LIETUVIŲ KALBOS MORFOLOGINĖ SANDARA IR JOS RAIDA

LIETUVOS TSR MOKSLŲ AKADEMIJA

ON THE DEVELOPMENT OF IMPERATIVE FORMS

D. SKURKAITĖ

§ 1. Lithuanian dialects and old writings have two types of imperative forms: with the formative -k (nešk, daryk) and without the formative -k (neš, darai). The singular 1st-person forms of the imperative without -k are still in use among the Eastern Dzūkians¹ (in Dūkštas, Ignalina, Švenčionėliai, Švenčionys, and Tverečius); elsewhere, only isolated forms of this type occur, e.g.: numėš Alks, Antz, Stlm, imi, neši Ds, Sv, Dgl, im Ssk, klausai Brd, Lknv, Pn, imį, bėgi, neši Krtn. Imperative forms without the formative -k are still used somewhat more frequently in the vicinity of Dieveniškės, Rodūnia, Lazūnai, Gervėčiai, and Apso.

§ 2. A. Schleicher and A. Becenberger were the first to present the singular 2nd-person imperative forms without the formative -k in the Lithuanian language of old writings and to attempt to explain their origin. Both they and other later researchers recognized that some of these forms derive from the optative, namely:

a) o-stem verbs, singular 2nd person: graištai “grab,” valgai, guldais rarely in Pn, Stlm, Ds; frequently in Dkšt, Rš, Sld, Lkm, Plš, Dglš, Ign, Tvr, Klt, Kret, Ršk, Ml, Dtn, Švnc, Žbl, Ad, Dv, Aps, Grv, Lz (here the reflexive form guldais is used), Žt. (These forms are most widespread in the dialects.)³

b) i-stem verbs, singular 2nd person⁴: ravì Dglš, Ign, Grv⁵, Lz,

¹ Dialect material for this article was collected from descriptions of individual dialects (dissertations, diploma theses, articles). Use was also made of the Lithuanian dialect atlas material collected at the Institute of the Lithuanian Language and Literature of the Academy of Sciences of the Lithuanian SSR.

² Žr. A. Schleicher, Handbuch der litauischen Sprache, Prag, 1856, p. 230 tt. (toliau: Handbuch...); A. Bezzenberger, Beiträge zur Geschichte der litauischen Sprache, Göttingen, 1877, p. 208 tt., 222 tt. (toliau: Beiträge...).

³ Besides imperative forms of o-stem verbs of this type, there are also paródz, šildzies in Grv, nekliūd in Klt; these are formed on the pattern of the imperative forms of (i)a-stem verbs (cf. § 5).

⁴ On the imperative of i-stem verbs, see E. Nieminen, Der urindogermanische Ausgang ai des Nominativ—Akkusativ Pluralis des Neutrums im Baltischen, Helsinki, 1922, p. 108 (hereafter: Ausgang...).

⁵ In Gervėčiai, the recorded singular 2nd-person imperative forms of i-stem verbs pažiūr, til, tilė, apparently have also been formed (cf. footnote 3) on the pattern of the imperative forms of -(i)a-stem verbs.

153

Zt, mi.lĭ Sv, ti.lĭ, tū.ri.s Dtn, ravī. Sld, Klt, Vsn, penī, turiēs Lkm, tū.ri, ravī, tū.ries, gulīēs Ad⁶, Plš.

c) In the dialects, we find almost no 2nd person singular imperative forms of (i)a-stem verbs with the formative te. This formative is characteristic of the 3rd person imperative (see § 13). J. Endzelīns mentions the form vedzie⁷ (cf. also the i-stem form tilīe, see footnote 5).

§ 3. The optative forms acquired imperative meaning, it seems, already in Proto-Baltic times: in Prussian, old optative forms are also used with this meaning (wedais, wedeys “lead,” endiris “look,” kirdijti “listen,” and so on)⁸; Latvian imperative forms darait, daraities, lieciet, liecieties and so on are likewise regarded as relics of the old optative⁹.

§ 4. While agreeing on the optative origin of the forms mentioned above, linguists have explained such forms very controversially, for example: gelb DP 78²⁹, 80⁹, dūd DP 84²⁸, 106¹⁶, 113³¹, dūdi DP 85²⁶, parōd DP 161⁸, atteid DP 113 ²⁸.¹⁰ In our opinion, F. Specht and E. Fraenkel¹¹ come closest to the truth: they regard the dialect form duō Dglš, Tvr, Ltk, Ml, Dtn, Ad, Grv, Lz... as an old imperative¹². There is reason to believe that the old root imperative of stōti was *stō; in the dialects it is represented by the forms atstō, stā Lkm, acistā Plš, Ad; cf. also testā (§ 13). The imperatives miēk, gied Plš, miēk Dtn, pāweyzd PK 66¹⁴, pawiżd, wiżd KN 11, may be equally ancient.

⁶ The reflexive imperative forms of i-stem verbs can also be stressed on the root: tu.ries Ad; cf. J. Kardelytė, “Double forms of the imperative mood and their use in the dialects of Eastern Lithuania,” LKK, vol. 1, Vilnius, 1957, p. 183 (hereafter: LKK, vol. 1).

⁷ Žr. J. Endzelīns, Latviešu valodas gramatika, Rīgā, 1951, § 691 (toliau: Gramatika...).

⁸ Žr. J. Endzelīns, Senprūšu valoda, Rīgā, 1940, p. 119 tt.

⁹ See A. Bielenstein, Die lettische Sprache nach ihren Lauten und Formen, II, Berlin, 1864, p. 160 (§ 450) (hereafter: Die lettische Sprache...); Ф. О. Фортунатовъ, Критическій разборъ сочиненія Г. К. Ульянова «Значенія глагольныхъ основъ въ лит. — славянскомъ языкѣ», «Сборникъ Отдѣленія русскаго языка и словесности Имп. Ак. Наукъ», LXIV, № 11, Петербургъ, 1897, p. 143–148 (hereafter: Критическій разборъ...); K. Brugmann, Grundriss der vergleichenden Grammatik der indogermanischen Sprachen, II, 3, Strassburg, 1916, § 469 (hereafter: Grundriss...); J. Endzelīns, Gramatika, § 692.

¹⁰ See A. Bielenstein, Die lettische Sprache..., p. 160; O. Wiedemann, Handbuch der litauischen Sprache, Strassburg, 1897, p. 112; J. Endzelīns, Gramatika..., § 691; J. Otrębski, Gramatyka języka litewskiego, III, Warszawa, 1956, p. 277 (hereafter: Gramatyka...); J. Zubatý, Baltische Miscellen, IF, vol. 4, p. 176 (hereafter: IF); A. Schleicher, Handbuch, p. 226 ff.; A. Bezzenberger, Beiträge..., p. 217 ff.; F. Specht, Širwyds Leben und Schriften, „Širwyds Punktay sakimu“, Göttingen, 1929, p. 41×–42× (hereafter: Širwyds Leben und Schriften...); by the same author, Zu den altpreussischen Verbalformen auf -ai, -ei, -sai, -sei, KZ, vol. 55, Göttingen, 1928, p. 174 (hereafter: KZ, vol. 55); A. Senn, Die athematischen Verba in Mikalojus Daukszas Schriften, St. Balt., vol. 4, Roma, 1934–1935, p. 86; Ф. О. Фортунатовъ, Критическій разборъ..., p. 160; В. К. Поржезинскій, Къ исторіи формъ спряженія въ балтійскихъ языкахъ, Москва, 1901, p. 30 ff. (hereafter: Къ исторіи...).

¹¹ See F. Specht, KZ, vol. 55, p. 182; E. Fraenkel, Der Stand der Erforschung des im Wilnagebiete gesprochenen Litauischen, BS, II, Wilno, 1936, p. 41 (hereafter: BS, II).

¹² Cf. V. Poržežinskis’s explanation of the origin of (te)ei; see B. K. Поржезинскій, Къ исторіи..., p. 30 ff.

154

wey, pawėzd SD 7, 243, gelb Vln 20₂₄, DP 78₂₇, gelb DP 79₂₈, gelb DP 309₁₅, gelb DK 15₁₇, gelb DK 15₃₄, 45, and so on.

§ 5. Dialects attest a great variety of 2nd-person singular imperative forms of (i)a-stem verbs, e.g.: krau̇, rau̇, apsiau̇, ragau̇, šūkau̇ Dkšt, Krd, Dglš, Lkm, Klt, Ad, Tvr, Grv; šluȯ (and šluȯ), dainuȯ Sld, Lkm, Plš, Švnč, Ad, Grv; grȧ, atsistȧ, kapȧ Krd, Ign, Lkm, Plš, Dglš, Ml, Tvr, Ad; gi̇, respectively gi̇, ri̇; padali̇ Plš, Ml, Dglš, Tvr, Dtn, Ad; bū̇, respectively bū̇, supū̇ Lkm, Klt, Ad, Ml; dūj, naktūj Dtn, Ad.

In our view, J. Kazlauskas correctly explains the origin of these forms: when some forms of the old root imperative (duo, išej) coincided with the verb’s infinitive stem, the latter began to be used with the meaning of the 2nd-person singular imperative. Once a connection arose between imperative forms and 2nd-person singular present indicative forms, J. Kazlauskas argues, alongside forms such as bū̇, riešutaū̇, etc., forms such as būni, riešutaūni, etc., arose on the model of the 2nd-person indicative forms: šluoj (šluoj, šluoj), meluoj Lkm, Plš, Dglš, Tvr, Dtn, Ad, Grv, Lz, griau̇n, šūkaūn Krd, Lkm, Plš, Klt, stoj (stoj), joj, kapȧj Lkm, Ign, Plš, Dglš, Dtn, Tvr, Ad, Lz, Aps, būn (būn), griūn Lkm, Klt, Ign. Alongside the forms stoj, šluoj, the forms būj, grūj, siūj arose in Plš, Lkm, Tvr, Dtn, Ad.

§ 6. The 2nd-person plural may likewise have imperative forms of root, optative, and contaminational origin¹⁴. In reflexive forms of this person, the reflexive particle may occur before or after the personal ending, e.g.: šūkuojtes, keltė̃s Ad, šilčtes “warm yourselves,” gultės Grv, gulties Dtn, Ml, liautės, liaunietės “stop,” Dglš, sėditės, klōjitės Lz, džiaugtės, džiaugietės Tvr, and kelieṡte Ad, liaunietėste Dglš, džiaugiėste Tvr.

These reflexive forms are frequently used in the dialects of Adutiškis, Daugėliškis, Mielagėnai, and Tverečius when commands or instructions are addressed to two or three people; when a command is addressed to a larger number of people, the 2nd-person plural ending is doubled, e.g.: džiauktėtės Ml, Tvr, keltėtės Dglš, keltėtės Ad, and keltėste Dglš¹⁵.

§ 7. Throughout Lithuania, the 1st-person plural present indicative form may be used with the meaning of the 1st-person plural imperative (with or without the formant -k), e.g. gaūdom “let’s catch,” vežam “let’s transport.” The use of indicative forms with the meaning of the 1st-person plural imperative, and the rarity of imperative forms for this person¹⁶, suggest that the 1st-person plural imperative forms originated later than the 2nd-person singular and plural imperative forms. For a long time, the forms used in the 1st-person plural imperative were probably injunctives; compare the athematic verb ‘to go’ forms that have survived in many parts of Lithuania: indicative forms used with the meaning of the 1st-person plural imperative: eima, eimà Prg, eimė, a(i)mė Grš, Vilkv, eimė Jrb, aimė Raud, Knv, Rgv, Kp, Pl, ai(i)mė Lkm, amė Dglš, Tvr, Ad, a(i)mà Zt, and so on; compare the 1st-person dual forms aivà Raud, Vl, Vilkv, eivà Jrb, Grš.

¹³ See J. Kazlauskas, Iš optatyvo istorijos, LKK, vol. 4, Vilnius, 1961, pp. 82—83.

¹⁴ For examples, see J. Kardelytė, LKK, vol. 1, p. 186. It need only be noted that the author’s claim that disyllabic 2nd-person plural imperative forms are stressed only on the ending seems unfounded; compare dėjste, laīte “allow,” rīcte, lañdže (and kemžtė), lipte, geñte, šākte (and larntė) Tvr.

¹⁵ Žr. ir J. Otrebski, Wschodniolitewskie narzecze twereckie, I, Kraków, 1934, p. 385.

¹⁶ E.g., rare 1st-person plural imperative forms without -k: gañaimė, kraūmė, grēbme “let’s rake,” vakariūjome Ml, Ad, rāšaimė, tilīme, paṣākme, siedziēmės Tvr, ūlgaime Švnč, lēidme, ragāūnme, stātāime Aps.

155

§ 8. In some Eastern Dzūkian dialects, first-person plural imperative forms ending in -me express a command directed at two people—the listener and the person being addressed (e.g., amė, padėsi vežimą nukrauti). When the ending -te is added to such a form, the command applies to several people (bōbos, amėte rīt viōgaut). Here, the ending -te, which gives the form plural meaning, has been added by analogy with the second-person plural.

§ 9. Nowadays, the particles te, tegu(l), testa, lai, laigul, etc., are generally used in third-person imperative (permissive) forms.

The origin of the particle te remains unclear to this day. A. Becenberger identified it with the deictic particle te, commonly used with verbs (Kodelei tu man tikrai viena burtą... tedavei BB Joz 17, 14, Schitie pafkucziau/ieghi wiena hadiną te dirba. BB Mt 20, 12, Jauniau/is tebėra pas mu/ų Tiewą BB 1 Moz 42, 13)¹⁷. J. Endzelīns and J. Otrębski explain te similarly¹⁸.

The particle te, when fused with a verb, may also be used as a separate word. Even in old writings, in sentences containing several third-person imperative forms, te usually occurs only before the first verb, e.g.: Czia te dūdie wiens wienam Wenčianōnės Žėdus o Klebonas fudedade abejū dešinę Ranką Daw Pam 9, Te ne lipa ing namus ney ineina Bt 13, 15.

In old writings, the particle gi is also used with te, e.g.: tėgistow BB 2 Krn 1, 9, tegiweifd BB 1 Moz 43, 33, tėgi parod (alongside te parodig), tėgi ne regis DP 264₃₄, tegi ėssi DP 433₂.

Old writings also contain examples in which the vowel of the particle te contracts with the first vowel of the verb or its prefix, e.g.: totei “let him come” KN 264, totliektis “let him remain” DP 70₃, tėssi “let it be” DP 51₄₁, tė/si DP 82₂₃, 194₅, 257₈.

The reflexive particle is usually found between te (ta) and the verb root, e.g.: tesidėd DP₂₄, te/įskiria BB 1 PvP 7, 15, tesided AK 42₉ || tęsispardai Jrb, tesi-suk Grdž, tesizi.nai Šl, Sv, Plš, tesizi.no Uog, Plv, tesizi.na Pnm, Ad, tasiprausia Skp, tasdārai Rmš, tesizi.nai Tvr. In some dialects, forms of this kind have the reflexive particle at the end, e.g.: tėpešas Krtn, temōkais Lkv, tegānos Gerv, tė.siuvas Ds.

§ 11. Much more often than with te, the third person is expressed by a verb form with the particle tegul (tagul) or its shorter form tegu (tagu), e.g.: tegul auga Smn, Lp, Glv, Bgs, Plš, Ad, Ds, tėgul būn Kal, Pgr, tegūl nėš Erž, Všk, tegul dāro Pgr, Grdž, Vs, tėgul ār Šl, Pkp, tegul’ lėkia Žsl, Kš, tegul dāro Knv, tegul kās Pš, Škn, tagūl eina Plv, Pnd, Dbk, tegūl tili Kp, Skp, Sv, Ktk, tegul sū.kas Aln and tegu nėša Gršk, Smn, Vs, Lp, Knv, tegū būn, tegu atimie Jrb, Grdž, tagū mōkos Krtn, Lkv, tegu sāko Všk, Kal, Erž, Gerv, tegu dirb ret. Brž, tegū praišia Glv, Aln, Ktk, Ad, tagū mirksta Kp, Skp, and so forth.¹⁹

¹⁷ See A. Bezzenberger, Beiträge..., p. 211. On one occasion te is even used with an adjective: tegaliau fioioje privartije BB Ez 40, 20 (cf. modern examples: Kartą ten tebuvau. Jis ten tebėgyvena; see J. Otrębskis, Gramatyka..., p. 368).

¹⁸ Žr. J. Endzelynas, Baltų kalbų garsai ir formos, Vilnius, 1957, p. 191; J. Otrębskį, Gramatyka..., p. 368.

¹⁹ The particle tegul is also used alongside te in old writings. It appears to be absent only from J. Chilinskis’s Bible and the Wolfenbüttel Postil; see A. Bezzenberger, Studien über die Sprache des preussischen Enchiridions, KZ, vol. 41, Göttingen, 1907, p. 112 ff. By contrast, in K. Sirvydas’s writings tegul is used especially frequently: in “Punktuose Sakymų,” only one form with te was found, te klausai 338₁ (F. Specht, Syrwids Leben und Schriften..., p. 42).

156

In Eastern Dzūkian dialects, the forms mentioned above with particles testą, testau, are still found, for example: testą eina Knv, testą lieka Ad, testą sėdas Tvr, testau pāravi “let them prepare” Ad. F. Kuršaitis²⁰ considered gulėti to be a permissive form of the verb²¹. A. B ecenbergeris discerned here a combination of the particles te, gu and lai. In this combination, gu is an interrogative particle, also found in other words, for example jeigu, negu, while lai corresponds to Prussian lai (cf. bōulai “that it may be,” ēilai “let it go, so that it might go”), and to Lithuanian and Latvian lai, used with imperative forms. This view of A. B ecenbergeris, later supported by other linguists as well, was rightly criticized by E. Hermanas²². E. Hermanas’s conjecture that tegu arose non-phonologically from tegul seems well-founded.

§ 12. Since the particle lai is used only in the northern dialects of Lithuanian (for a long time it was thought to occur only in Samogitian dialects²³), it was regarded as a Curonianism or a Latvianism²⁴. K. Būga, noting that the Prussians also had the particle lai, cf. ēilai, auotilai “they wish,” conjectured that it was of non-Latvian origin²⁵.

There are two views concerning the origin of the particle lai. Some linguists regard Latvian lai and Lithuanian lai as a remnant of the old imperative of the verb (Latvian laist, Lithuanian leisti)²⁶. They rely on the form laid, used in old Latvian writings instead of lai, and on corresponding particles in Slavic languages (Russian пусть (from пустить), пускай (from пускать), Polish niech, niechai, Ukrainian нехай, Belarusian няхай), which are also old imperatives.

A. B ecenbergeris and K. Būga did not agree with this view; they considered laid, found in old writings, to have arisen from lai and the postposition da²⁷. The latter explanation,

²⁰ Žr. F. Kurschat, Grammatik der litauischen Sprache, Halle, 1876, p. 372 (§ 1369).

²¹ Similarly, testą is regarded as a permissive form of the verb stovėti; see J. Zubatý, IF, vol. 4, p. 474.

²² E. Hermann, Litauische Studien, Berlin, 1926, p. 255 tt.

²³ F. Kuršaitis (see F. Kurschat, Grammatik..., p. 301, § 1160) stated that lai is used in Klaipėda and Kretinga, while the eastern Samogitians have tegul. K. Būga (see K. Būga, Kalbos mažmožiai, “Švietimo darbas,” 3–4, Kaunas, 1921, p. 47) indicates that lai is used in Skuodas, Mosėdis, Alsėdžiai, Rietavas, and Kvėdarna. It may be added that lai (or lei) is also found in Kretinga (lai eit), Viekšniai (lei miliė), Kalnalis (lai skaīto), Šelėniai (lai dirb), Kuršėnai (lei nōrie, rarely lai einie), Šakyna (lai būnie, lei dirb), Šiauliai, Pakapė (lai ār'), Steigviliai (lei kabalīoji), Biržai (lei peīk), Pabiržė (lei paliek), and Svėdasai (lai gu.li).

²⁴ Cf. J. Endzelīns, Gramatika..., § 694.

²⁵ See K. Būga, Славяно-балтийские этимологии, “Rinktiniai raštai,” I, Vilnius, 1958, pp. 452–453. (hereafter: RR, I).

²⁶ See A. Bielenstein, Die lettische Sprache..., pp. 164, 208, 365; J. Zubatý, Über die sogenannten Flickvokale des lettischen Volkslieds; Věstník královské České společnosti náuk, Třída filosoficko-historicko jazy kopytná, vol. 19, Prague, 1855, p. 13; by the same author, IF, vol. 6, p. 294; J. Endzelin, Lettische Miszellen, KZ, vol. 42, Göttingen, 1909, p. 375; by the same author, Gramatika..., § 690; by the same author, Kritische Anzeiger, AfslPh, vol. 32, Berlin, 1911, p. 295. E. Fraenkel, BS, II, p. 64; by the same author, Litauisches etymologisches Wörterbuch, Göttingen, 1955, p. 328 ff.; Chr. S. Stang, Das slavische und baltische Verbum, Oslo, 1942, p. 245 (hereafter: Verbum...); J. Otrębski, Gramatyka..., III, p. 365. K. Brugmanas, A. Trautmanas, E. Bernekeris, and others did not accept either of the two views on the origin of lai mentioned in the article (see K. Brugmann, Slavisch li und lettisch lai, IF, vol. 15, Strassburg, 1903–1904, p. 380 (hereafter: IF, vol. 15); by the same author, Grundriss..., II, 3, p. 974; A. Trautmann, Altpreussische Sprachdenkmäler, p. 285; E. Berneker, Slavisches etymologisches Wörterbuch, I, Heidelberg, 1913, pp. 716, 698; F. Solmsen, KZ, vol. 44, p. 184.

²⁷ A. Bezzenberger, KZ, vol. 41, p. 122; K. Būga, RR, I, pp. 452–454.

157

which E. Nieminen later also agreed with,²⁸ seems more accurate²⁹. For Russian пусть, пускай are clearly of later origin: in Old Russian, the particle да was used in their place (replacing forms without a particle). In Latvian, the third-person imperative is used only with lai. That the imperative form lai became a particle in very ancient times (as, for example, perhaps later tegul or the aforementioned Slavic particles changed), is doubtful.

§ 13. Forms of the third-person imperative of o-stem verbs with the old optative ending -ai occur, though rather rarely, in Samogitian and Western and Eastern Aukštaitian dialects, e.g.: tarašai, tamokais Lkv, tegūnai Vds, terodai Prg, tesakai Šl, tesispardai, tegu suvalgai Jrb, Grdž, tesizinai GĮ, tedarai Pls, Dglš, Tvr, Dtn, Ad, tadarai Rš, tesiži.nai ret. Sv, Ktk.

Older imperative forms are currently also used with verbs of other stem classes, e.g.: leimūlė “let him/her love,” leibūnė, tevežie Vks, teneši, tepraušis Gerv, tanešie, tanaudie Lkv, tebūnie, težiūrie Pgr, tegulie, teguli, tedirbi ret. Kps, VIkv, tebėgie, tegu atimė Jrb, testovį, ret. teeinie Sv, testovį, teneši ret. Pnm, testą, testau, temiek, tevažiuoj, testovį, tepriraū, teišmok, tegied Dglš, Tvr, testą, téi “let him/her go,” testau, tepadali, teguli Ad.

§ 14. Much more frequently, the third-person imperative meaning is expressed by third-person present indicative forms with a particle (te, tegu(l), testa, lai, and the like). Such forms are used throughout Lithuania and are also abundant in old writings. In some dialects, the third person of the imperative can also be expressed by a third-person future indicative form, e.g.: tėgu pamatīs Gerv, lai būs Škn, tegul važiuoš, tegū pavalgis Lp, testà ateis Knv; compare also forms of this type found in old writings: tegul bus SP I 25₂₀, 66₄, 152₁₇, tegul ne bus SP II 100₁₀, tegul bus SE 276; tegul karis paštatītus KN 25.

In the Kaniava dialect, the third-person imperative can also be expressed by a conditional form with a particle: testà ateit.

§ 15. As we have seen, the third person of the imperative differs from the other imperative persons through the presence of a certain particle and the use of indicative forms. There is reason to believe that third-person imperative forms are of later origin than second-person singular and plural forms. For a long time, the imperative meaning of the third person in Lithuanian and Latvian was apparently expressed by indicative forms (cf. the fixed expressions rūna tave giltinė, Latvian paraj vinu Jupis).³⁰

Later, second-person imperative forms became established in their place; compare dies duo tau³¹, trauk tave kelmai “may the stumps drag you away,” šūk tave ubagas Krtn, rāu(k) aną pikį “may the devil take her,” trauk jum paraliai, trenk tave perkūnas Škn³² and in old writings...

²⁸ In E. Nieminen’s view, the particle lai may derive from *lāi or *lāi (E. Nieminen, Ausgang..., pp. 112 ff.).

²⁹ Cf. Lithuanian lai, Latvian lai, and Prussian -lai; the comparison with Russian ли (see A. Leskien, Die Deklination im Slavisch-Litauischen und Germanischen, Leipzig, 1876, p. 49; F. Solmsen, KZ, vol. 44, p. 184; W. Vondrák, Vergleichende slavische Grammatik, II, Göttingen, 1928, p. 288; K. Brugmann, IF, vol. 15, p. 380; K. Büga, RR, I, pp. 452–453; E. Nieminen, Ausgang..., p. 113.

³⁰ See J. Endzelīns, Gramatika..., § 694; J. Endzelynas, Baltų kalbų garsai ir formos..., p. 191 (§ 398).

³¹ See J. Endzelīns, Gramatika..., § 694.

³² Cf. Russian Отохни твой язык.

158

(Karalius tho neded (chirdie)fp BB 2 Sam 19, 19), atmete “let him reject” BP 26₂₅.³³ Perhaps the circumstance that the forms derived from the optative mentioned in § 2 began to be used in place of non-paradigmatic (injunctive or other) third-person imperative forms helped them become established with this meaning. With the loss of the third-person ending *t (*darai-f), the forms *darai(s) and *darai(t) became indistinguishable. Therefore, alongside them, the particle te began to be used to express the third person of the imperative (and elsewhere, perhaps initially everywhere (?), lai). This particle was intended to specify the third person of the imperative. Over time, it was replaced by the semantically bleached imperative forms tegul, testa, laigul, and the like. They could also be used alongside the older third-person imperative form with the particle te; cf. tegul tedūmo DP 221₁₅, tegul te diūdes DK 120₁₀ (in E. Zittig’s edition), kožnas tegul paties fawo darbą tedatina Bt PvG 6, 4, o tegul tenudalija Bt PvG 6, 6, tegul tedaro taip KIG 521³⁴.

§ 16. The word sau may be used with the third person of the imperative, having greatly weakened the force of command, e.g.: tegū sau rėkauna Tvr, tegul sau nė.šas Pnm. A. Brückner regards such a form from the Merkinė dialect (tegul sau eina) as a Slavic calque (cf. Polish niech sobie idzie)³⁵. Yet the imperative tesižinai, tes ižino is very frequently used in Lithuanian dialects (cf. § 9). In many places (Raguva, Palėvenė, Svėdasai, Alanta, Kuktiškės, etc.), the particle te has survived only in this form, alongside tegul, tegu, used in all other cases tegul, tegu. As can be seen, the particle te survived in the form tesižinai, tes ižino because of the reflexive particle, which emphasizes the permissive nuance: tesižino =“let him know for himself”. Therefore, the use of sau with the third person of the imperative might also be considered to have arisen independently of the analogous Polish niech sobie idzie.

§ 17. How did the imperative with the formant -k arise in Lithuanian?

In the Indo-European languages, various nuances of command in the imperative can be emphasized by certain particles³⁶. These particles can sometimes develop into genuine inflections (e.g., -dhi in Greek and Sanskrit). The Lithuanian imperative formant -k may have arisen in a similar way.

What nuance of meaning did the particle -k impart to the imperative form? In J. Otrębski’s view³⁷, forms with -k may have expressed a softened command (which is why their future-tense meaning arose: a softened command does not require immediate execution). This view, however, does not seem well founded—if the old imperative expressed a strict command, it is unclear why it should have disappeared.

As already noted, some imperative forms disappeared in Lithuanian, and new forms began to be used in their place. The imperative of optative origin did not replace all the earlier imperative forms. Some of those forms resembled the second-person present indicative forms (cf. § 5). Apparently, the new particle was intended to strengthen the force of command and thereby more clearly distinguish some second-person imperative forms from second-person present indicative forms.

³³ If these are not calques, as may perhaps be the case in M. Daukša’s Catechism: Kuri mūmus duok’ Wieszpatie (Ktore nam day Panie) (in E. Zittig’s edition, 86²⁰).

³⁴ Cf. F. Kurschat, Grammatik..., p. 301 (§ 1160), 372 (§ 1369). Alongside the forms tegul pasilieket, teit, tegul eit, one may also say tegul tepasilieekt, tegul teit, tegul teina.

³⁵ Žr. A. Brückner, Litu-slavische Studien, I, Weimar, 1877, p. 164.

³⁶ E.g., Greek ἄγε, φέρε, Latin age agedum (strengthening the exhortation), Latin amabo sis (=si vis) (softening the exhortation); see K. Brugmann, Grundriss..., II 3, § 717.

³⁷ See J. Otrębski, Gramatyka..., III, p. 226 (§ 504).

159

In dialects, a connection can be observed between the imperative and the future indicative forms: the first (and third, cf. § 14) person plural of the imperative without -k may also be expressed by the future indicative form (cf. nešma ‘let us carry’ Lkv, pasidalìsma ‘let us share’ Skp); in some dialects, the first-person plural forms of the imperative with -k have developed on the model of the future indicative forms (see § 21)³⁸.

Traces of the use of two imperative tenses can also be observed in the Kretinga dialect³⁹. It may therefore be reasonable to suppose that for some time there were two imperative tenses in other speech varieties as well.

§ 18. In both dialects and old written texts, the second-person singular forms of the imperative most often end in the formant -k. Fuller forms are very rare. They may end in⁴⁰:

-i: šūki rare Gršk, Plv; padėki Sb, see BM I 104, gulkį, palaukig Zt;

-ie: šūkie Smn, Jrb, darýkie Lp, Ršk, Grv, atadúokie Dglš, Tvr, Švnč, nesėdėkie Lz;

-a: vėskia Knv, baika ‘stop’ Kp;

-ai: sakíkai Krtn⁴¹, tilėkai Ds, pa-gi-kinķkai VoL 490₃, pa-gi-žiūrėkai VoL 490₅, palaukai VoL 364₆,₇;₇, pasakýkai VoL 366₂₁.

In present-day dialects, reflexive second-person singular imperative forms most often end in -is, e.g.: neškis, juokis Ms, Vkš, Gerv, Jrb, KzR, Prn, sū.kis Rk, Sv, Tršk, Pnd, Ktk, Mlt, and so on. Reflexive forms with -ies are quite rare, e.g.: juókies rare Grš, Vlkv, nėškies Jrb, liáukies Dglš, Ršk, Tvr, or with -ias, e.g.: kiškias Vkš, Trk, sūkias Kal, nėškias Lkv, kėlkias Krž.

§ 19. It is difficult to say what the full particle was like that became the imperative formant -k. A. Schleicher and K. Brugmann (later, Ch. Stang and J. Otrębski made a similar conjecture) thought there had been a particle k⁴². V. Poržežinskis⁴³ argued that there had been two particles—ki and kie. F. Solmsen⁴⁴ thought it might have been ko or ka.

The origin of the formant -k is also explained in various ways. Nearly all linguists⁴⁵ connect this formant with the Slavic particle ka. In Russian, this particle is used with verbs

³⁸ In addition, one may suppose that the conjugation paradigm of the imperative with -k (eik(i), eik-i-me, eik-i-te...) developed on the model of the future-tense conjugation (eisi, eisime...), cf. § 19.

³⁹ See J. Aleksandravičius, Kretingos tarmės veiksmažodis, “Vilniaus Valstybinio pedagoginio instituto Mokslo darbai”, vol. 8, 1959, pp. 117 ff. (for some of these forms, see footnote 41).

⁴⁰ For examples found in old written texts, see A. Bezzenberger, Beiträge, p. 220 ff.; В. К. Поржезинский, Къ исторіи, p. 22 ff.

⁴¹ In the Kretinga dialect, these forms have the meaning of the present-tense imperative (see J. Aleksandravičius, cited work, pp. 117 ff.). Here -ai is probably a particle, also used with other verb forms, e.g.: reikiai ‘is necessary’, eitai ‘goes’, reiksai ‘will be necessary’, ėsaị ‘will be’ Krtn. This particle cannot be a remnant of the optative imperative, since it occurs in a-stem forms (as we have seen, and a-stem imperatives are not inclined to mix either).

⁴² A. Schleicher, Handbuch..., p. 230; K. Brugmann, Grundriss..., II, 3, p. 957, § 889; J. Otrebski, Gramatyka..., p. 226; Ch. S. Stang, Verbum..., p. 248.

⁴³ В. К. Поржезинский, Къ исторіи..., p. 34.

⁴⁴ F. Solmsen, Etymologien. I. Slav. ka, ko, griech. κᾶ, κέ(v), KZ, vol. 35, 1899, p. 463 ff. (hereafter: KZ, vol. 35).

⁴⁵ See the literature cited in notes 42–44; see also F. Miklosich, Vergleichende Grammatik der slavischen Sprachen, IV, Wien, 1867, p. 798.

160

imperatives and subjunctives, e.g., расскажи-ка, ложись-ко, поёшико, попейко, по-ёдему-ко, постоиму-ко⁴⁶.

The Lithuanian plural imperative endings are -kime, -kite (not *-mek(i), *-tek(i)). Similar infixation of the particle has been observed in the Yaroslavl dialect of Russian: in colloquial speech — полюкаме frequently (instead of полюме-ка) and in song выго-няйкаме⁴⁶.

As can be seen, the relationship between Lithuanian -k and Slavic -ka has been correctly observed. K. Brugmann further considers Lithuanian -k (as well as Old Indic kam, kám, Avestan kam, Greek χέν, χέ, χά, Doric κά and Old Slavic κъ, Russian ко, ка) as deriving from the Indo-European pronouns *quom, *quem, *quā, while V. Poržezinskis derives it from Indo-European pronouns with the stems⁴⁷*qui (> Lithuanian ki), *quā (> Lithuanian -kie).

A. Schleicher⁴⁸ compared the particle -k with p, po (e.g., taip, Old taip, kitaip, kaip, derived from the stems ta, kita, ka)⁴⁹. E. Fraenkel holds a similar view; he connects the imperative formant -k with the Lithuanian postposition pi⁵⁰.

V. Toporov and O. Trubačiov⁵¹ put forward an entirely new idea. In their view, the Lithuanian and Russian imperative with -k was formed on the model of the Finno-Ugric imperative, which has the formant -ka.

§ 20. To which forms did the particle -k attach? In A. Schleicher’s and K. Brugmann’s opinion, the particle was inserted into imperative forms of the type duodi, imi. However, duodi, imi are present-stem forms. For this reason, A. Schleicher thought that the particle was inserted into those forms in which the present-tense and infinitive stems coincided. Thus, alongside the old imperative gelbi (1st person singular), gelbite (2nd person singular), the new gelbki, gelbkite arose (later gelbėk, gelbėkite).

Ch. Stang (and J. Otrębski) believe that the basis for imperative forms with -k was such root-imperative forms as ei⁵². The particle -k apparently attached last to forms that ended in a consonant (p, š or the like). In individual dialects even today, the formant -k in the 2nd-person imperative is sometimes not pronounced if preceded by a consonant, e.g.: nepil(k), paiń(k), kirp(k) (but gėrk) Jrb (in the Jurbarkas dialect k can drop after any consonant and remains only after r); mė.s(k), lū.p(k), nė.š(k) Sy, at-nė.š(k) Plvn, atkimš(k) Kp, vėš šieną, nupėš vytą Dbk.

§ 21. In some Aukštaitian dialects, the 1st-person plural forms of the imperative with -k may have the ending -ium; cf. bėki'um rarely in Dkš, NŪt, Vlė, neš̌kium, neškiumės Prn, eikium Mrk, Prl, kėlki'um, kėlkiums, kėlkiumėms Slm, t. lėkium Plvn, duokium, diokiumus, prašksiumėms mostly in Pnm, vė.ški'um, vė.ški'um(e)s, vė.ški'umėms Sy⁵³.

⁴⁶ Cf. F. Solmsen, KZ, vol. 35, p. 463 ff.

⁴⁷ Other linguists also hold differing views; cf. H. Osthoff, Perfekt, p. 355; A. Bezzenberger, Beiträge..., 217 ff.; F. Solmsen, KZ, vol. 35, 463 ff.

⁴⁸ See A. Schleicher, Handbuch..., p. 231.

⁴⁹ Compare also Slavic pronouns: Russian мне-ка, тебе-ка (forms occurring in Russian bylinas), Serbian тебе-ка ‘to you’, naske ‘us’, vaske ‘you’, Bulgarian азика, азка ‘I’, менека ‘me’, тебека ‘you’.

⁵⁰ See E. Fraenkel, BS, II, p. 41 ff.

⁵¹ See В. Н. Топоров и О. Н. Трубачёв, Лингвистический анализ гидронимов Верхнего Поднепровья, Москва, 1962, p. 249—250.

⁵² This is particularly confirmed by the form aik, which in Daugeliškis and Tverečius expresses both the imperative present and the imperative future tense.

⁵³ See also išmuškium (and atadūokiem) in Sb, išmuškium, eikium; see F. Specht, Litauische Mundarten, gesammelt von A. Baranowski, II, Leipzig, 1922, p. 193.

11. Lietuvių kalbos morfologinė sandara ir jos raida

161

Such forms arose on the model of the first-person plural of the indicative future tense: in those dialects where the first-person plural of the imperative ends in -kium, the future-tense forms also have -siu⁵⁴. This once again confirms the connection between the imperative and indicative future-tense forms (cf. § 7).

§ 22. As we have seen, forms of the imperative without -k are used not only in the Eastern Dzūkian dialects; they are also occasionally found in Kretinga, Linkuva, Panevėžys, Švedasai, Aleksandrãvelė, and elsewhere.

The oldest surviving Lithuanian imperative forms are the so-called root imperative (e.g., duo, etc.). These forms are now found in former athematic and (i) a-stem verbs. During the common Balto-Slavic period, the optative form began to be used with imperative meaning. The imperative of o- and i-stem verbs used in Lithuanian dialects today is of optative origin. The optative imperative of (i)a-stem verbs has almost entirely disappeared. It is unclear when the contaminated imperative forms duod, šluoj, būn arose in Lithuanian.

The old third-person imperative forms may have been identical to the second-person singular imperative forms. To specify the third person of the imperative, the particles te, ta, lai began to be used with this form, and later tegu(l), tagu(l), testa(u), laigu(l), leigu(l). Today, the third person of the imperative is most often expressed by the third-person present indicative form with a particle.

The first-person plural imperative form is of later origin than the second-person singular and plural forms. The first-person plural can also be expressed by the first-person plural form of the present indicative.

Imperative forms with -k arose well before the appearance of the first written texts. A certain connection can be observed between the imperative with the formant -k and the indicative future-tense forms.

A B B R E V I A T I O N S⁵⁵

AfslPh — Archiv für slavische Philologie, Berlin (since 1876).

AK — Anonymous catechism of 1605.

BB — (with biblical text references) Biblia tatai esti Wissas Schwentas Raschtas Lietuwischkai pergulditas per Jana Bretkuna Lietuwos Pleboną Karaliaucziuie (1590).

BM — Lithuanian dialects, collected by A. Baranowski. Volume I: Texts from the Weber estate, edited by Dr. Franz Specht... Leipzig 1920.

BP — Postilla tatai esti Trumpas ir Prastas Ischguldimas Euangeliu... Per Jana Bretkuna... Karaliaucziuie... 1591, I.

BS — Balticoslavica, I—III, Vilnius (1933—1938).

Bt — (su biblijos teksto nurodymais) S. Bythner, Naujās Testamentas... Pirma karta ant Svieto Lietuwiskzoj kalboj... Su didžiu dabojimu pergulditas. Lietuvos žmonėms ant iszganytingos naudos Iszspāustas. Drukawõjo arba iszspaude Spāustuwõj Reusnėro Karaliaucziuj’, Métu MDCCI.

⁵⁴ The future-tense ending -sium most likely arose by analogy with the relationship aisi-aisit = aisiu: x=aisium; see F. Specht, KZ, vol. 55, p. 177.

⁵⁵ In addition to these abbreviations, the article also uses the abbreviations for place names introduced in the Dictionary of the Lithuanian Language; see Dictionary of the Lithuanian Language, vol. 5. Vilnius, 1959.

162

Daw Pam — A guide for those who wish to serve faithfully in the holy office of pastor in Lithuania. Königsberg, 1730.

dažn. — often, the most commonly used form.

DK — Lithuanian Catechism by N. Daukša, according to the 1595 edition, reprinted and furnished with explanations by E. Wolter. St. Petersburg, 1886.

DP — Catholic Postil, that is: An exposition of the Gospel for every Sunday and feast throughout the year. By the priest Mikalojus Daukša, canon of Medininkai, translated from Polish. With the permission and privilege of the authorities, in Vilnius, at the printing house of the Academy... 1599.

Dtn — Dotenėnai, Švenčionys district.

Gerv — Gerviniai, Raseiniai district.

IF — Indo-Germanic Research, Strassburg, Berlin and Leipzig (since 1892).

KlG — The First Grammar of the Lithuanian Language, Vilnius, 1957.

KN — (S. J. Telega). Book of Christian Devotion... reprinted for the second time in Kėdainiai... 1653.

Krd — Kirdeikiai, Utena district.

Kret — Kretuonys, Švenčionys district.

KZ — Journal of Comparative Linguistic Research in the Field of Indo-Germanic Languages, founded by A. Kuhn, Göttingen (since 1852).

LKK — Issues in Lithuanian Linguistics, Vilnius (since 1957).

Ltk — Latakiškė, Ignalina district.

Nūt — Naujoji Ūta, Prienai district.

PK — Polish-Lithuanian Catechism, or a Brief Collection in One Place of the Christian Faith and Duties... Published at the expense of His Grace Lord Málcher Pietkiewicz, land scribe of Vilnius. In Vilnius, printed by Stanisław Wierżeyski, in the year 1598.

Plvn — Palėvenė, Kupiškis District.

ret. — rare, a less frequently used form.

Ršk — Riešutėnai, Švenčionys District.

SD — Dictionarium trium lingvarum, In usum Studiosae Iuventutis, Avctore... Constantino Szyrwid... Quinta editio... Vilnae... MDCC.XIII.

SE — Summa Abā Trumpas iszguldimas Ewangeliw... Kiedaynise... 1653.

SP — Syrwids Punktay sakimu (Punkty kazań). Teil I: 1629. Teil II: 1644... herausgegeben von Dr. Franz Specht. Göttingen 1929.

St. balt. — Studi baltici, Roma, Instituto per L’Europa Orientale.

ukr. — Ukrainian.

Vln — Enchiridion, Catechismas maszas, dael paspalitu Plebónu ir Koznadiju... isch Wokischka lieszuwia ant Lietuwischka pilnai ir wier­nai pergulditas, per Baltramieju Wilentha Plebona Karalauczuje ant Schteindama. Ischspaustas Karalauczu per Jurgi Osterbergera, Metu... MDLXXXIX.

ON THE DEVELOPMENT OF IMPERATIVE FORMS

D. SKURKAITĖ

Summary

In some Lithuanian dialects and in old written sources, two types of imperative forms are used: with the formative -k and without it. The imperative without the formative -k is widespread in the dialects of the Eastern Dzūkai (Dūkštas, Ignalina, Švenčionėliai, Švenčionys, Tverečius); in other dialects, only isolated imperative forms occur—

11*

163

tive without -k: imti, neši (Dusetos, Svėdasai, Daugailiai), im (Šešikai), klaušai (Bridai, Linkuva, Panevėžys), imti, begi (Kretinga).

The oldest surviving imperative form is the imperative of the duo type. Forms of this type are now formed from verbs with athematic stems and stems ending in -(i)a.

Since the Common Baltic period, optative forms have entered the imperative system. At present, in Lithuanian dialects, the imperatives of verbs with stems ending in -o and -i are optative in origin.

The 3rd person imperative is now expressed by the singular indicative form with the particles te, ta, lai, lei, tegu(l), tagu(l), testa(u), laigu(l), leigu(l). In many dialects, other 3rd person imperative forms formed with the particles te, tegu(l), lai, laigu(l), testa, among others, and 2nd person imperative forms without the formative -k have also been preserved.

First-person plural imperative forms arose in Lithuanian later than the second-person singular and plural forms. Before they arose (and still today), first-person plural present indicative forms were used in the function of the first-person plural imperative.

The imperative with the formative -k arose long before the first written sources appeared. It can be stated with certainty that these formations are connected with future indicative forms.