Naujažodžiai Vytauto Didžiojo universiteto kompiuteriniame tekstyne

ACTA LINGUISTICA LITHUANICA LIETUVIŲ KALBOTYROS KLAUSIMAI XLII (2000) Jurgita MIKELIONIENĖ Vytautas Magnus University, Kaunas NEOLOGISMS IN THE COMPUTER CORPUS OF VYTAUTAS MAGNUS UNIVERSITY Search, selection, classification New words are constantly emerging in language. Words that enter a language in various ways (morphological word formation, borrowing of lexemes, abbreviations) and as a result of various changes—changes in social relations and advances in science, technology, and culture—are called neologisms; they are perceived as new and unfamiliar (Gaivenis, Keinys 1990: 132, Jakaitienė 1980: 81, Urbutis 1978: 27–29). They are not stable, and it is difficult to say which of them have non-linguistic (extralinguistic) reasons to remain and which will disappear along with the phenomena they name. The emergence of neologisms is a continuous, intensive process in language development that is difficult to generalize. Works examining this lexical stratum (Schippan 1992: 244) often raise such questions as: how long should a word be considered a neologism; is every coinage a neologism, and vice versa; how should individual authors’ occasional constructions be treated; should calques and semantic innovations be considered neologisms; what status should be assigned to new sememes of familiar lexemes? It is also worth considering whether, from the perspective of contemporary usage, words that have long been part of the passive vocabulary but have returned to active use can conditionally be regarded as new. Considerable attention is devoted to the analysis of neologisms worldwide. Linguists working on both major languages (such as English and French) and smaller languages (such as Danish) conduct research on the documentation, analysis, and evaluation of this lexical stratum. Since the major Western European languages have already been adapted to computer technologies, new vocabulary can be collected automatically from large computer corpora using various methods. In Lithuania, neologisms were not systematically and consistently studied for a long time, for understandable reasons: computerization of the Lithuanian language began only quite recently, so automated searches for and analysis of neologisms could not be carried out. An automated search for Lithuanian neologisms was conducted in a section of the corpus being compiled at the Centre of Computational Linguistics of Vytautas Magnus University. This section contains 30 million words and covers journalistic texts from various fields published in 1991–1996. 15 At the initial stage of the study1, after adapting V. Zinkevičius’s neologism recognition program (the general-purpose program does not recognize these words because they are not included in computer dictionaries), about 750,000 linguistic units were selected: unidentified single letters, abbreviations, and words. Although the program makes it possible to automatically extract words containing foreign-origin derivational elements and helps compile lists of the newest words, it became clear that a second stage of automated selection was needed to filter out material irrelevant to the analysis. Words with fewer than four letters (mostly various abbreviations) and all words beginning with a capital letter—that is, proper nouns—were excluded. The final result of the automated search for and selection of neologisms was 230,000 headwords or word forms. Thus, the computer eliminated more than two-thirds of the words. However, this was not an ideal solution. The third selection step concerned the complex system of Lithuanian inflection and word formation: not all inflected forms of neologisms were needed for the study (more precisely, for compiling the list). A single occurrence is sufficient. Unfortunately, the computer dictionaries of word roots included in the program supplied the full paradigm of inflected neologisms. For example, the nominative singular of the coinage aplinkosaugininkas occurred 4 times, the nominative plural 8 times, and the instrumental 14 times. It was impossible to automatically eliminate unnecessary word forms. In addition, it became clear that the list of neologisms contained quite a few words with proofreading errors. Here, too, the computer and its spell-checking programs were powerless, since they treated every word in the list and the adjacent indication of where that word occurred in the database as an error. The list was cleaned manually, and about 17,000 words that had become current, been coined, or been borrowed from other languages during the years of independence were retained for further analysis. For example: answering machine c:\fox25a\lkt\lkt\00\00\03\57\9495BR.TXT most memorable c:\fox25a\lkt\lkt\00\00\02\44\D28GYVEN.TXT separatism c:\fox25a\lkt\lkt\00\00\00\21\FIL-142.TXT printout maker c:\fox25a\lkt\lkt\00\00\04\25\ROLZU25.TXT convert-like c:\fox25a\lkt\lkt\00\00\02\46\D36TRISK.TXT imported (feminine plural) c:\fox25a\lkt\lkt\00\00\03\98\GR049D.TXT separator (accusative) c:\fox25a\lkt\lkt\00\00\05\00\6VARIK.TXT restorer (genitive plural) c:\fox25a\lkt\lkt\00\00\02\82\ACIU.TXT reflection c:\fox25a\lkt\lkt\00\00\04\17\LEON20.TXT 1 In order to determine the sources of neologisms, the reasons for their borrowing or creation, and their conformity with the norms of standard Lithuanian, it was decided to undertake a research project entitled “Analysis of Neologisms (Loanwords and Coinages) in Standard Lithuanian during the Period of Independence.” Recognizing the relevance of the topic and the likelihood of obtaining new results, the State Science and Studies Foundation supported this initiative. 16 These words are the principal object of the study presented here, and other problems that arose during the work are also related to them. Thus, to sum up the first stage—automatic word search and selection—it can be said that the main difficulty in such work stems from the incompatibility of the user environment and computer dictionaries, and from the fact that not all word forms can be automatically reduced to a single lemma. In the second stage of the study, the selected material was classified. It immediately became clear that the list contained more than just neologisms. In addition to new coinages and loanwords, several other groups of unrecognized words can be distinguished. First, there are words that have regained relevance over the past decade and moved from the passive to the active vocabulary, as well as words from old written texts (kumpelninkas, nuopilas, prekionė, rodastis), dialectal words (dzienelės, keimas), derivatives of proper nouns (klaipėdienis, from the newspaper title “Klaipėdos diena”; donelaitiana, plėckaitiada, neoozolizmas, makdonaldizacija, derived from surnames); there were also names for residents of countries, cities, and villages (dalasietis, neringiškis, kvebekiečiai). The list included quite a few names of ethnic groups and tribes not included in dictionaries (kurdai, tutsis, svao). There were also crude vulgarisms and children’s words (niam, popieliukas). Second, there were instances in which the program selected not words, but only certain unrecognized forms or variants of words included in dictionaries, for example, the archaic inessive forms indicating movement toward the interior—the illative (kūriniuose, vestuvėsna)—and the archaic inessive forms indicating movement toward the exterior—the allative (lietuvininkump). The most numerous were unrecognized singular illative forms used alternately with prepositional constructions and moving from spoken into written language (klasėn, kojon, katedron instead of į klasę, etc.). The many examples of this type of locative collected during the study challenge the view that the illative is already disappearing from usage and acquiring merely a particular stylistic value (Paulauskienė, Tarvydaitė 1986: 119). Third, the program failed to recognize some diminutives (poemėlė, prakartėlė, gėrimukas), pronominal forms (nepusiasvireji), and the like. Fourth, spelling, morphology, and word-formation variants of words included in dictionaries but not listed there are also quite common. A considerable number of compounds were recorded that either lack a linking vowel or, conversely, have one where it is not expected (not dvylikametis, but dvylikmetis; not mėlynakė, but mėlynaakė); gender may differ (not eglišakė, but eglišakis), as may number (not žudynės, but žudynė). All these cases require separate investigation and conclusions. Clearly, to determine the true significance of such variant phenomena, the full range of usage must be taken into account (Miliūnaitė 1995: 40). After separating out these groups (the separation of lexical items that have regained currency can be discussed only conditionally), the remaining words (about 9,000) were grouped according to the traditional classification of neologisms—into loanwords and new formations. Each group, in turn, consisted of several subgroups. New skolinius are customarily divided into international words and foreignisms. Both these lexical strata are traditionally classified by origin and degree of adaptation to the Lithuanian language. Until recently, linguists used to say that most new loanwords entered the language from Russian, since the greatest amount of 17 informacijos dalis (Jakaitienė 1980: 84), o dabar surinkti faktai aiškiai rodo, kad rusizmai ar apskritai slavizmai vyrauja tik senųjų svetimybių, kurioms neretai būdinga stilistinė ar vertinamoji konotacija, pogrupyje (skaurada, rubežius, cielas, pieška), o pirmauja iš anglų kalbos atėję žodžiai. Bearing in mind that the borrowing process itself is certainly not a new phenomenon and that an abundance of international words is one of the linguistic features of journalistic writing (Bitinienė 1995: 2), its intensity during the period under study is still striking. Since the computerized version of the Dictionary of International Words is included in the neologism-recognition program, following the hypothesis that a neologism is a word that has not (yet) made it into a dictionary (Busse 1996: 649), one might conclude that the international words automatically selected from the corpus are precisely that—neologisms. Unfortunately, for reasons of “economy,” the dictionary’s editors omitted a considerable number of derivatives, although, as reviewers noted (Gaivenis 1987a: 69, Palionis 1987: 68), they did not always consistently observe the established rules. Therefore, when looking at the list of words unrecognized by the computer, it remains unclear whether those derivatives were already in use in the language at the time or were, in fact, neologisms. For instance, the preface states (TŽŽ 1985: 3) that names of persons derived from fields of science containing the element -logy were not included in the dictionary. (It is unclear, then, why patologas, pedologas, etc., were included.) All that remains is to speculate and trust that endokrinologas, konchiliologas, kriminologas, politologas and the like, although not included in the dictionary, are words that have been in use for a long time. Clearly, the term neologism can be applied to them only with certain reservations. The situation is different with those internationalisms for which almost the entire derivational family is absent from the dictionary, for example: edukologija, edukologas, edukologinis; kultūrologija, kultūrologas, kultūrologinis; faktografija, faktografas, faktografinis, faktografiškas, and so on. Medicine, economics, theology, music, and sport are fields particularly rich in borrowings. Some have already been included in the Commission of the Lithuanian Language’s lists of inadmissible foreignisms and have Lithuanian equivalents—neologisms (distributor(i)us – platintojas, nuskenuoti – nuskaityti); suitable replacements have not yet been created for others (keisbolas, softbolas), while for a third group, already very widespread, there may be no point in creating them (jogurtas, singlas, hitas). Not all borrowings found in the press are equally adapted to Lithuanian. Some (not many) still do not conform to the phonetic and morphological rules of Lithuanian (disco [diskol, papaya [papaja], atache [atašė]), so we have examples of “the simplest (and worst) treatment of the newest foreignisms” (Girčienė 1997: 64). Slang examples are also classified as foreignisms (fuksauti, chebrantas, psichuš̌kė, pacanas, diskonas). Another group of neologisms consists of coinages. They are classified according to word-formation patterns. Examples of all four word-formation patterns are recorded among the coinages2: 2 None of the coinage examples presented appears in the comprehensive Dictionary of the Lithuanian Language. However, there are quite a few of them in the list under analysis. Since we do not have a computerized version of this dictionary, 18 a) prefixal derivatives: bepilotis, bevizis; b) suffixal derivatives: coinages were found not only with the suffixes -imas (muziejavimas), -ykla (kaupykla, pynykla), -ininkas (pilstukininkas, virgul ininkas), -ėlis (prasilošėlis), -tojas (šiukšliautojas), -iklis (kaupiklis, nešiklis, nuėmiklis, valdiklis), -ynas (pamestinukynas, svarainynas, kiniavarpynas), but also with suffixes that are not particularly productive (sliekidė); c) ending derivatives: įmitis, jautra, irsmas, nuodribos; d) compounds: coinages of this type are almost on a par with suffixal derivatives in quantity and variety (įvairiagrūdis, plonatinklis, savišaudis, laivagrybis). Some compounds encountered are terms: laižligė (veterinary), banglentė (sports), kraujasekys (zoology), etc. Words formed according to such typical patterns are called systematic neologisms (Barauskaitė et al. 1995: 43). Clearly, a considerable proportion of them should be regarded as potential neologisms—ones whose novelty we do not perceive. Examples of potential neologisms might also include compounds such as auksagalvis, basagalvis, dešragalvis, drambliagalvis, plienagalvis, žvynagalvis. Their abundance in the list of unrecognized words could be explained by the fact that “lexicographers dislike potential neologisms and try not to include them in dictionaries” (Gaivenis 1987b: 33). Not all coinages, however, are systematic. Examples of authorial, stylistic coinages were also found: džinsuma, plaukuosena, skiedrija, žaidynas. Some are immediately perceived as creative departures from prevailing lexical norms (Busse 1996: 647), while others require context to assess. Statistical data also help distinguish systematic coinages from occasional ones3. Words occurring dozens or even hundreds of times are genuine candidates for inclusion in future dictionary editions. Examples include: aplinkosauga (used 420 times), pasaulėjoka (44), ploviklis (20), pilstukas (176), vandenvála (88). Words recorded only once, on the other hand, are probably unlikely to become entries in a general-language dictionary. This applies not only to individual coinages, but also to technical terms in narrowly used specialist vocabulary (ezofagogastroduodenoskopinis, alfafelandrenas). Words with a Lithuanian or international-word root and an element of foreign origin, which some also call a combining form (avia-, auto-, mikro-, tele-, termo-, etc.), occupy a somewhat intermediate position between borrowings and coinages. At present this is a very popular method of word formation. The following are most often cited as reasons for the free spread of neologisms of this type: a pragmatic attitude, an undefined place in the Lithuanian language the assumption is therefore adopted that words that more or less conform to word-formation patterns but are absent from the normative computerized Dictionary of Contemporary Lithuanian are neologisms, although some of them may be derivatives from an earlier period that have recently regained currency. 3 In addition to systematic and occasional neologisms, a semantic type is also distinguished—a new meaning of a familiar word. Since the word is included in the computerized dictionary, the automatic neologism-search program does not detect it. 19 in the system, cases of usage that have not been discussed, and an exaggerated emphasis on spelling issues (Kupčinskaitė 1997: 58). The available data clearly show that they are spreading. Previously, only a few words with the component video- were recorded (videomagnetofonas, videokasetė, videokamera), whereas now we also find videoklastotę, videomadas, videomeną, videopoeziją, and so on, although this element is often strongly recommended to be replaced by the Lithuanian equivalent vaizdo. In many instances involving already familiar components (hidrokostiumas, mononatris, multivitaminai, antiteroras, stereovaizdas, pseudomitinis), their number is almost matched by those whose components are not yet recorded in dictionaries (ekspresolimpiada, technobiurokratija, ritmenbliuzas, vertebroneurologija, ekspolinija, popkaralienė). At present, a particularly frequent component is euro-. These are not only euroentuziastas or euroskeptikas, eurolyga, eurorinka, but also europudingas, eurotenisas, euroįvartis, eurofakultetas, and so on. The emergence of neologisms is a dynamic and continuous process. It is therefore clear that neologisms must be recorded and studied continuously. Language innovations selected for analysis should be monitored until, having become part of the general lexicon and general linguistic awareness, they lose their salience and the distinguishing mark “new” and are transferred (or only some of them are) to dictionaries (Busse 1996: 654). In Lithuanian linguistics, we can speak only of the very beginnings of neologism lexicography. This study clearly demonstrates that when large amounts of data need to be processed automatically, computer science can successfully assist linguists. At the same time, the research methodology is fundamentally changing: in the future, automated selection of neologisms will be much faster, because after a word-tagging system was applied to the Lithuanian language corpus, the need to search for words in computerized dictionaries was eliminated. Of course, the analysis of neologisms is not limited to their recognition and classification. The next stage is the study of syntactic combinations and collocations (a particular lexical environment) involving these words. Received 1998 12 16 LITERATŪRA Barauskaitė J. ir kt. 1995: Lietuvių kalba 1, Vilnius: Mokslo ir enciklopedijų leidykla. Bitinienė A. 1995: Publicistinio stiliaus ypatybės. – Gimtasis žodis 12(84), 1–4. Busse U. 1996: Neologismen. Der Versuch einer Begriffsbestimmung. – Euralex'96 Proceedings II: Göteborg University, Department of Swedish, 645–658. Gaivenis K. 1987a: Keli žodžiai apie naująjį „Tarptautinių žodžių žodyną“. – Kalbos kultūra 52, 66–70. Gaivenis K. 1987b: Margas žodžių pasaulis, Vilnius: Mokslas. Gaivenis K., Keinys S. 1990: Kalbotyros terminų žodynas, Kaunas: Šviesa. Girčienė J. 1997: Naujosios svetimybės spaudoje. – Kalbos kultūra 69, 64–68. Jakaitienė E. 1980: Lietuvių kalbos leksikologija, Vilnius: Mokslas. Kupčinskaitė A. 1997: Nelietuviškų kalbos elementų (dėmenų) vartojimo polinkiai. – Kalbos kultūra 69, 56–61. 20 Miliūnaitė R. 1995: Variantiški bendrinės kalbos vartosenos reiškiniai (fiksavimas ir vertinimas). – Kalbos kultūra 67, 39–48. Palionis J. 1987: Naujasis „Tarptautinių žodžių žodynas“. – Kultūros barai 7, 67–69. Paulauskienė A., Tarvydaitė D. 1986: Gramatikos normos ir dabartinė vartosena, Kaunas: Šviesa. Schippan T. 1992: Lexikologie der deutschen Gegenwartssprache, Tübingen: Niemeyer. TŽŽ 1985: Tarptautinių žodžių žodynas, Vilnius: Vyriausioji enciklopedijų redakcija. Urbutis V. 1978: Žodžių darybos teorija, Vilnius: Mokslas. NEOLOGISMS IN THE COMPUTERISED CORPUS OF VYTAUTAS MAGNUS UNIVERSITY UNIVERSITY Summary The paper deals with the analysis and classification of the most recent Lithuanian neologisms that have been found in a part of Vytautas Magnus University corpus of 30 million words (periodicals). All the lexical items identified as neologisms are subdivided into loan-words and derivational neologisms. The latter are further classified according to the patterns of their affixation into compounds and affixation derivatives as well as into potential and occasional neologisms. 21