kern der Sache vorbei. Die Hügelform resultiert nämlich nicht aus einem unterliegenden Erdhügel oder Stubben, sondern aus der Art des Mooses (Zweige oder Sträucher sind vollkommen ausreichend als Skelett) oder der Grasart (hier kann von einer Unterlage überhaupt keine Rede sein) zu wachsen (seltene Ausnahmen sind möglich, die würde man aber in der Regel auch kelmas ‘Stubben’, kauburys ‘Huckel, Haufen’, kupstas ‘Haufen’: kaupì ‘anhäufen’ bezeichnen). Die Moose, um die es sich handelt sind wesentlich kimìnai ‘Torfmoose, Sphagnum spp.’ und daher ist die Etymologie auch mit diesen zu verknüpfen: die Mooshügel sind nämlich ausgesprochen weich und geben nach, lassen sich zusammenpressen, gehören also zu kimšti ‘stopfen’, evtl. auch als Isoliermaterial im Bau und Verwendung in Kopfkissen etc. Weich, im Vergleich zu einem Erdhügel, sind auch die Grasbüschel (als Hindernis bei der Mahd eventuell ‘verstopfen’ sie vielleicht die Sense – ebenfalls eine zu prüfende Motivation). Daher ist die Etymologie von Fraenkel (1962: 239) „wohl zu der [unter] kemėras ‘unförmiger, großer Gegenstand, ungeschlachte Person’ [...] zusammengestellten Familie“ nur auf die Form orientiert und daher zurückzuweisen, wenngleich weitläufige etymologische Verwandtschaft besteht. 26 In Lithuania, however, after land improvement, this can often no longer be reconstructed from observation. 27 Semantically equivalent formations with -r- and -l- suffixes occur, for example, in kėmelis ‘difficulty, inconvenience’, kėmeris ‘same; something tangled, a misshapen thing’ (LKŽ, 557). Lit. kāmanos, kamienas, kumėlė | 21 named after the winding river. Alongside corresponding river names Lankà (F), Lañkenas (F), Lankupė (F) has also been proposed (Vanagas 1981: 180), presumably to distinguish it from the meadow once named after the river. Likewise, Kamárkos upėlis (F) may have arisen alongside Kamárka (meadow). If one sets aside the loanword kamarā as the source of the name, one can readily assume Kamarā to be the older, original form, rather than having to regard it as a “late artificial variant” (Vanagas 1981: 145). In this connection, I would also like to point to a riddle: Kur tu bėgi, kambanka? – Kas tau darbo, praskustine? (LKŽᵥ, 173), which encodes a river and a mown meadow, as do the variants: Kur tu bėgi, kumpuringe? – Kas tau darbo, kirpse!; Kur tu bėgi patekolė...; Kinka ringa, kur tu bėgi...; Kreivarangi, kur tu bėgi... (Korsakas 1968: 461). It is obvious that, apart from patekolė: tekėti ‘to flow...’, each refers to the bends in the river. With regard to Kinka ringa, one should also note kinkai ‘movable connections; joints’, kiñka ‘leg; door hinge; calf; knee; thigh...’ (LKŽᵥ, 813). The suffix -anka is considered Slavic and appears, for example, in Liaukiánka (farmstead), Mazuránka (village), Sabaliánka (farmstead) (Razmukaitė 1998: 164, 197). 7. Another kamantinis is likewise found in a riddle: Vytinis kamantinis, aukštyn lėkdamas, kiaušius deda or: Susuktinis pakunktinis lipa aukštyn kiaušinių dėti (LKŽᵥ: 170, Korsakas 1968: 491f.). The answer is hop (Humulus lupulus L.), specifically the female plant, whose cone-like fruiting clusters are compared here to eggs. While the usual names apynys, apynis and slightly differing variants clearly refer to its twining growth as a climbing plant, on the basis of what has been said so far there are two possible explanations for kamantinis. On the one hand, it could be compared with river names kam-ant-inis *‘that which winds, meanders’, which would be a semantic equivalent of apynys. On the other hand, the derivation proposed for kamantinis ‘riding horse’ is also possible, but here not with the transferred meaning ‘to tame’; rather, it would derive from the more original *kam- ‘to form into a lump’, to which kãmas ‘lump, ball’ belongs, with kam-ant-inis meaning *‘that which is equipped with cones; that which forms cones’. This second possibility has a parallel in German Hopfen ‘Humulus lupulus L.’, which originally named only the cones, an important raw material for medicine and food (Marzell 2000: II 904). The botanical name humulus is considered a loanword from a Germanic language (Marzell 2000: II 902–903); the plant seems to have been unknown in antiquity. Written sources mention hops in Central Europe in the 8th century, but then immediately in reference to humlonariae ‘hop gardens’ (Marzell 2000: II 902). Hehn mentions its use in the Grand Duchy of Lithuania in the 15th century: at the wedding of Grand Duke Alexander, his bride, the Russian princess Elena, was showered with hops (Hehn 1963: 485). From this perspective, one might be inclined to agree with ESSJ’s account, according to which hops came to Europe from the east and could therefore certainly have been designated by a Wanderwort (discussed in detail in ESSJᵥᵢᵢᵢ 141–145). Pollen analyses, however, show that in the Žemaitian uplands, near Lake Biržulis, hops grew abundantly over an extended period (at least 1,500 years) around 7000 BC; during this warmer climatic interval, even beeches briefly 22 | B E R N D G L I W A nachgewiesen (Guobytė, Stančikaitė 1998: 125, Fig. 4). In signifikantem Maße erscheint der Hopfen wieder (gemäß Pollenanalyse) in den ersten Jh. AD (Guobytė, Stančikaitė 1998: 124, Fig. 3). Daraus darf nun aber nicht geschlossen werden, dass der Hopfen in der Zwischenzeit gar nicht vorhanden war. Sicher ist jedoch, dass sich die artspezifische Vegetationsgrenze verschob, möglicherweise zwischenzeitlich bis südlich des heutigen Litauens. Angesichts des vormaligen ausgedehnteren Siedlungsraumes der Balten (sobald man in dem betrachteten Zeitraum von Balten reden kann), kann der Hopfen in diesem aber mit hoher Sicherheit als vorhanden angesehen werden. Šimkūnaitė berichtet, dass Hopfenreste in Ausgrabungen steinzeitlicher Siedlungen gefunden werden (2001: 110)28. The ritual use of hops, or even their comparison with Soma or Haoma (ESSJVIII 144), cannot be maintained in light of pharmacological data either. Hops have a calming effect and are used as a sleep aid. An anaphrodisiac effect is also known—and it was precisely for this reason that hops were first added to beer in monastic communities: to ward off the devilish urges. The purity law of the 16th century made this practice law, establishing hops as a beer seasoning ø and banishing, for example, henbane from beer (Rätsch 1998: 269–271). Marzell regards Germanic *humilo as an inherited word, on account of Old English hymele. It is also noted that, besides hops, field bindweed ‘Volvulus spp.’ and white bryony ‘Bryonia spp.’ are also designated by Old English hymele. He therefore concludes that Slavic *chъmelъ is borrowed (Marzell 2000: II 903). Seebold remarks laconically that the direction of borrowing is disputed (Kluge 1999: 382f.). If one takes into account that field bindweed and white bryony are also climbing plants, it is reasonable to suspect an associated concept behind these terms, something unlikely to arise through borrowing. Germanic *hum- < Indo-European *ḱm̥- can itself be regarded as a zero-grade form of Indo-European *kem-. Correspondences involving Baltic k-, Slavic ch-, and Germanic h- are usually explained by borrowing (e.g. Büga 1959: 216). ESSJ takes a different view: if a mobile s is assumed for the Slavic cognates, then ch- is expected regularly here (ESSJVIII: 70 et passim). If Baltic k-, Slavic ch-, and Germanic h- (in some cases) can be explained by common inheritance, it is then significant that Slavic -ъm-, too, may be the result of an Indo-European zero-grade form *-m-. In Lithuanian, this grade would point to kum- or kim-. Lithuanian kamantinis, however, continues the o-grade, and could therefore at most be a parallel formation29. As long as a concept remains etymologically transparent, lexicalization can certainly take place on the basis of different ablaut grades and need not necessarily indicate borrowing, e.g. kamane ‘bumblebee, Bombus spp.’ < Indo-European *kom-: German Hummel, Czech čmel ‘bumblebee’ < Indo-European ḱm̥- (although the multitude of etymologies 28 This is a popular-science article without references, but since the author is known to have often undertaken the analysis of plant remains from archaeological excavations, I see no reason to question this statement. 29 Relevant to further investigation should be German dialectal (Transylvania) Hamê ‘Humulus’, which according to Marzell comes from Romanian hmei, hamei ‘same’ (Marzell 2000: II 906). However, it is hardly a Romance inherited word, and borrowing from Slavic also seems unlikely. Lit. kāmanos, kamiēnas, kumelē 23 could be interpreted in such a way that there is no relationship here; cf. ESSJV 145 and Lit.—a view I nevertheless doubt). A further example from the same semantic field: ‘fly’ in Lithuanian is musė (zero grade *-u-), whereas Slavic uses the o-grade (*-ou-): Polish mucha. Indeed, even within the dialects of a language, different ablaut grades can be used to designate identical situations: Lithuanian kaukas ‘goblin, house spirit’: kukas ‘same’, and likewise Old Prussian cawx ‘devil’, cucumbrast ‘Devil’s Ford’ (Maziulis 1993: 148ff., 295f.; for further examples, cf. Kazlauskas 1968: 109). Put somewhat differently: there may initially be several competing designations, such as dialect variants, for similar or identical situations, whose subsequent development may proceed in diametrically opposite directions. A particularly striking example is the Avestan and Old Indic designations for gods and demons: Old Indic devá- ‘god; king…’, ásura- ‘demon’ but also ‘living, divine’; Avestan daēva- ‘demons’, ahura- ‘god’ (cf. Puhvel 2001: 99; Mylius 2001: 58). Thus the affiliation of kamantinis can certainly be justified. On the other hand, the corresponding zero grade may occur in the phraseological expression kumelės šlapimas ‘boiled malt or warm beer with honey and milk’ (Šimkūnaitė 2001: 29). Since beer is nowadays usually hopped, it is conceivable that the name derives from hops. And kumelė corresponds to Germanic *humilo and Slavic *chomelb not only in ablaut grade but also in suffix. As Šimkūnaitė has shown elsewhere, names meaning šlapimas ‘urine, moisture’ or šūdas ‘dung’ are to be understood metaphorically; the named thing—in this case, the sclerotia—does indeed resemble mouse droppings, as in pelėšūdžiai ‘literally: mouse droppings’ = ‘ergot, Secale cornutum’ (2001: 72). Although in the case of kumelės šlapimas the name could be justified by a certain resemblance, the primary motivation remains the use of an archaic word, as is quite common in riddles—and the recipe is a riddle, a trade secret, after all. 8. Another case in which Slavic ch- has a Baltic k- correspondence is found in Russian хомяк ‘hamster, Cricetus cricetus’, Czech chomík ‘the same’, etc., which are regarded as back-formations from the reconstructed *choměstorb ‘hamster’ (+ a new suffix) (ESSJVIII 67). For Old Russian хомпсторъ, хомпстаръ the attested designation ‘an insect’ is considered erroneous; here too, the hamster is taken to be the actual meaning (ESSJVIII 67). The German equivalent Hamster < OHG hamustro is to be regarded as a borrowing from Slavic (< Russian Church Slavonic choměstorb ‘the same’, Kluge 1999: 353); by contrast, the relationship between the Slavic words and the Baltic ones—on the one hand, Lithuanian ³kamas ‘large rat’, kamūklis ‘the same’ (LKŽV 172ff.), Latvian kāmis ‘hamster, Cricetus cricetus’ (Būga 1958: 581); on the other hand, Lithuanian stãras ‘an animal, Citellus or Spermophilopsis’³⁰ (LKŽXIII 668), with the divergent meaning ‘the same; hamster’ (Fraenkel 1965: 896)—is disputed. Fraenkel on kāmas: “The Baltic designations can be reconciled with the Slavic ones, at most, in the way Būga notes: that they independently go back to a foreign source.” (1965: 213). ESSJ: “It is not entirely clear whether the Lithuanian compound *kamastaras < Proto-Slavic choměstorb existed, or whether kāmas and stãras were separate from the outset” 30 Probably the European ground squirrel, ‘Spermophilus citellus’. 24 | BERND GLIWA were borrowed separately” (ESSJVIII 67). As far as the hamster is concerned, the reference to its cheek pouches (ESSJVIII 68), used to carry supplies to its store, is highly apt: the hamster is notorious for stuffing its pouches full and building up large stores. The action of stuffing is aptly expressed in Lithuanian by kimšti ‘to stuff; eat greedily; pickle; gather’ (LKŽV 805); with ablaut, compare kamštis ‘cork, stopper’, kamšlys ‘one who eats a lot’, < *‘stuffs oneself’, kamšėklis ‘same’, kamasuoti ‘eat with a hearty appetite, gobble, stuff’ (LKŽV 172ff.), ²kimštinis ‘insatiable eater, glutton’³¹ (LKŽV 805). A comparison with Slavic *skomiti ‘compress, press’ (ESSJVIII 68) is therefore entirely justified. Regarding Lithuanian kamūklis, it should be noted that -uklis is a typical suffix for forming deverbal nouns denoting instruments (nomen instrumenti) or agents (nomen agentis) (Ambrazas 1993: 156, 178). Although denominal use does occur occasionally, for example Ežerūkis ‘pond’ (Bilkis 1998: 83), it is probably based on double diminutive formation < *Ežer-iuk-elis. miegūklis ‘sleeper’, cited as a diminutive (Ambrazas 2000: 110), belongs to the category mentioned above (nomen agentis), whereas barzdūklis, barzdžiūklis ‘field horsetail, Equisetum arvense L.’ (Ambrazas 2000: 184) should, in my view, be regarded as a blend of barzdà ‘beard’ and asiūklis, asiukai ‘horsetail’ (on blending, cf. Kabašinskaitė 1998: 63ff.) (or as a worn-down compound *barzdaasiuklis); cf. kaukabarždis ‘literally: Kaukas’s beard’ = ‘horsetail’ (LKŽV 419). With some probability, then, kamūklis can be understood as *‘stuffer’, referring either to the stuffed store of provisions or to the cheek pouches; in that case, naturally, it would be an originally Baltic formation from a verb *kam- ‘to stuff’. Būga (1958: 581f.) argues that Lithuanian kāmas, dialectal kāms ‘rat’ and especially Latvian kāmis ‘hamster’ are not borrowings from Russian on the basis of variation in vowel quantity, which has parallels in Lithuanian lāsas; Latvian lase, lāse ‘drop’, Lithuanian vēlės: Latvian veļi, vēli ‘souls of the dead’, and elsewhere. He also points out that Russian хомяк denotes not only the hamster but also the rat (1958: 581), and that Old Polish chomyk denotes the ‘mole’ (ESSJVIII 66). Thus, designations for rat and hamster appear readily transferable; this is also supported by the synonym žiurkėnas ‘hamster’, now also ‘steppe marmot, Marmota bobac’, formed on the basis of žiurkė ‘rat, worthless useless person, mouse, wooden clog’ (LKŽXX 810f.). Lit. stāras wird einerseits mit stirti ‘erstarren’ verglichen (Fraenkel 1965: 896 und Lit.), was für den Hamster: „Begegnet der Feldhamster einem Feind auf dem Boden, dann richtet er sich auf und bläht die Backentaschen auf. Begleitet wird das ganze von wütendem Knurren und Fauchen.“ (www.das-tierlexikon.de/eigentliche-wuehler.htm) wie auch für den Siebenschläfer (auch Steppenmurmeltier, lit. švilpikas, Marmota bobac) plausibel erscheint, da diese am Bau wachen und sich dazu auf den Hinterbeinen aufrichten, bei Gefahr durchdringend pfeifen und dann in den Bau abtauchen 31 As Urbutis has shown on the basis of Lithuanian gomùs ‘desirous, inclined’ and Latvian gāma ‘glutton’, such a development based on Indo-European *gem- ‘press, squeeze; heap up…’ is entirely familiar, so that kamaros ‘lust’ can also be placed alongside greedy eating (Lithuanian kamuóti, kamuŕkyti) (cf. Urbutis 1981: 62-68). Lit. kāmanos, kamiēnas, kumėlė 25 (cf., for example, www.darwin.museum.ru/expos/ZooGeo/8_07.htm). On the other hand, I would like to point to starinti ‘to pull, tug, carry heavily; to cut into, weigh heavily; to stiffen’, starỹs ‘one who pulls under strain’ (LKŽXIII 668t.), sterinti ‘to pull under strain’ (LKŽXIII 754), stirénti ‘to strut slowly and awkwardly; to carry heavily; to shiver (from cold); to squirm; to die; to rage; to sob’, stīrlioti ‘to lift or carry heavily; to quarrel’, stīrninti ‘to lift, carry heavily’, stirnóti ‘ditto, to shiver with cold’, stirnùs ‘strong, firm; rigid’, styrúoti ‘to carry heavily, toil, torment oneself; to stagger’ (LKŽXIII 817–830). It is not entirely clear whether these are always related designations, or whether the individual lemmata should be split into homophones. Without wishing to discuss this in detail32, I follow Fraenkel’s view: he treats all these terms together (insofar as he mentions them) and places them with German starr, etc. (1965: 896; Pokorny 1994: 1022); according to Büga (1959: 259), stōras ‘thick’ also belongs here—a highly stratified and differentiated word family. What is especially interesting here, however, is starinti ‘to pull, tug, carry heavily...’: given the more or less synonymous forms stīrlioti, styrúoti, sterinti, it cannot be regarded as a unique formation from stãras, suspected of being a loanword, on the model of Hamster > hamstern ‘to bring together a lot, to hoard’. If, therefore, we assume common inheritance, stãras could be understood as *‘one who or that which carries, hauls heavily’, an apt designation for the hamster. The Slavic compound chomèstōrō would thus be understood as < *‘that which hauls heavily when stuffed full’. The uncertain assignment of the names to the hamster remains problematic, on the one hand, as does, on the other, the question of whether the hamster or the other animals named actually lived in the settlement area of the Balts and Slavs during the period in question. Unfortunately, I know of no concrete data on this, so I offer a few considerations. As a steppe rodent, the hamster would have depended on arable land as a cultivated steppe in wooded regions. Since agriculture spread with the Indo-Europeans at the latest, this condition cannot be considered an obstacle. The existence of further names žiurkėnas, šalčias, balėsas ‘hamster’ may be taken as evidence of long-standing familiarity. On the other hand, the dormouse and the souslik are absent from Lithuania, so one may assume that the original meaning of both stãras and kamùklis was ‘hamster’—unless one insists on a borrowing after all. 9. Lithuanian ¹kamas ‘simple fellow; serf; boor’, which, given kēmeras, kemèras ‘ugly, coarse thing; difficulty; fool, clumsy person’ (LKŽV 556), kamėga ‘clumsy, ragged person’ (LKŽV 175), kēmas ‘indistinct stuff, scarecrow’, kemėklis ‘one who works clumsily; one who speaks with difficulty or awkwardly’, kemėklis ‘useless, obstructive stuff; scarecrow’, kemėšius ‘clumsy person’ (LKŽV 554ff.), among others of similar meaning, is scarcely a borrowing from Polish cham, Russian хам ‘ditto’33, as LKŽV (172) suggests. Compare also Slavic *ch- here with Russian хомяк. 32 One example: with stãrãs ‘wind with rain squalls’ and stãruoti ‘to rain with gusty wind’ (LKŽXIII 669), German steife Brise ‘strong wind’ may be cited as a related, though not identical, expression. 33 The relationship to the biblical Ham (Hebrew HM or, with vowels, HAM; Septuagint Χαμ, Polish Cham, as opposed, however, to Chus, Hebrew KWS, Sept. Χους, Polish Chus) is unclear. The biblical text (Genesis 9f.), however, seems to offer an ethnic and religious rather than a social classification; the linguistic categories Semitic and Hamitic derive from this from the eighteenth century onward (Körner 1990: 13). The interpretation of the Polish diminutive— 26 | B E R N D G L I W A ‘junk, shapeless stuff’ (Kardelytė 1985: IV 533), Chechen chomeček ‘old person’, < *‘hunchbacked person’, Chomek ‘personal name’, čumák ‘insatiable person’ (ESSJVIII 66); with a different suffix, Czech chomor ‘rubbish, junk’ (ESSJVIII 68); and possibly German Hampelmann ‘initially meaning “simpleton”, of uncertain origin, probably an imitative formation’ (Kluge 1999: 353). Latvian kamana ‘crooked woman’ (Mühlenbach 1927: 148) should also be mentioned here once again. It remains to show that clumsy, inactive people or cumbersome objects are given designations that go back semasiologically to ‘crooked’. This can be seen from the following Lithuanian examples: Kerėpla ‘a branched tree used as a trestle; a saddle on harness; an ugly, sprawling thing; a sprawling plant (especially a bush or tree); a clumsy person’, kėrė ‘stump; sod; layer on an overgrown pond, spongy bog soil; part of a spinning wheel; a small, slow person...’, kerėva ‘bow-legged, limping person’ (LKŽV 600ff.), kerėti ‘to grow, proliferate, branch out; to become strong’ and, antonymically, ‘to grow poorly, vegetate’ (LKŽV 605); cf. Latvian čerba ‘curl, person with tousled hair’, Russian kopenь ‘root’ (Fraenkel 1962: 241)³⁴. The other ablaut grades survive only in traces: kárka ‘(of livestock) knee; thigh; thighbone’ (LKŽV 289), kirtmuõ ‘bend’, kirùlis ‘hook for stirring; a tool for tensioning the fabric on a loom’, kìrlas ‘crutch’, kirkšnis ‘groin’, kirkšnìmas ‘ditto, armpit’ (LKŽV 838ff.), among others. Kevérza ‘a small, crooked tree with tangled branches; a crooked thing; a bow-legged, clumsy, awkward person; a fool; a small, crooked person...’, kėvalas ‘shell, husk; chaff; awkward person’, keveĩša ‘fool’, keveliuóti ‘to veer aside; to roll about; to come together; to roam’, kevėlsės ‘legs (derogatory)’ (LKŽV 671ff.), probably from Indo-European *keu- ‘to bend’ (Pokorny 1994: 588). Kreĩvas ‘crooked; crosswise, bent; uneven; wrong; primitive, bad’, kreĩvis ‘something crooked (a tree or person); inaccurate’, kreĩvys ‘bow-legged person; hunchback; left-handed person’, kreivùklis ‘one who walks crookedly (drunkards)’ (LKŽVI 508ff.). Lìngius ‘tall, bent-over person’, linginėti ‘to bend over, nod; to wander around without working’, lingýnas ‘boggy ground, swamp’, lingė ‘bow’, lingti ‘to bend’ and lìngti ‘to love one another’ (LKŽVII 526ff.); compare semasiologically German jemandem geneigt sein ‘to be inclined toward someone’, Neigung ‘inclination’: neigen ‘to incline’: nicken ‘to nod’. Similar cases are found in German, for example krummes or linkes Ding ‘crooked or left thing’ versus gerade, recht, richtig ‘straight, right, correct’; also links, linkisch ‘clumsy’, Swedish linka ‘to limp’ (Kluge 1999: 521), and Gelenk ‘joint’, compared with Lithuanian leĩkti ‘to bend’ as a g/k variant (Kluge 1999: 310). Better, perhaps, to point directly to lingti ‘ditto’, among others (see above). nobility, according to which the plebs descend from Ham, is therefore more likely to reflect an adaptation of biblical interpretation to existing terms than the creation of a term on the basis of this biblical passage. I am grateful to Prof. A. Holvoet, who drew my attention to the possible influence of this biblical passage on a borrowing and kindly sent me the name spellings in the Hebrew, Greek, and Polish Bible. 34 Der erwogene Anschluss an *(s)ker- ‘schneiden’ (Fraenkel 1965: 241) kann nicht überzeugen, vgl. die obigen Erörterungen zu kamienas. Lit. kāmanos, kamienas, kumėlė | 27 10. SYNTHESIS This yields a series of correspondences, each beginning with Lith. k-, Germ. h-, Sl. ch- / k-. The most straightforward way to reconcile them is to assume s-mobile in the Slavic forms, from PIE *(s)k- > Sl. chrustěti, Latv. skrāustēt ‘to crackle, crunch’, for Slavic via an intermediate stage *ksroust-, and Sl. chrupati < ksroup-: Lith. skùpsnūs, skrūpsi, skrupsėti ‘to crackle, crunch’, skràubis ‘fear, fright’ (Būga 1958: 443). Mel’ničuk questions this development, considering the development PIE *sk- > Sl. *ch- “недоказана и по чисто теоретическим соображениям невероятно”, apart from a few exceptions (Mel’ničuk 1968: 215–6). Accordingly, Lith. skujà ‘pine cone; needle of a conifer; pine branch...’ (= Sl. chvoj-) should be regarded as an old formation involving metathesis of initial *ks- > sk- before Baltic *ks- > š- (Mel’ničuk 1968: 215–6). On the other hand, the primary existence of *ks- is not always obvious, and the obligatory intermediate step *sk- > *ks- > Sl. ch- seems rather artificial. Therefore, the pragmatic position of ESSJ (e.g., ESSJᵥᵢᵢᵢ 22 et al.), PIE *(s)k > Sl. ch-, is plausible, while it is open to anyone to posit a corresponding formation such as that found in the pair skujà ‘pine cone...’: Sl. *chvoja > Czech chvoja ‘pine...’ et al. (ESSJᵥᵢᵢᵢ 125f.), with all the implications concerning antiquity and metathesis. For hops, we have Lith. kamantai, while kumėlė is uncertain, Sl. *chъmelъ > Russ. хмель: Germ. *humilo- > Dan. humlekop, Lat. humulus (Marzell 2000: II 906). In view of OE hymele ‘hops’ and ‘bindweed, bryony’, a PIE base *(s)kem- ‘to wind, bend’ should be assumed. For ‘yoke’, we have Lith. kamantāi ‘yoke’, Polish chomąto / Upper Sorbian chомот ‘same’, and German Hamen ‘yoke’. It should be noted that terms for the net, such as German Hamen ‘net’ and Slovenian homôt ‘net’ (ESSJᵥᵢᵢᵢ 69), may also belong here, with a different ablaut, as may perhaps kumelys: ‘9. mouth of a trawl; 10. float on a net’. As ESSJ already stated, this provides sufficient grounds to rule out borrowing and justifies comparison with Lith. kāmanos (ESSJᵥᵢᵢᵢ 70), assuming a more general meaning based on PIE *(s)kem- ‘to restrain; tame’ (cf. Pokorny 1994: 555). Latv. kāmis ‘hamster’ and Russian хомяк ‘same’ should be regarded as cognate, subject to uncertainty about the rodent’s historical distribution, and as based on PIE *(s)kem- ‘to press, squeeze’ in the specific sense ‘to stuff’, as found in Lith. kimšti—thus, literally, ‘the one who stuffs its cheeks full’. Lith. kamas ‘simple fellow; serf; boorish peasant’, kamėga ‘clumsy, ragged person’, kemėšius ‘clumsy person’, Polish cham, Russian хам ‘same’, Czech chomeček ‘old person’ (with a hump?), and German Hampelmann are regarded as a cognate group based on *(s)kem- ‘to bend, curve’. Zu den lit. Flussnamen Kamainė, Kāmė, Kamoja, Kamatis, Kāmona (Vanagas 1981: 144), Kumė, Kùmpė, Kumpālė (Vanagas 1981: 171), vgl. lit. kùmpti ‘biegen, krumm werden’, lett. kùmt ‘bucklig, krumm werden’, ebenso Kumėlupis (Vanagas 1981: 171), Kamarià, Kamārė, Kamarà (Vanagas 1981: 144) und den Rätselworten kambanka, kumpuringė ‘Fluss’ stelle ich die ukr. Flussnamen Хомо́ра, Хомор, Хомур, (ESSJᵥᵢᵢᵢ 28 | BERND GLIWA 69). Further derivatives from a base *(s)kom- ‘to bend’ include Slovak dialectal chomola ‘neck’ (ESSJVIII 68); semantically compare Lith. kākłas ‘neck’ (Fraenkel 1962: 205f.) and kumelvš ‘13. vertebral column’, Czech chomol ‘peak, treetop’: Lith. kaukurė ‘hill, peak’ < PIE *kou-k- ‘to bend, arch’ (cf. Pokorny 1994: 589), Old Polish chomla ‘wreath worn on the head to hold the hair’, Ukrainian χόмля, хомівка ‘women’s headdress’ (ESSJVIII 68f.). Another possible way to integrate Sl. ch- into this context may be to posit Sl. ch < *kh₂, which Kortlandt (1988: 303) cites without reference to the literature. Regarding the example cited, Russian xanatʹ ‘to seize, grab’ (Kortlandt 1988: 303), ESSJ states, without mentioning an etymology involving a laryngeal, “Основано, в конечном счете, на звукоподражании” and compares it with analogous Sl. *capati, *gabati (ESSJVIII 19). In the cases above, the analysis of ch- < kh₂- has a significant drawback: the laryngeal cannot be optional, and Slavic doublets beginning with k- or ch- should not occur35. Thus, two distinct roots, homophonous (at least at a late stage), can be clearly identified: *(s)kem- ‘to bend, curve, wind’36 and *(s)kem- ‘to press, squeeze, hinder’37, from which also derive ‘to tame, suppress’ and ‘to torment’: kamuiti, kamuoti, kamuoti ‘to weary, torment; to work hard, serve as a slave…’, on the one hand, and kimšti ‘to stuff’, on the other. It should not be overlooked that homophony can lead to semantic interference and contamination; for example, ‘to tame’ may also develop from ‘to bend, make inclined’: e.g., Lith. palankùs ‘submissive, flexible, inclined’ (LKŽIX: 239) in the context Kartais net sunku patikėti, kad laumė – tai suasmeninta pirmykščiam žmogui pavojinga ir grasi gamtos, visų pirma, vandens, stichija, kurią jis stengiasi pažaboti, palenkti savo labui (Vėlius 1979: 9), as well as the expression to bend someone to the yoke ‘to tame, dominate, educate someone’. LITERATURE38 AMBRAZAS, S. 1993: Daiktavardžių darybos raida, Vilnius: Mokslo ir enciklopedijų leidykla. AMBRAZAS, S. 2000: Daiktavardžių darybos raida II, Vilnius: Mokslo ir enciklopedijų leidybos institutas. BASANAVIČIUS, J. (Hrsg.) 1998: Lietuviškos pasakos įvairios IV, Vilnius: Vaga. BILKIS, L. 1999: Lietuvių priesaginiai diminutyviniai helonimai, Acta Linguistica Lithuanica 41, 51–95. BŪGA, K. 1958: Rinktiniai Raštai I, Vilnius: Valstybinė politinės ir mokslinės literatūros leidykla. BŪGA, K. 1959: Rinktiniai Raštai II, Vilnius: Valstybinė politinės ir mokslinės literatūros leidykla. DINI, P.U. 2000: Baltų kalbos. Lyginamoji istorija, Vilnius: Mokslo ir enciklopedijų leidybos institutas. 35An approach involving metathesis, Indo-European *k/HVm- (>k-) > Slavic *k/HVm- (>ch-), where H as in l, r in Russian голова, Polish głowa: Lithuanian galvà ‘head’, Russian голова ‘city’, Polish gród ‘fortress’: Lithuanian gardas ‘enclosure’, would formally suffice. However, there is no evidence for such laryngeal metathesis. 36 Hierzu lit. kumpis, ne. ham ‘kumpis’, fnhd. Hämchen – karka (Drotvinas 1987: 206), lett. kamps ‘krumm’, ai. kmárati ‘ist gebogen, krumm’ < ig. *kam-(?). *kṃ- ‘gebogen, krumm sein’ (Fraenkel 1962: 309), lit. kė́mos ‘Schenkel’ (LKŽV: 561), kìmbra ‘Bein (abwertend)...’ (LKŽV: 800). 37Related are Armenian kamel ‘to press, wring’, Old Norse hemja ‘to bridle, restrain’, Low German hamme ‘enclosed field’ (Pokorny 1994: 555). 38I would like to express my heartfelt thanks to Dr. A. Judžentis for his assistance in obtaining literature. LIT. kāmanos, kamienas, kumėlė 29 DROTVINAS, V., red. 1987: Lexicon Lithuanicum. Rankraštinis XVII a. vokiečių-lietuvių kalbų žodynas, Vilnius: Mokslas. ESSJ – Etymological Dictionary of the Slavic Languages, 4, 8, 10, 13, ed. O. N. Trubachev, Moscow: Nauka, 1977, 1981, 1983, 1987. EULER, W., 2002: Indogermanische Hirten und Krieger – ein Mythos oder Beginn europäischer Kultur? Res Baltica 8, 35–65. FRAENKEL, E. 1962: Litauisches etymologisches Wörterbuch 1, Heidelberg: Carl Winter Universitätsverlag, Göttingen: Vandenhoeck & Ruprecht. FRAENKEL, E. 1965: Litauisches etymologisches Wörterbuch 2, Heidelberg: Carl Winter Universitätsverlag, Göttingen: Vandenhoeck & Ruprecht. GIMBUTIENĖ, M. 1996: Senojĭ Europa, Vilnius: Mokslo ir enciklopedijų leidykla. GLIWA, B. 2003: Witches in Baltic Fairy Tales, Onomasiology Online 4, 1–14. [www.onomasiology.de] GLIWA, B. 2003a: Nešinas, vedinas, tekinas, Acta Linguistica Lithuanica 48, 19–34 GLIWA, B. 2003b: Die Hexe und der Junge (AaTh 327 F) und Der Junge im Sack des Freßdämonen (AaTh 327 C). Ein kulturgeschichtlicher Deutungsversuch litauischer Märchen. Fabula 3/4 Bd. 44, 272–291. GUOBYTĖ, R., STANČIKAITĖ, M. 1998: Traces of Human Activities in Pollen Diagrams from Biržulis Lake with a Short Review of Geological and Geomorphological Conditions in the Biržulis Area, Archaeologia Baltica, 3: 121–130. HEHN, V. 1963: Kulturpflanzen und Haustiere, Darmstadt: Wissenschaftliche Buchgesellschaft. HERTEL, J. (Hrsg. u. Übers.) 1998: Indische Märchen, Augsburg: Weltbild (Bechtermünz). HOAD, T. F. 1996: Concise Dictionary of English Etymology, Oxford, New York: Oxford University Press. JUŠKA, A. 1955: Lietuviškos dainos 1–2, Vilnius: Valstybinė grožinės literatūros leidykla. KABAŠINSKAITĖ, B. 1998: Lietuvių kalbos liaudies etimologija ir artimi reiškiniai, Vilnius: Mokslo ir enciklopedijų leidybos institutas. KAEGI, A., bearb. 1990: Benselers Griechisch-Deutsches Wörterbuch, (19. Auflage), Leipzig: Verlag Enzyklopädie. KARALIŪNAS, S. 1987: Baltų kalbų struktūrų bendrybės ir jų kilmė, Vilnius: Mokslas. KARDELYTĖ, J. et al., red. 1985: Rusų–lietuvių kalbos žodynas IV, Vilnius: Mokslas. KARULIS, K. 1992: Latviešu etimoloģijas vārdnīca I–II, Rīga: Avots. KAZLAUSKAS, J. 1968: Lietuvių kalbos istorinė gramatika, Vilnius: Mintis. KAZLAUSKAS, J. 1970: (Rez.) Baltų ir slavų kalbų ryšiai (Lietuvių kalbotyros klausimai X), Baltistica 6 (1), 130–132. KERBELYTĖ, B. 1999: Lietuvių pasakojamosios tautosakos katalogas 1 – Pasakos apie gyvūnus. Pasakėčios. Stebuklinės pasakos, Vilnius: Lietuvių literatūros ir tautosakos institutas. KLUGE, F. 1999: Etymologisches Wörterbuch der deutschen Sprache, 23. erw. Auflage, bearbeitet von Elmar Seebold, Berlin, New York: W. de Gruyter. KÖRNER, J. 1990: Hebräische Studiengrammatik, Leipzig: VEB Verlag Enzyklopädie KORTLANDT, F.1988: The laryngeal theory and Slavic accentuation. Bammesberger, A. (Hrsg.) 1988: Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems, Heidelberg: Winter, 299–312. KORZAKAS, K. (Red.) 1968: Lietuvių tautosaka V – Smulkioji Tautosaka, žaidimai ir šokiai, Vilnius: Mintis. KULAKOV, V. 2002: Pferdeopfer am Dollkeimer-Berg, Res Balticae 8, 209–218. LKŽ – Dictionary of the Lithuanian Language I–XX, Vilnius: Mintis, Institute of Science and Encyclopedia Publishing, LKI, 1956–2002. LUCHTANAS, A. 1992: Rytų Lietuva I. tūkst. pr. m. erą, Lietuvos Archeologija 8, 56–85. LÜTHI, M. 1996: Märchen, 9. durchgesehene und ergänzte Auflage, bearbeitet von H. Rölleke, Stuttgart, Weimar: Metzler. LYBERIS, A. et al. (Hrsg.) 1979: Pirmasis lietuvių kalbos žodynas, Vilnius: Mokslas. 30 | Bernd Gliwa Mel’ničuk, A. S. 1968: Корень *kes- и его разновидности в лексике славянских и других индоевропейских языков. Этимология 1966, 194–240. Marzell, H. 2000: Wörterbuch der deutschen Pflanzennamen I–V, Köln: Parkland. Mažiulis, V. 1988: Prūsų kalbos etimologijos žodynas I (A–H), Vilnius: Mokslas. Mažiulis, V. 1993: Prūsų kalbos etimologijos žodynas II (I–K), Vilnius: Mokslo ir enciklopedijų leidykla. Mažiulis, V. 1997: Prūsų kalbos etimologijos žodynas IV (R–Z), Vilnius: Mokslo ir enciklopedijų leidybos institutas. Morkūnas, K. et al. (Red.) 1977: Lietuvių kalbos atlasas 1. Leksika, Vilnius: Mokslas. Mühlenbach, K. (1925–) 1927: Latvian-German Dictionary II, edited, supplemented, and continued by J. Endzelin. Riga: Latvian Cultural Foundation. Mylius, K. 2001: Langenscheidts Handwörterbuch Sanskrit-Deutsch, Berlin etc.: Langenscheidt. Otrębski, J. 1961: Beiträge zur baltisch slavischen Namenkunde, Beiträge zur Namenforschung 3, 262–265. (zit. Vanagas 1981: 144) Otrębski, J. 1965: Gramatyka języka litewskiego II. Nauka o budowie wyrazów, Warszawa: Państwowe Wydawnictwo Naukowe. Pokorny, J. 1994: Indogermanisches etymologisches Wörterbuch I, (3. Aufl.) Tübingen, Basel: Francke. Puhvel, J. 2001: Lyginamoji mitologija, Vilnius: Lietuvių literatūros ir tautosakos institutas. Rätsch, Ch. 1998: Enzyklopädie der psychoaktiven Pflanzen, Aarau: AT-Verlag Razmukaitė, M. 1998: Priesaginiai oikonimai iš asmenvardžių ir apeliatyvinių asmenų pavadinimų, Acta Linguistica Lithuanica 41, 159–201. Renfrew, C. 2000: Die Indoeuropäer – aus archäologischer Sicht. Spektrum der Wissenschaft. Dossier: Die Evolution der Sprache, 40–48. Rix, H. et al. 2001: Lexikon der indogermanischen Verben. Die Wurzeln und ihre Primärstammbildungen, Wiesbaden: Dr. Reichert. Sabaliauskas, A. 1968: Baltų kalbų naminių gyvulių pavadinimai. Lietuvių kalbotyros klausimai 10, 101–190. Skardžius, P. 1996: Lietuvių kalbos žodžių daryba. Id., Rinktiniai raštai 1, Vilnius: Mokslo ir enciklopedijos leidykla. Skardžius, P. 1998: Lietuvių vandenvardžiai su -nt-. Id., Rinktiniai raštai 4, Vilnius: Mokslo ir enciklopedijos leidybos institutas. Slančiauskas, M. 1974: Šiaurės Lietuvos pasakos, Herausgegeben von N. Vėlius und A. Seselskytė, Vilnius: Vaga. Smoczyński, W. 2000: Untersuchungen zum deutschen Lehngut im Altpreußischen, Kraków: Wydawnictwo Uniwersytetu Jagiellońskiego. Sruoga, B. 1957: Dainų poetikos etiudai. Id., Raštai 6, Vilnius: Valstybinė grožinės literatūros leidykla. Šimkūnaitė, E. 2001: Gyvenimo receptai, Vilnius: Žuvėdra. Toporov, V. N. 1975: Русский язык. Словарь (А–Д), Москва: Наука. Toporov, V. N. 1980: Русский язык. Словарь (I–K), Москва: Наука. Ulvy das, K. (Red.) 1965: Lietuvių kalbos gramatika 1, Vilnius: Mintis. Urbutis, V. 1972: [Rez.] Этимология 1966, Москва: Наука. Baltistica 8 (2), 207–213. Urbutis, V. 1981: Lie. gomus, -i „norus, noringas, linkęs“, la. gama „kas pernelyg daug valgo“ ir kiti giminiški žodžiai. Id., Baltų etimologijos etiudai, Vilnius: Mokslas. Vanagas, A. 1981: Lietuvių hidronimų etimologinis žodynas, Vilnius: Mokslas. Vanagas, A. 1988: Ringuoja, linguoja, gumba mūsų upės, Gimtasis kraštas, 25.02–02.03, 8. Varnas, A. 1998: Horse Burials in Lithuania, Archaeologia Baltica, 3: 291–294. Vėlius, N. 1979: Laumių dovanos, Vilnius: Vaga. Lit. kāmanos, kamienas, kumėlė | 31 Vėlius, N. 1987: Chtoniškasis lietuvių mitologijos pasaulis, Vilnius: Vaga. Vėlius, N., (Hrsg.) 1996: Baltų religijos ir mitologijos šaltiniai I, Vilnius: Mokslo ir enciklopedijos leidykla. Volkaitė-Kulikauskienė, R. 1999: „Žirgų kapai“ ir jų simboliai, Augalų ir gyvūnų simboliai, Vilnius: Gervelė, 306–334. (Senovės baltų kultūra, 5) Bernd Gliwa Received 2003 10 16 Sargeliai LT 60433 Žaiginio pšt.