Lexicography and diachronic derivational morphology

Until now, Lithuanian and Latvian lexicography lacks a coherent methodology.17 No operative principles have been formulated of what is to count as a lexical unit, nor is there any theoretical foundation of the way lexical units interact with grammatical distinctions. There are no theoretical guidelines concerning a share between lexicon and grammar. Consequently, the vast “gray zones” between lexicon and grammar have hardly been reflected upon, let alone integrated into lexicographic work. An analogous problem concerns a differentiation between lexicographic accounts of the standard language vs. dialects (or other non­

standard varieties); as concerns Lithuanian, cf. the discussion of Kardelis and Wiemer (2003: 47–54, 66–68).

Work on the largest Lithuanian dictionary (LKŽ, 20 volumes, Internet version at http://lkz.lt/; henceforth LKŽe) started before World War II, the last volume was issued in 2002.18 The biggest problem with this dictionary is not that its first volumes had become obsolete by the time the last ones appeared, but that there have never been any clear principles of selection and description. As a conse­

quence, one can find promiscuously various dialect data reaching back to the nineteenth century, even without any qualification. Murmulaitytė (2000) cri­

ticized LKŽ’s practice of listing nominal derivatives in the entry of the deriving verbs (verb stems). The lexicographic practice does not satisfactorily distinguish between regular and more idiomatic (less predictable) items. As concerns speech act verbs, Zaikauskas (2006) reports that in LKŽe, their semantics was described incoherently.

A motivated argument concerning the lexicographic treatment of motion verbs (primary usage vs. figurative use in which they are stative) has been given by Mikulskas (2006a). In a sense, the mirror image to verbs, i.e., the lexicographic treatment of so­called verbal particles in northern and western Lithuanian dia­

lects was analyzed by Mikulskas (2003) (see also Section 4).

The existence of the frequency dictionary of Lithuanian based on a 1 million token annotated corpus should be mentioned here as well (cf. Utka 2009, available online at http://donelaitis.vdu.lt/publikacijos/).

17 For an overview of current standards in lexicography oriented toward Lithuanian, see the handbook Leksikografija by Jakaitienė (2005).

18 Work, headed by Juozas Balčikonis, started in 1930. The first volume appeared in 1941, the second in 1947. Then the work on the dictionary was held up by Soviet authorities, the editorial board changed, and the third volume of the LKŽ, based on principles of officially accepted Soviet lexicography, appeared only in 1956. Later, in 1968−1969, the already published first two volumes were considered to reflect “bourgeois­nationalistic” ideology and re­edited on “new” principles.

Traditionally, lexicographers have been devoting much attention to etymo­

logy and the diachrony of word semantics as well as of the system of derivational affixes, many of which became unproductive a long time ago. As for derivational morphology, Saulius Ambrazas (1993, 2000a) presented onomasiologically orien­

ted monographs on the diachronic formation of derivational categories of nouns with verbal or nominal origin, respectively.19 As a partial diachronic equivalent on the side of the verb lexicon and the involved derivational suffixes of Lithu­

anian, one may regard Kaukienė (1994, 2002). This approach has been applied more broadly to the entire Baltic area by Kaukienė and Jakulis (2009). Ostrowski’s (2006) selection of studies focuses on the diachrony of aspectually relevant suf­

fixation and denominal verbs in Lithuanian. Larsson (2002) deals with nominal compounds from a diachronic perspective (with an Indo­European background).

Fraenkel’s etymological dictionary (Fraenkel 1955–1965) is quite well known, but one has to have in mind that Fraenkel was not able to account for a great many of important lexical items, because when he was writing the dictionary only the first few volumes of the LKŽ had been issued (Sabaliauskas 1990: 5).

The recent etymological dictionary by Smoczyński (2007, an expanded English version, to be published by Peter Lang, is under preparation) comprises a smaller amount of lexemes than Fraenkel’s, but the selection is based on the entire LKŽ (since 2000), and the author deliberately included borrowings.

To our knowledge, apart from work on derivational morphology and Mikul skas (2002a,b) (see above), no onomasiologically oriented studies of lexical fields have been undertaken, although one occasionally finds discussions of word meanings arranged by onomasiological fields scattered over the lexicon (see, for instance, in Gudavičius 1985). Furthermore, Sabaliauskas (1990) subdivided his annotated dictionary into lexical groups that correspond to periods beginning with common Indo­European heritage and ending with layers restricted to Lithuanian. The last third of his book is dedicated to different layers of borrowed lexemes, among which Slavicisms occupy the most prominent place. Despite this fact, a coherent methodology for the lexicographic treatment of Slavicisms, in general, and for the differentiation of different Slavic languages as particular sources still waits for its master (cf. Kardelis & Wiemer 2003: 46–54). Kardelis (2003) gives a survey of the

19 The sections of these books are organized according to notional types (e.g., nomina actoris, resultati, instrumenti, actionis for deverbal nouns, collective, diminutives, etc., for denominal nouns). S. Ambrazas (2001) deals with the provenance of certain Lithuanian adjectives derived from numerals. S. Ambrazas (2000b) discusses the most striking differences in the derivation of nouns between Lithuanian and Latvian. First of all, they concern nomina actionis, diminutives, and collective nouns.

problems connected to establishing the concrete source language of Slavicisms and applies a principled method to the chronology and phonological integration of Slavic loans into Lithuanian.

Admittedly, the problem of identifying the specific Slavic source language is partly rooted in objective difficulties, and it is even aggravated by the fact that often one can hardly discern between borrowings from Slavic, on the one hand, and root morphemes and derivational affixes from a common Slavic­Baltic stock, on the other hand. This issue proves particularly problematic in the lexicon and morpheme layers of Lithuanian dialects whose speakers have for centuries been in intense contact with speakers of (East) Slavic (cf. Wiemer 2009b: 358–385 for a comprehensive investigation; see Section 4 for further discussion of contact phenomena).

Modern Latvian lexicography starts with the dictionary commonly referred to as ‘Mühlenbach­Endzelin’ 1923–1932 (with two supplement volumes: Endzelin and Hausenberg 1934–1946), a dictionary covering the nineteenth­ and early twentieth­century written language, the dialects, and the language of oral folklore, started by Karl Mühlenbach and, after his death, completed and provi­

ded with brief etymological notes by Jānis Endzelīns. It is still the only dictionary of any use for historical linguists as it marks syllable accents, a tradition since abandoned in Soviet Latvian dictionaries.

The Soviet period saw the compilation of a comprehensive dictionary of the modern Latvian literary language (LLVV). Although obviously valuable as the main lexicographical source on modern written Latvian, especially the language of the post­war period, it has several drawbacks: its normative character leads to exclusion of large parts of the lexicon, such as loanwords considered undesirable, much of the colloquial vocabulary etc.; there is no phonetic or prosodic information.

Latvian historical lexicography is still in its childhood, but the compilation of a corpus of Old Latvian texts from the sixteenth and seventeenth centuries (LVSTK) is to be the foundation of an Old Latvian dictionary, work on which started in 2004.

Several dialect dictionaries have been compiled in Lithuania since Vitkaus­

kas (1976) as well as in Latvia during and after the Soviet period. The most useful among them is probably that of the High Latvian dialect of Kalupe, compiled by Antoņina Reķēna (1998), as it gives an image of the lexical stock and also (through its illustrative material) of the morphosyntax and syntax of one of the dialects of Latgalian, for which but few descriptions are available at this moment (see Section 1.2). As long as no comprehensive Latgalian dictionary is available (see, however, Bukšs 1969, Bērzkalns 2007), Reķēna’s work will remain the principal gap filler.

In addition to the etymological notes in Mühlenbach and Endzelin’s dictionary, there is a separate etymological dictionary of Latvian by Konstantīns Karulis (1992).

It does not quite meet modern standards, being based on the Neogrammarian para­

digm of Indo­European reconstruction and offering mostly root etymologies, but it certainly is a valuable work, with lots of useful information on word history, and it offers a synthesis of earlier research as well as an overview of the relevant literature.