ARTICLES

 

Grammatical Data in the Dictionary of Montenegrin National and Literary Language

 

Grammatikale data in die Dictionary of Montenegrin National and Literary Language

 

 

Sonja Nenezić

Faculty of Philology, University of Montenegro, Nikšić, Montenegro (sonjan@ucg.ac.me)

 

 

---

ABSTRACT

Dictionaries are generally consulted to ascertain the meaning of a word. However, the meaning is inseparable from its grammatical features, which often determine it. Therefore, this article examines the type, scope, and method of presenting grammatical data in a comprehensive general dictionary. The aim is to analyze and ascertain the morphological and syntactic characteristics of all types of words recorded in what is currently the only such dictionary of the newly standardized Montenegrin language. Attention is also given to the applied metalanguage, representing a combination of transparent abbreviations and natural language. The initial hypothesis about the heavy reliance on the inherited Serbo-Croatian lexicographic practice is confirmed, but certain deviations from this tradition are also noted, which aligns with the dictionary's goal of presenting grammatical data more accessibly and comprehensively to its target users.

Keywords: grammatical data, grammatical marker, general descriptive dictionary, dictionary of montenegrin national and literary language, serbo-croatian language dictionaries

---

OPSOMMING

Woordeboeke word gewoonlik geraadpleeg om die betekenis van 'n woord te bepaal. Die betekenis kan egter nie van die grammatikale kenmerke, wat dikwels die betekenis daarvan bepaal, geskei word nie. Daarom word die tipe, bestek, en metode van aanbieding van grammatikale data in 'n omvattende algemene woordeboek in hierdie artikel bestudeer. Dit het die analisering en bepaling van die morfologiese en sintaktiese eienskappe van al die tipes woorde wat tans in die enigste sodanige woordeboek van die nuut gestandaardiseerde Montenegrynse taal opgeneem is, ten doel. Aandag word ook geskenk aan die toegepaste metataal, wat 'n kombinasie van deursigtige afkortings en natuurlike taal verteenwoordig. Die aanvanklike hipotese dat daar sterk gesteun word op die oorgeërfde Serwo-Kroatiese leksikografiese praktyke word bevestig, maar sekere afwykings van hierdie tradisie word ook waargeneem, wat ooreenstem met die doel van die woordeboek om grammatikale data meer toeganklik en omvattend vir sy doeltaalgebruikers aan te bied.

Sleutelwoorde: grammatikale data, grammatikale merker, algemeen beskrywende woordeboek, dictionary of montenegrin national and literary language, serwo-kroatiese taalwoordeboeke

---

 

 

1. Introduction

It is well acknowledged that in dictionary-making, lexicographers primarily focus on the meanings of lexical units, but the grammatical features inherent to them are equally crucial for their usage. Proponents of integral linguistic theories, with Apresjan (2010) being one of the most notable, advocate for the integration of dictionaries and grammar towards a unified linguistic description, emphasizing the need for their mutual consistency in terms of the data included and the method of its recording. Achieving this requires the collaboration of their authors, applying the same theoretical approaches and principles of "identification, classification, and interpretation of linguistic units," which is quite rare (Topolinjska 2002: 33). Typically, a dictionary and a grammar, as two fundamental works describing a language, where the former lists lexical items and the latter prescribes the rules for their combination, are produced independently of each other. Moreover, there is an issue where grammar and other linguistic disciplines do not provide suitable solutions for dictionary compilation, and conversely, the extent to which lexicography is receptive to the existing linguistic literature is discussed (Tafra 2005: 167). However, one thing is certain: linguists, even those not advocating for specific integral models, recognize the necessity and importance of grammatical information in the lexicographic processing of lexemes (Gortan-Premk 1980, Katičić 1994).

The central place in a dictionary is indeed reserved for lexical data, as it is most commonly consulted to verify the meaning of words (Engelberg and Lemnitzer in Kostić-Tomović 2017: 21). The question arises as to how much grammar should be included and by what principles, to ensure that the lexicographical description is as precise and purposeful as possible. It is essential to remember that although a dictionary is complementary to grammar, it is still a separate entity, distinct from grammar, hence it should contain a certain amount of grammatical data, without which it would otherwise be unusable or very difficult to use (Kačić 1994: 302). This necessary minimum must be primarily determined for any large monolingual descriptive dictionary, especially one intended for a diverse range of users such as native speakers, foreign users who have a (greater or lesser) command of the language, and experts - e.g. lexicographers using it as a basis for developing other types of dictionaries (Kostić-Golubičić 1997: 458). Therefore, our aim in this article is to analyze the interrelationship between grammar and the dictionary, exemplified by the only such lexicographic work of the newly standardized Montenegrin language to date. The lexicographic work in question is the Dictionary of Montenegrin National and Literary Language (DMNLL) published by the Montenegrin Academy of Sciences and Arts, in which the lexemes, as the subject of description, are presented "in the entirety of its grammatical forms and meanings" (DMNLL 2016: IX).

For more political than linguistic reasons (see Šubarić and Đurčević 2023), work on this fundamentally important lexicographical work was suspended immediately after the release of the first volume. The long-prepared second volume has therefore not yet been published. This, of course, causes delays in establishing lexical norms for Montenegrin (Šubarić and Đurčević 2023: 72), and there is uncertainty on when and how this issue will be resolved. However, since it is currently the only general dictionary of the Montenegrin language, it is still in use, and linguists, despite its incompleteness, utilize it in their research, especially since it will undoubtedly serve as a basis for many different academic lexicographical endeavors. We believe that it is necessary to promptly address the identified and other potential linguistic shortcomings, to which no lexicographical work is immune, finalise the second edition, and continue working on this crucial project for Montenegro, which is why we have carried out this research. The evaluation of the grammar presented in the DMNLL will take into account the type and amount of grammar as well as how the grammatical data is presented. The impact of grammar on the meaning of the headword will also be considered.

Montenegrin¹ was standardized after Serbian, Croatian, and Bosnian, following the dissolution of Serbo-Croatian as the common language of Serbs, Croats, Bosniaks, and Montenegrins, triggered by the disintegration of the Yugoslav state community. It is presumed that the authors of the DMNLL heavily relied on inherited Serbo-Croatian lexicographical practices. Therefore when indicating the lexicographic treatment of lexemes in the DMNLL, reference will occasionally (when it is necessary to highlight differences) be made to the lexicographical treatment of grammatical data in the most relevant multi-volume dictionaries of Serbo-Croatian: The Dictionary of Serbo-Croatian Literary and National Language² (1959−) (DSCLNL) and the Dictionary of Serbo-Croatian Literary Language³ (1967−1976) (DSCLL). However, considering the temporal distance between DSCLL and the early volumes of DSCLNL, on the one hand, and DMNLL, on the other hand, it is expected that the DMNLL is somewhat more contemporary and purposeful.

Some brief remarks on the DMNLL and a general discussion on grammatical data as a part of lexicographical data and its representation in a general monolingual dictionary follow. Subsequently, the morphological and syntactic features which are included in the lexicographical description of headwords in the DMNLL are investigated. Additionally, we will examine whether there are deviations in the presentation of grammar from the inherited Serbo-Croatian lexicographical practice, and what these deviations are. We will highlight any shortcomings and omissions that should be addressed in the continued work on the development of the Montenegrin dictionary. The most significant findings of our research will be succinctly presented in the conclusion.

 

2. Dictionary of Montenegrin National and Literary Language - general remarks

The basic information about the corpus of our research can be found in its preface (DMNLL: IX-XII). In 2011, the Montenegrin Academy of Sciences and Arts established the Council for the Compilation of the Montenegrin Language Dictionary, which included experts from various fields. Aiming to encompass the diverse lexicon of the Montenegrin area, the Council opted for the title Dictionary of Montenegrin National and Literary Language, which is also rooted in lexicographical tradition (cf. Dictionary of Serbo-Croatian Literary and National Language of the Serbian Academy of Sciences and Arts). This dictionary represents a collective endeavor in all phases of its creation: from the selection of sources, the excerpting of lexical material, to its processing. The first volume of the DMNLL, printed on the tenth anniversary of the restoration of Montenegrin statehood, contains 12,018 words listed in alphabetical order, starting with the Cyrillic letters A, B, and V. It was initially planned for this dictionary to have around 100,000 entries. It is conceived as a general descriptive dictionary, hence the lexicon it covers includes, in addition to the commonly used lexicon of the literary language, dialectal, terminological, onomastic, and all lexicon that is temporally and expressively marked with any semantic or morphological peculiarity. Pertinent to our study is the emphasis that it reflects current literary language norms at all levels of language structure - orthographic, orthoepic, grammatical and lexical, but at the same time, it can serve as a supplement to it, contributing to the resolution of dilemmas present in prosody, orthography, phonology, morphology, word formation, and syntax. Arguably the dictionary is therefore not only descriptive, but also prescriptive⁴.

 

3. Grammatical data in a general monolingual dictionary

The amount of grammatical data to be included in a dictionary and the lexicographic solutions to be applied depends largely on the type of dictionary and the target audience it is intended for, the characteristics of the language whose lexicon is being presented, and the lexicographic tradition (cf. Atkins and Rundell 2008: 399, Zgusta 1991: 115-116). In some dictionaries, typically bilingual ones, where the focus in processing the headwords is on equivalents, grammatical overviews of one or both languages are often provided among its supplementary sections, although the grammatical data do not always need to be so comprehensive and can be limited to specific key categories, such as tables of irregular verbs (Kostić-Tomović 2017: 45-46). However, in general monolingual dictionaries, which aim for comprehensiveness, much more data, not only about grammatical, but also about semantic, pragmatic, and other properties of the headword can be found within the dictionary entry itself.

Grammatical data is often presented or contained in the comment on form, comprising the morphological and syntactic features of the headword and having a normative character. Morphological features pertain to parts of speech, their grammatical categories, inflection, and word formation. Syntactic features, on the other hand, include syntagmatic relations, primarily valency, especially of verbs, but also of other morphological classes, the functions of individual words, etc. It should be noted that the system of parts of speech and their categories is equally important for syntax, hence it is also referred to as the morphosyntactic system. It is a fact that traditional Serbo-Croatian lexicography devotes more space to morphological than to syntactic data, as highlighted by one of the more significant Yugoslav and Serbian syntacticians, Popović (2003: 204), who emphasizes that it deals with words, not the syntactic units formed from them, and does not pay enough attention to the syntactic aspects of lexeme usage. However, it is clear that some types of words, such as conjunctions and prepositions, can only be defined by their function. The greater prevalence of morphological data in the dictionaries of many Slavic languages is influenced by their intricate inflection, within whose paradigms various accentual and morphological deviations from the canonical form of headwords occur. In English, for instance, inflection is less developed, and its paradigms are generally quite predictable, hence they are usually seldom displayed lexicographically. This is why, for example, Atkins and Rundell (2008: 218-221) do not mention it in The Oxford Guide to Practical Lexicography, where they highlight three types of grammatical data: (1) word class, (2) constructions or syntactic relationships, specifically of nouns, verbs, adjectives, and adverbs, as these are the four main word types, and possibly (3) data that directly depends on the headword's class, such as countability for nouns, indication of action or state for verbs, attributive or predicative function for adjectives, etc.

In addition to other elements, grammatical data contributes to the value of a dictionary and should meet some basic requirements: it should be unambiguous, complete, and economical (Gruszczyński in Kostić-Golubičić 1997: 458). It could be said that the grammatical description of the lexicon illustrates the relation between macrogrammar and microgrammar. This is especially reflected in the presentation of exceptions because grammar is, in principle, "more oriented towards rules, the system, analogy [...], while the dictionary records individuality, anomalies" (Tafra 2005: 68). A user will consider a dictionary the best if it includes everything they are looking for. Besides the basic forms, users are often, for example, interested in other inflected forms, grammatical peculiarities of a word, etc. Since every dictionary has its grammatical scheme, it should also contain, usually in the introductory section, some kind of guide through it, which would include grammatical markings and explain how they are used (Atkins and Rundell 2008: 218).

When discussing indicators of grammatical data, it is important to remember that establishing a metalanguage is one of the major lexicographic challenges. Lexicographers from different countries theoretically consider and practically resolve the issue of creating a universal metalanguage by using semantic primitives and the symbolic means of their denotation. They aim to construct interconnected interpretations of words from different languages. However, in domestic lexicography, preference is given to the natural language for accurate and consistent dictionary interpretations (Kozyrev and Černjak 2015: 74). Therefore, lexicographic data, including grammatical ones, can be encoded and decoded. Most printed dictionaries use abbreviations, while in electronic dictionaries, grammatical terms are usually given in full (Atkins and Rundell 2008: 218). Codes can be transparent, opaque with multiple keys, and opaque with a single key (Kostić-Tomović 2017: 47). Transparent ones are based on abbreviations and symbols that the target user group is already familiar with, having encountered them during their education, such as abbreviations for cases, parts of speech, etc. Codes whose meaning we cannot infer from prior knowledge or from the context are considered opaque. They are not desirable for the user even when used repeatedly with the same meaning, let alone singly, when they must be deciphered anew each time. Often, we learn about the grammatical properties of a headword indirectly, through illustrative examples. One example can convey more than ten symbols that the user does not understand (De Caluwe and Van Santen 2003: 82).

The grammatical markers themselves can be direct or indirect indicators of a grammatical category (Gortan-Premk 1980: 108-109). For instance, in traditional Slavic dictionaries, the direct marking of masculine, feminine, or neuter gender indirectly indicates the association of the headword with nouns. Although such an approach ensures the economy of grammatical data, there is an increasing demand for it to be explicit nowadays. This is the case, for example, in electronic English dictionaries (Atkins and Rundell 2008: 219), as well as in Apresjan's Active Dictionary of the Russian Language (2010). A dictionary is expected to play a more active role, and the lexicographical data provided should be as comprehensive and direct as possible, thereby becoming more accessible and useful to the user. Finally, it is worth noting that various typographical and nontypographical structural markers play a significant role in presenting grammatical data: the sequence of markers, different types of fonts, the use of various kinds of brackets, etc.

 

4. Grammatical data in the Dictionary of Montenegrin National and Literary Language

The creation of the DMNLL is based on the Handbook for Processing Entries in the Dictionary of Montenegrin National and Literary Language, which establishes a unique practice for the lexicographic processing of the selected lexemes. This handbook provides exemplary descriptions for all types of words with a complete physiognomy of the dictionary entry, which includes the way of presenting the headword, its pronunciation form, the sequence of grammatical data, etymology, syntagmatic and phraseological expressions, the method of determining meanings, the choice of appropriate examples, and the listing of sources. The mentioned Handbook, as well as the dictionary's macrostructure, which includes the Preface, Tabular Overview of Lexicographical Processing (XIII-XVII), Instructions for Using the Dictionary (XIX-XXVII), Abbreviations (XXIX-XXXII), and Symbols and Punctuation Marks (XXXIII), provide a good insight into the lexicographical solutions related to the scope and manner of presenting grammatical data in the DMNLL, although some of them were modified and refined during the creation process. The use of abbreviations and symbols, especially pronounced in the grammatical processing of headwords, indicates a consideration for the economy of lexicographical description, as it pertains to a print edition. It should be noted that these are transparent codes, mostly traditionally established, and there is also information that is not coded. The grammatical block in the dictionary entry comes after the possible pronunciation form, which is provided along with the standard accented form of the headword, and etymological data, but grammatical data, especially if related to particular meanings of headwords, also appears in the section reserved for the definition (see the examples in Appendix A).

It is entirely logical that from a grammatical perspective, different types of words are not processed in the same way; that is, the grammatical data is conditioned by the type of word. If it belongs to the variable word classes, the headword, as the first segment of the dictionary entry, is presented in its basic, canonical form, which represents the entire paradigm (Zgusta 1991: 115). For nominal words (nouns, adjectives, pronouns, and numerals) in Montenegrin, as in many other languages, this is the nominative singular, less often plural, and for verbs, the infinitive. Invariable words: prepositions, conjunctions, interjections, and particles, appear only in one form, and therefore they are accompanied by a marker of the morphological class to which they belong. The same lexicographic treatment is applied to adverbs, although some of them, mainly adverbs of manner, have comparative forms. This solution ensures their differentiation from neuter gender adjectives, with which many adverbs coincide in form (e.g., brzo (fast) as both adverb and neuter gender adjective), but also from other word types that derive from them (e.g., više (more) as both an adverb and a preposition). Among the variable word types, nouns, adjectives, numerals, and verbs are presented without a type marker, which is indicated through grammatical categories or definitions. Pronouns are an exception, presumably primarily because of the adjectival pronouns, which users commonly confuse with adjectives, and also because they are classified similarly in English grammar. Generally, the same practice is traditionally applied in Serbo-Croatian and in Slavic dictionaries at large, whereas in English dictionaries, each word type is assigned its qualifier since verbs and nouns often share the same form, and adjectives do not acknowledge gender categories, thus eliminating any ambiguities that might arise for the dictionary user (Marković 2014: 71-72).

In the subsequent sections, we will analyze the representation of grammatical features of all word types in the DMNLL, except for numerals which are not included in the first volume, whilst also highlighting certain deviations from the DSCLL and DSCLNL.

4.1 Nouns

According to traditional lexicographic practice, nouns do not carry a mark of the class they belong to. Instead, they are equipped with data on grammatical categories such as gender, number, and case, which are inherent to them (Čirgić et al. 2010: 68). However, it is important to note that the data about the word type is part of the typical grammatical definition for verbal nouns, for example, abdiciranje (abdicating),⁵ a verbal noun derived from the verb "abdicirati" (to abdicate).

Nouns have the most comprehensive grammatical data regarding the category of gender. Alongside all of them, the gender is initially indicated by an abbreviation: "m." (masculine), "ž." (feminine - henceforth f.), or "s." (neuter - henceforth n.). In the DSCLNL and DSCLL, however, the singular genitive form is provided before the gender, if necessary, without a case marker, while the other cases, introduced by an abbreviation, are listed after the gender mark and in round brackets. Since grammatical gender, unlike with adjectival words, is a classification category for nouns, its markers indirectly suggest that their bearers belong to the noun class of words. Therefore, this sequence in the DMNLL seems to be a better solution.

Two grammatical gender markers appear alongside pairs of headwords that differ in their endings, for example, brigàntīn and brigantína m. and f. (a light sailing ship with two masts), but each variant corresponds to one gender and paradigm. There are also rarer cases where one form has two genders without a difference in paradigm and meaning, for instance, bijenále (biennial) n. and m., and also with a noted difference, e.g., bôl m. and f. 1. (usually in masculine) the feeling of physical pain due to injury or illness 2. (usually in feminine) the feeling of mental pain; sorrow, sadness. However, the markers m. and f. refer not only to grammatical but also to natural gender or sex. This dictionary, unlike the DSCLNL and DSCLL, attempts to separate and mark the grammatical and natural gender of nouns where they do not coincide. Without delving into the complex issues of distinguishing gender and sex here, although it should be noted that here grammatical gender is treated as a morphological, not a syntactic category, we have observed several types of nouns that receive two gender markers, specifically when: (a) forms of the masculine gender denote people of both male and female sex (nomina professionalis), e.g., akadèmik (academician) m. (+f.), where the female form akàdemkinja is also recorded; (b) forms of the feminine gender denote people of female and male sex (nomina attributiva), e.g., avétinja (fool) f. (m.); (c) they refer to a person or animal of male sex, but are grammatically of feminine gender, e.g., burègdžija (a person who makes burgers) m. (gram. f.); (d) variants of masculine and feminine genders represent male and female persons, e.g., ànglist(a) (anglicist) m. (f.), with the female form ànglistkinja also presented. In the DSCLNL and DSCLL, the mentioned nouns only receive the first of the two indicated markers, although there are examples of common-gender nouns, mainly nomina attributiva, with two markers, e.g., budala (fool). However, inconsistencies are observed in the DMNLL as well, as, for instance, the headwords analìtičar (analyst), ankètar (polister), apsòlvēnt (graduate) etc., are marked only as masculine nouns even though they are used for female persons as well.

Regarding declension, the data about it is nowhere near as exhaustive as in the case of the gender category. Not all case forms of nouns are noted, primarily for efficiency reasons. Moreover, it is debatable whether it is necessary to occupy dictionary space with listings of patterns that users already know or can easily predict (De Caluwe and Van Santen 2003: 82). Therefore, nouns either lack any data on case forms, or, grouped in square brackets, only those suffixes or complete case forms of nouns that differ in accent or morphology from the basic form are noted. These forms deviate from systematic predictability, and the case abbreviation informs about which form of the word is in question, e.g., bàstadūr [gen. sg. bastadúra, voc. sg. bȁstadūre, instr. sg. bastadúrom] (a person who is resourceful, skilled; everything is going their way). Although such practice is also present in Serbo-Croatian dictionaries, it is noticeable from a general overview that the DMNLL justifiably devotes more space to case forms, especially in comparison with the DSCLL. For the mentioned lexeme, for example, both the DSCLL and DSCLNL provide only the genitive singular ending.

Duplex forms are also noted, separated by a slash, e.g., alkohòličār [voc. sg. ‑e/-u, instr. sg. -om/-em] (alcoholic). Rare nouns that lack declension, mostly loanwords, are marked with the qualifier "nepromj." (invariable), e.g., Bȁntu (Bantu), vònderbra (wonderbra) ... The case forms of nouns that change meaning when inflected thus gain the status of a separate defining word, for example, većìnōm (mostly).

When it comes to grammatical number, it is well-known that common nouns, besides being listed in the singular, also have a plural form, which is usually not noted except in cases of accentual or morphological deviation from the basic form. In such instances, under the label "nom. pl." (nominative plural), its full form or just the ending is specified, either alone or together with other case forms, e.g., àviopark [nom. pl. aviopàrkovi] (airpark), bȁmbrēk [voc. sg. bȁmbrēče, nom. pl. bȁmbrēci] ... (a short fat person; chunky person). In the DSCLNL and DSCLL, however, only the abbreviation "pl." is used, without a case marker, even when the plural nominative form is provided along with other case forms.

If a noun has two accentual and/or morphological plural forms, both are listed and separated by a slash, for example: vȅlegrad [nom. pl. velegràdovi/ vȅlegrādi] (metropolis), buldožèrist(a) [gen. sg. -a/-ē, nom. pl. -i/-e] (catskinner). A noun that occurs only in the plural is marked with the Latin abbreviation "pl. t." (pluralia tantum), e.g.: Apeníni pl. t. (The Apennines). We consider this designation to be more precise than the mn. label used in the analyzed Serbo-Croatian dictionaries. However, nouns predominantly used in the singular, which are mostly material and abstract, including verbal nouns, are not marked as such either in the DMNLL, DSCLNL or DSCLL, even though such data would be valuable for users.

Only collective nouns receive a special qualifier: "zb." (italicised) if they are collective both morphologically and lexically (e.g., bȁlavčād zb. from balavče), and "zb." (not italicised) if they are collective only lexically (e.g., balavùrdija). This unusual difference in font style is not noted in the list of abbreviations. If a noun is not collective in all its meanings, then the "zb." label is placed before the specific sense, for example, bižutèrija a. zb. jewelry made from cheap materials and fake stones. This indirectly indicates that these lexemes or sub-lexemes do not have plural forms. On the other hand, as a suppletive form, for example, the plural of the noun brȁt [supl. pl. brȁća] (brother) is marked, which is also provided as a separate defining word with the "zb." qualifier, a more user-friendly approach as it does not require grammatical knowledge from the user.

The grammatical category of number can significantly impact the meaning of a lexeme, as seen in the case of zoological and botanical terms, where the plural form denotes a species, while the singular form refers to an individual representative. For example, àlbatros in plural refers to oceanic birds from the Diomedeidae family with a large wingspan; in singular, it denotes such a bird. Bèscvjetnica in singular means a plant that lacks flowers or does not bloom; in plural, it refers to such plants, known as Cryptogamae. Other instances of this type are very rare, but we can illustrate with the lexeme vrijème (time), one of whose meanings is related only to the singular: 1. (in sg.) a. philos. one of the two dimensions of existence (alongside space), manifested in the form of continuous duration.

Many nouns, either entirely or in a specific meaning, are accompanied by data about their colligation tendency⁶ towards the plural form, for example, bjȅčva (usually in pl.) refers to a type of short sock or leg warmer, bjèlača 1. (usually in pl.) short socks made of white wool, bàtina 2. (only in pl.) received beatings: to get a beating.

The DMNLL also notes specific functions of nouns, which are enumerated as separate senses, for example: vozàrica 1. a female carter, 2. (in attributive use) propelled by rowing, with oars (about a boat): boat ~; vrućìna 1. high air temperature, scorching heat. 2. (in adverbial use) hot, excessively warm; automàtik 1. a device or machine that independently performs the action for which it is designed. 2. (in the role of an invariable adjective) operating automatically, independently: ~ gearbox, etc. Data about the noun's rection (see batàljōn, battalion etc.) is not provided, neither in the DMNLL, nor in Serbo-Croatian dictionaries.

4.2 Adjectives

Like nouns, adjectives do not have a part-of-speech marker, except for invariable adjectives (marked with the qualifier "nepromj. pridj."), which are almost all of foreign origin, like the adjective bàtāl (cannot be used anymore, outdated) from Turkish and blȉnd (which protects, shields) which is from German. The canonical form of variable adjectives is the nominative singular masculine form, in the indefinite aspect if it exists, and in the positive, if it features a category of degree.

Gender is fully represented, as with nouns, so the forms for feminine and neuter genders are always provided, specifically their endings if they are distinguished only by these, or their full forms or occasionally complete forms if they accentually and morphologically deviate from the base form, e.g., vòlovskī, -ā, -ō (bovine), bèščūlan, -lna, -lno (senseless), bȉjel, bijèla, bijèlo (white). This indirectly indicates the morphological class of the defining word. With rare adjectives, data is provided in round brackets indicating that they are predominantly or exclusively used in the feminine gender, either entirely or in one of its senses, e.g., brȅmenit, -a, -o 1. (only in fem. gender) meaning pregnant.

Data on case and number is not provided because there are no deviations from the usual paradigm; instead, data on aspect and degree are provided in square brackets. Specifically, for all adjectives that have forms in both aspects, the definite aspect form is given after the marker "odr.", alongside the canonical indefinite form, for example, ȁlāv, -a, -o [odr. ȁlavī] (a person who eats a lot, insatiable, voracious, greedy), whereas in the DSCLNL and DSCLL, this is only done if they deviate in accent or phonemes from the base form. The dependency of a particular meaning of an adjective on its definite aspect is also regularly noted, e.g., vìsok, visòka, visòko 10. a. (only in def. aspect) which refers to someone in a prominent position in service; related to persons in such positions, as well as the tendency of certain adjectives to favor this aspect, for instance: vȉšeglasan, -sna, -sno and višèglasan, -sna, -sno [odr. vȉšeglasni and višèglasni] (usually in def. aspect) composed of multiple voices; performed with several voices, polyphonic. Adjectives that have only the indefinite or only the definite aspect do not carry information indicating the absence of forms in both aspects, although the latter indirectly signify this fact through their canonical forms.

Comparative forms are also provided in square brackets, but only, as in the DSCLL, if there have been phonetic alterations in their formation, for example, bȉjel, bijèla, bijèlo [comp. bjȅlji] which is the color of milk, snow. In contrast, in the DSCLNL, they are always recorded. Suppletive comparatives have their own dictionary entry, for instance, bȍljī, -ā, -ē comp. of good. The conditionality of a certain sub-meaning by the tendency towards the comparative form is noted as well, e.g., vìsok, visòka, visòko 6. b. (usually in comp.) aspiring to something noble, sublime, significant; possessing such qualities: higher ideals; higher principles. Superlative forms are justifiably not listed among the forms or as defining words, since their formation follows the rules without deviation.

Data about the rection with which certain adjectives are distinguished is noted by listing complements in the form of indefinite pronouns for people and things in the corresponding case, provided in round brackets before the definition of the lexeme or sub-lexeme, e.g., vȉčan (to something) a. accustomed to something, used to something. b. skilled in something, experienced, adept, proficient. c. well-acquainted with something, informed about something; vjȅšt 3. (at something) having great knowledge about something, very adept at something, accustomed to something, skilled, etc. There are also omissions in listing, as the data on rection is missing, for example, with the adjective blagonáklon (benevolent, favorable).

Noun usage of adjectives, which is often accompanied by restrictions related to aspect or number, is regularly noted as a separate meaning/sense. For example: bȍgāv, -a, -o 2. (in nominal usage) m. poor soul, weakling; vȍljen, -a, -o 2. (in nominal usage) (in definite aspect) a person who is loved; blȋžnjī, -ā, -ē and blȉžnjī, -ā, -ē 3.b. (in nominal usage) (usually in pl.) a person who is in close kinship with someone, a relative, kin; a person in general. However, certain substantivized adjectives have separate dictionary entries, e.g., Bȕgarskā (Bulgaria), indicating a lexicographic issue with the unclear boundary between polysemy and conversion.

4.3 Pronouns

Alongside pronouns, a qualifier for their morphological class ("pron.") is provided, and in the definition, sub-classes are specified, while characteristic case forms are given in square brackets: including accentual forms, enclitic forms, and even those characteristic of dialects, marked with the abbreviation "nar.". For example: vî pron. [gen. vâs, encl. vas, dat. vȁma, encl. vam, nar. vi, acc. vâs, encl. vas, nar. ve] 1. personal pronoun for the second person plural; the people to whom the speaker is addressing in speech or writing. 2. (Vi) for polite addressing of an individual, as an expression of respect: thank you. 3. (in dat.) functioning to enhance the meaning and draw attention. Adjective pronouns, as they are characterized by motion, are listed in the masculine gender, with endings for feminine and neuter, and with data about their class and subclass, e.g., vȁš, -a, -e possessive pronoun for 2nd person pl. 1. belonging to the larger number of people we address (to you). 2. (Your) in addressing someone out of respect, meaning "your". 3. (in nominal usage) m. (in pl.) family members, relatives; like-minded individuals, members of a party or society, etc. From these examples, we see that data is provided not only about the formative characteristics but also about the usage of the pronouns.

4.4 Verbs

Verbal entries are presented in the infinitive as the basic form, without a morphological class marker. Rare verbs that lack an infinitive are listed in the first person present tense and marked with the abbreviation "incomp." (incomplete verb), for example, vèlīm (complete and incomplete) [pres. vèlīš, vèlī, imperf. vȅljāh] incomp. meaning to say, tell, speak.

Information about the verbal aspect, which can be perfective (svrš.) (perf.) or imperfective (nesvrš.) (imperf.), is a mandatory grammatical detail. These markers also serve as indirect indicators of the word type. Verbs of different aspects but the same meaning are listed as separate entries, e.g., búpati (imperf.), bȕpiti (perf.) (to hit something making a dull sound, smash). Verbs with two aspects, predominantly of foreign origin, receive both markers: perf. and imperf., for instance, blokírati (block), vizuèlizovati (visualize). Subsequently, in square brackets, paradigmatic forms are registered, typically the 1st person singular present tense, and if there is a duplication of forms, then the 3rd person plural, as well as other forms if they deviate accentually or morphologically from the headword. For instance, alongside the verb vȉdjeti (see), besides the present tense, the imperative, imperfect, active participle, passive participle, past adverbial participle, and present adverbial participle are provided, indicated by abbreviations: [pres. vȉdīm, imp. vȉdi, imperf. vȉđāh, act. part. vȉdio, vȉdjela, vȉdjelo, pass. part. vȉđen, -a, ‑o, past adv. part. vȉdjēvši, pres. adv. part. vȉdēći]. The DSCLL records only the suffix for the 1st person singular present tense, without a marker, while the DSCLNL additionally provides the aorist, but lacks the passive participle and adverbials. For impersonal verbs, marked with the abbreviation 'impers.', the mandatory present tense in the grammatical block is provided in the 3rd person singular, which is the only form used, for example, bjȅnūti (perf.) [pres. bjȅnē] impers. meaning to clear up without full brightness, to become slightly beautiful, to improve (regarding the weather).

The dictionary entry for verbs that appear in both non-reflexive and reflexive forms, that is, without and with the particle "se", is divided into two parts marked with Roman numerals I and II. The first part deals with their non-reflexive use, and the second part with their reflexive use, as can be seen in the verbs like vaspitaváti (to raise, educate), vézati (to tie), vijòriti (to fly (a flag)) etc. Impersonal reflexive verbs are also presented under number II. Within the headword, the particle se is placed in round brackets if the verb has the same meaning with or without it, e.g., vijùgati (se) (to wind).

As with the other mentioned Serbo-Croatian dictionaries, the DMNLL dictionary does not provide explicit data about the (in)transitivity of verbs since this is included in the definitions and examples. Only occasionally, when it conditions the realization of a certain meaning, is direct data given about the government or the right valency of verb lexemes and, more often, sub-lexemes, for example, àdaknuti (someone) to expel, to drive away; bògatiti I (someone, something) to make rich, to enrich materially or spiritually; adresírati 2. (to someone) to intend/aim, usually a criticism or objection; vjenčávati 3. (with something) to adorn, to decorate, etc. As seen from the provided examples, complements are listed in the form of indefinite pronouns for people and things in the appropriate case, placed in round brackets before the definition. However, they can also be found within the definition, e.g., bŕziti 1. to hurry, to urge (someone) to move or work faster. Regarding the left valency, only occasionally, after the data on aspect and forms, is data provided about the logical subject, e.g., with bàstati (with the logical subject in dative, rarely in accusative), but it is absent, for example, with bòljeti, although this verb in every mentioned meaning is realized in a construction with the accusative of the logical subject.

As with nouns and adjectives, attention is occasionally given to the collocation of verbs. Predominantly, this concerns their propensity for a certain form or combination with specific types of words, for instance, brȉsāti 5. (usually in imperative) to disappear, to flee, to move away; bèndati (usually with negation) to consider, to heed, to attach importance, to notice; to respect, to care, to worry; blagovòljeti (usually with the infinitive) to honor someone with something, to express a willingness to do something, to condescend (in expressing respect, sometimes ironically), etc. In rarer instances, a particular semantic realization may be conditioned by a specific form, and this is also noted, for example, vȉdjeti 10. a. (in the 2nd person singular imperative) in an indefinite meaning, when expressing astonishment, surprise, warning, threat, reproach. b. (in the 2nd person singular present) when emphasizing the content of the statement, etc.

4.5 Adverbs, prepositions, conjunctions, interjections and particles

In addition to adverbs, prepositions, conjunctions, and interjections, corresponding abbreviations are used as part of speech markers (e.g., "adv." for adverb, "prep." for preposition, "conj." for conjunction, "interj." for interjection), for example:

- avanturìstički adv. meaning in an adventurous manner, or in the spirit of an adventurer;

- van prep. (with genitive) indicating 1. outside the boundaries of a space; beyond the framework of something. 2. exceeding boundaries and frameworks, not adhering to them: a. above, over. b. beside, against. 3. archaic for excluding: except, besides;

- a conj. meaning 1. adversative: a. to connect independent sentences in a complex one. b. to link words that are in direct opposition: old yet naive. 2. cumulative: a. to connect independent sentences in a complex sentence: and. b. to link words in a parallel relationship: thin but tall. 3. disjunctive, to connect sentence parts of opposite meaning from which one must choose: or. 4. (when emphasized: ȁ) traditionally temporal: when; as soon as, just;

- âj interj. is used to express various emotional states - most often pain, sorrow, etc.

Generally, as seen from the examples provided, their meanings and usage domains are elaborately given. For prepositions, the cases they are used with are always indicated in round brackets. With particles, the full name of the word type is provided within the definition, e.g., vàljda particle for indicating or highlighting hope, probably.

Since adverbs of manner can be compared, their comparative forms are only listed if they exhibit accentual deviations and phonetic alternations, for example, bȓzo [comparative bȑže/bȑžē] (fast). Suppletive comparative forms have separate dictionary entries, for instance, bȍlje and bȍljē 1. comparative of good (dobro) ...

Other functions of invariable word types are also noted, such as the adverb alègro 1. quickly, cheerfully, lively. 2. (in a nominal function) masculine, a fast musical movement, a piece performed quickly; the interjection bȁmbajāt 1. upon falling. 2. (in the function of an invariable adjective) at death's door, barely alive; dead, etc.

 

5. Conclusion

The lexicographic marking of the grammatical features of lexemes is one of the more significant issues in both theoretical and practical lexicography. It is particularly crucial for a general dictionary of a language, as it is essential to both describe and prescribe the language. The type, scope, and manner of presenting grammatical data primarily depend on whether the dictionary is in electronic or print form. In the case of an electronic dictionary, the data should ideally be as comprehensive and explicit as possible, enhancing its informative power. In contrast, for a print dictionary, the presentation of grammatical data tends to be more economical and, therefore, often indirect due to space constraints. Consequently, it requires a higher level of lexicographic competence from potential users.

The DMNLL began as a print version, although the identification of sources and the extraction of lexemes were performed using modern computer programs. Therefore, its authors had to be mindful of space-saving and the economy of the lexicographical description. As for the grammatical treatment of headwords, this implies the mandatory indication of the word type, either directly, with a special marker, or indirectly with a marker of some comprehensive grammatical category or through the definition. The first case involves invariable parts of speech (adverbs, adjectives, conjunctions, interjections, and particles) and nominal pronouns. The second case pertains to nouns, adjectives, and adjectival pronouns, which are consistently marked by grammatical gender, and verbs, which receive obligatory aspect notation. Other types of grammatical data including categories of case and number, comparative forms of adjectives, verbal reflexivity, valency, specific functions, and collocational preferences of headwords are generally provided only if they are systematically unpredictable or if the meaning of the lexeme depends on them. Only adjectives that distinguish aspect have the form of the definite aspect indicated, although from an economy standpoint, this is unnecessary because it is predictable.

The described approach reflects a considerable reliance on traditional Serbo-Croatian lexicography, which is expected given the developmental trajectory of the Montenegrin language standard and Montenegrin studies. However, on the one hand, there was a missed opportunity to correct some omissions in the dictionary of the Serbo-Croatian language already identified in the literature, such as the non-marking of singularia tantum nouns (Ristić 2003: 128). Of course, since certain mass and abstract nouns, as uncountable, can also have plural forms beyond their basic meaning, when they are associated with e.g. something concrete, this should also be noted in their lexicographical description. We also consider the omission of noun valency to be an inherited shortcoming that should be corrected in future work. On the other hand, nominal case forms are grouped within the grammatical block in square brackets, as are verb forms, which seems to be a better solution than traditional. A useful innovation is the marking of the natural gender of nouns when it does not align with the grammatical gender.

Despite the clear intention in DMNLL to consistently provide grammatical features, the systematic nature is occasionally disrupted due to the lexicographers' oversight, which we particularly indicated in our analysis. Nevertheless, it can be stated that, in terms of grammatical data presented through transparent metalanguage, it is characterized by a high degree of consistency in the compilation of dictionary entries, and, along with removing observed omissions, it represents a solid foundation for the continuation of a serious and necessary lexicographic project such as the creation of a general dictionary of the Montenegrin language.

 

Endnotes

1. In 2006, Montenegro restored its independence and the following year proclaimed Montenegrin as the official language. However, since Montenegro is home to not only Montenegrins but also members of the Serbian, Bosniak, and Croatian peoples, Serbian, Bosnian, and Croatian languages are also in official use, being very close linguistic standards that emerged through a kind of layering of Serbo-Croatian after the dissolution of the Yugoslav state, in whose different versions Montenegro had existed since 1918.
2. As a thesaurus and academic dictionary, DSCLNL is one of the most important and demanding projects of the Serbian Academy of Sciences and Arts. The first volume was published in 1959, and to date, 21 volumes have been released. Once completed, it will contain over 35 volumes with more than 500,000 headwords. Although the sources and methodology of work are presented in the introduction of the first volume, it is entirely expected that they have been supplemented and changed during its many decades of compilation.
3. DSCLL is a six-volume dictionary, with 150 000 entries. The first three volumes were published in 1967 and 1969 in collaboration between Matica srpska and Matica hrvatska, while the latter three were independently published by Matica srpska in 1971, 1973, and 1976 after the Croatian side withdrew due to disagreements over the dictionary's concept and the nature of the common literary language.
4. Dictionaries, even if they claim to be solely descriptive or prescriptive, generally combine these two approaches because by describing the lexicon, they are essentially describing the linguistic norm, and the public expects from them at least a certain degree of normativity (Vrbinc et al. 2020: 576).
5. Alongside the headwords that we take as examples, from the dictionary entry we cite data only about those features that we wish to illustrate, and we do so in the manner presented in the DMNLL. In addition to the original language, headwords are provided in parentheses in English whenever possible, while their definitions are always translated.
6. Hoey (in Atkins and Rundell 2008: 304-305) views collocation as a "midway relation between grammar and collocation," explains that, for instance, a countable noun that almost always appears in the plural and never at the beginning of a sentence represents "a prima facie case of colligation - an observable preference for a subset of the available grammatical options." Besides nouns, they discuss the colligational tendencies of verbs and adjectives.

 

References

A. Dictionaries

DMNLL (Dictionary of Montenegrin National and Literary Language). 2016. Rječnik crnogorskog narodnog i književnog jezika I−. Podgorica: CANU.         [ Links ]

DSCLNL (Dictionary of Serbo-Croatian Literary and National Language). 1959−(2020). Rečnik srpskohrvatskog književnog i narodnog jezika I−(XXI). Beograd: SANU.         [ Links ]

DSCLL (Dictionary of Serbo-Croatian Literary Language). 1967−1976. Rečnik srpskohrvatskoga književnog jezika 1−6. Novi Sad: Matica srpska (1−3, Zagreb: Matica hrvatska).         [ Links ]

B. Literature

Apresjan, J.D. 2010. Prospekt aktivnogo slovarja russkogo jazyka. Moskva: Jazyki slavjanskih kultur.         [ Links ]

Atkins B.T. and M. Rundell. 2008. The Oxford Guide to Practical Lexicography. Oxford/New York: Oxford University Press.         [ Links ]

De Caluwe, J. and A. van Santen. 2003. Phonological, Morphological and Syntactic Specifications in Monolingual Dictionaries. Van Sterkenburg, Piet (Ed.). 2003. A Practical Guide to Lexicography: 71-82. Amsterdam/Philadelphia: John Benjamins.         [ Links ]

Čirgić, A., I. Pranjković and J. Silić. 2010. Gramatika crnogorskoga jezika. Podgorica: Ministarstvo prosvjete i nauke.         [ Links ]

Gortan-Premk, D. 1980. O gramatičkoj informaciji i semantičkoj identifikaciji u velikom opisnom rečniku. Naš jezik 24(3): 107-114.         [ Links ]

Kačić, M. 1994. Rječnik i gramatika. Filologija 22-23: 297-302.         [ Links ]

Katičić, R. 1994. Leksikografija i gramatika. Filologija 22-23: 281-286.         [ Links ]

Kostić-Golubičić, M. 1997. Gramatika i rečnik: teorija i praksa, tradicionalno i netradicionalno. Naučni sastanak slavista u Vukove dane 26(2): 457-463.         [ Links ]

Kostić-Tomović, J. 2017. Savremena nemačka leksikografija. Beograd: FOKUS - Forum za interkulturnu komunikaciju.         [ Links ]

Kozyrev, V.A. and V.D. Černjak. 2015. Leksikografija russkogo jazyka. Vek nynešnij i vek minuvšij. Sankt-Peterburg: Izdatelstvo RGPU im. A.I. Gercena.         [ Links ]

Marković, M.A. 2014. Gramatika u srpskim rečnicima. Dragićević, R. (Ed.). 2014. Savremena srpska leksikografija u teoriji i praksi: 69-91. Beograd: Filološki fakultet Univerziteta u Beogradu.         [ Links ]

Popović, Lj. 2003. Integralni rečenični modeli i njihov značaj za lingvistički opis i analizu korpusa. Naučni sastanak slavista u Vukove dane 31(1): 201-220.         [ Links ]

Ristić, S. 2003. Leksikografski metajezik i srpska deskriptivna leksikografija. Naučni sastanak slavista u Vukove dane 31(1): 119-130.         [ Links ]

Šubarić, S. and J. Đurčević. 2023. The Names of the Balkan Peoples and the Names of the Inhabitants of Balkan Countries in Lexicography (on the Example of the Dictionary of Montenegrin National and Literary Language). Lexikos 33: 68-89.         [ Links ]

Tafra, B. 2005. Od riječi do rječnika. Zagreb: Školska knjiga.         [ Links ]

Topolinjska, Z. 2002. Šta tražim u rečniku? Deskriptivna leksikografija standardnog jezika i njene teorijske osnove: 33-37. Beograd: Institut za srpski jezik SANU, Novi Sad: Matica srpska.

Vrbinc, M., D. Šipka and A. Vrbinc. 2020. Normative Labels in Two Lexicographic Traditions: A Slovene-English Case Study. Lexikos 30: 561-582.         [ Links ]

Zgusta, L. 1991. Priručnik leksikografije. Sarajevo: Svjetlost.         [ Links ]

 

 

Appendix A

Examples of the dictionary articles barikadirati, barjaktar and blijed in the DMNLL.