Pragmatic functions of anglicisms in the Montenegrin language Círculo de Lingüística Aplicada a la Comunicación

. This paper deals with the traditional distinction between necessary and unnecessary loanwords as a problematic one because it marginalises the functions of so-called unnecessary loanwords. By adopting a cognitive-linguistic approach, this paper aims to highlight the importance of both types of loanwords from a language user’s point of view. To this end, we examine anglicisms in Montenegrin on the basis of a new pragmatic model which distinguishes between catachrestic and non-catachrestic loanwords (Onysko, Winter-Froemel 2011). Our study has shown that the pragmatic distinction of anglicisms is possible in Montenegrin, thereby proving it is unacceptable to divide them into necessary and unnecessary ones. The results also show that anglicisms do not always have all the characteristics typical of their categories, which brings us back to the cognitive-linguistic approach we have taken in our study.


Introduction
There are numerous classifications of loanwords within the enormous corpus of literature on contact linguistics. A very common classification, based on the semantic relationship these words have with the existing lexicon of the recipient language, distinguishes loanwords that do not have a semantic equivalent in the recipient language from loanwords that do have an equivalent in the recipient language. Terminology varies according to the different interpretations of various authors, some calling them necessary and luxury (Tappolet 1913;Danesi, Rocci 2009), cultural andcore/prestigious (Myers-Scotton 2002;Haspelmath 2009;MacKenzie 2012), functional and modern (Melchers & Shaw 2011), etc. Regardless of the terminology, the aforementioned distinction actually sees loanwords either as necessary ones, which denote new concepts by introducing new terms into the recipient language, or unnecessary ones, which have semantic equivalents in the recipient language. As such, this distinction is seen here as controversial because it is based on prescriptive linguistics which prescribes what is needed in one language and what is not, without taking into account all the functions that the "unnecessary" loanwords have from a usage-based perspective.
Taking a cognitive-linguistic approach that studies language from a usage-based perspective, this paper aims to emphasise the importance of these loanwords, as well as the importance of both types of loanwords from a language user's perspective. To that end, the paper questions the traditional classification of loanwords, highlighting the fact that loanwords are important not only when they fill lexical gaps but also because they have different semanto-pragmatic functions. In order to examine the semanto-pragmatic nature of loanwords, we are investigating anglicisms in the Montenegrin language. In doing so, we are adopting Gottlieb's broader definition of anglicism(s) as "any individual or systemic language feature adapted or adopted from English, or inspired or boosted by English models, used in intralingual communication in a language other than English" (Gottlieb 2005: 163). As a theoretical model for analysis, we have taken the pragmatic classification of anglicisms into catachrestic and non-catachrestic (Onysko, Winter-Froemel 2011), which is grounded in Levinson's pragmatic theory (Levinson 2000) and which distinguishes the anglicisms that bear the implicature of informativeness (I-implicatures) from those that bear the implicatures of manner (M-implicatures). On the basis of the described theoretical model, we hypothesise that anglicisms in Montenegrin will demonstrate they are all necessary from a semanto-pragmatic aspect, which will show that the traditional distinction of loanwords into necessary and unnecessary is unacceptable. Imagined in this way, the research ultimately aims to draw attention to the importance of studying the pragmatic nature of loanwords and contribute to the unexplored field of the pragmatics of loanwords in general and in Montenegrin in particular.
The paper is organised as follows. Chapter 2 gives an insight into literature that explains the problem of the traditional distinction of anglicisms into so-called necessary and unnecessary loanwords, emphasising the importance of a cognitive-linguistic approach that aims to describe loanwords from a usage-based perspective. In this regard, section 2.1 offers a pragmatic framework for the analysis of anglicisms in Montenegrin, based on a new and terminologically more appropriate distinction of anglicisms into catachrestic and non-catachrestic. Chapter 3 is the central part of the paper and it presents the proposed analysis of the pragmatic functions of anglicisms in Montenegrin. It first describes the corpus and the methods of analysis, and then illustrates the categorisation of anglicisms according to their pragmatic functions. Chapter 4 offers conclusive remarks about the pragmatic nature of anglicisms as a field that has never been explored before in Montenegrin, and it highlights the comparability with the results of the study of anglicisms in German that served as a model. As the field of pragmatic nature of anglicisms still remains widely unexplored, this chapter finally emphasises the necessity of similar research in other languages as well.

Pragmatic functions and classification of loanwords
As we mentioned, the traditional classification of loanwords recognises only loanwords motivated by a need to designate new concepts and fill lexico-semantic gaps in the recipient language. Factors motivating the transfer of luxury or "unnecessary" borrowings have traditionally been treated as less relevant, and it is the same case with the functions that these loanwords have in their use. The importance of linguistic and non-linguistic factors behind the process of borrowing, that is, the denotative and connotative, semantic and pragmatic meaning of loanwords is, however, something that has been recognised by many linguists, even by the aforementioned ones who see loanwords either as necessary or unnecessary. Haspelmath (2009: 46), for example, explains that there is no real need for cultural (necessary) borrowing since languages can form words on their own, while MacKenzie and Myers-Scotton emphasise the importance of conversational or stylistic effects of luxury loanwords (MacKenzie 2012: 31-32) pointing out that these also arise from a need -various social and psychological needs of the speakers (Myers-Scotton 2006: 265).
However, loanword studies focusing on their usage and pragmatic nature, that is, the pragmatic functions they have from a usage-based perspective are rare. In one of the first such papers, Hans Galinsky (1963: 134) offers a list of stylistic functions of anglicisms in German: "1) providing national American color of settings, actions, and characters; 2) establishing or enhancing precision; 3) offering or facilitating intentional disguise; 4) effecting brevity to the point of terseness; 5) producing vividness, often by way of metaphor; 6) conveying tone, its gamut ranging from humorous playfulness to sneering parody on America and ʻAmericanized' Germany; 7) creating or increasing variation of expression.ˮ Examining the functional properties of anglicisms in Spanish, Rodríguez González (1996) relies on Halliday's functional components (Halliday 1978) -ideational, interpersonal and textual, emphasising the importance of examining the interpersonal and textual, as these components are actually pragmatic in their nature. Unlike the ideational (or referential or denotative) function, which is characteristic of loanwords filling lexical gaps in the recipient language, the interpersonal and textual are much more complex. Loanwords with an interpersonal component are stylistically marked and expressive in meaning, which means they can fulfill a variety of functions such as expressing affectation, irony, positive or negative associations, euphemism etc., while those with the textual component aim to achieve clarity, precision, economy of expression, variation of expression, a foreign atmosphere etc. In this way, as Rodríguez González (1996-1: 125) explains, the interpersonal and the textual component speak in favour of loanwords being motivated by prestige rather than need, as well as in favour of prestige itself, which is, from a sociolinguistic point of view, as important a reason for borrowing as need. His study is, therefore, a reminder that linguistic and extra-linguistic causes are equally important for the process of borrowing and that anglicisms, due to their pragmatic nature (interpersonal and textual component), have a much more complex function than filling gaps (ideational component) in the recipient language. So far, Branka Drljača Margić (2011) seems to be the only linguist from the pluricentric BCSM (Bosnian, Croatian, Serbian and Montenegrin) language area who is taking a stand in defence of luxury or so-called unnecessary loanwords. She points out that the process of borrowing is caused by a variety of linguistic and extra-linguistic factors, which is why not only denotation but also connotations and associations of loanwords need to be taken into account when these words are compared to their equivalents in the recipient language (Drljača Margić 2011: 54, 63).
On the basis of all this, it can be concluded that a different approach to "unnecessary" loanwords and their functions is rare in loanword studies. The classification of loanwords as necessary and luxury, functional and modern, denotative and connotative, or, in other words, necessary and unnecessary, is the result of a traditional, puristic approach to the study of loanwords that ignores the extralinguistic reasons behind their transfer as well as their semantopragmatic characteristics. However, expansion into the domain of loanword research has been enabled with the development of cognitive linguistics, which sees linguistic meaning as being formed in contact with the extralinguistic world, and therefore aims to describe language in its actual use. Meaning, which is the primary focus of cognitive linguistics (Langacker 1987: 12), is understood as the conceptualisation of the experience from the extralinguistic world. It is classified into radial categories with loose boundaries and a centre-periphery structure, which means that concepts or members are organised in relation to a central or prototypical concept. Due to the fact that language is an instrument of conceptualisation, the conceptual approach aims to describe language from a usage-based perspective, which is why cognitive linguistics is a usage-based model of language (Kemmer, Barlow 2000). Given the importance of usage data here, quantitative, corpus-based research represents an indispensable part of the methodology of cognitive linguistics. As such, this methodology soon becomes useful for loanword studies that seek to challenge the structuralist perspective on loanwords and its traditional methods of analysis which relied on small corpora and dictionary observation. By linking up to the cognitive linguistic approach, research into loanwords has started to shift its attention to a usage-based perspective on borrowing and quantitative, corpus-based methods. Recent papers that rely on large corpora, as well as sociolinguistic and psycholinguistic experimental methods, speak in favour of this tendency (Rohde, Stefanowitsch, and Kemmer 1999;Kemmer and Barlow 2000;Zenner, Speelman, and Geeraerts 2011;Backus 2012;Winter-Froemel 2013;van Meurs, Hornikx and Bossenbroek 2013;Daems, Heylen and Geeraerts 2015;Serigos 2017...). Examining the usage of loanwords and their semanto-pragmatic nature as a field that has been neglected in traditional studies, the cognitive-linguistic approach has made a great contribution, both theoretically and methodologically, to the study of loanwords. It is thanks to this approach that loanwords that duplicate already existing words can be finally seen as equally important as loanwords that fill lexical gaps, as they can also be necessary in communication from a speaker's perspective. This is how it has become possible to adopt a new approach to the dichotomy between necessary and unnecessary loanwords, as well as to describe it with different, more objective terminology. Such an approach and terminology was first offered in the study of Onysko and Winter-Froemel (2011).

Catachrestic and non-catachrestic loanwords
In order to emphasise the importance of studying loanwords and their functions from a usage-based perspective, as well as the necessity of more objective terminology for them, Onysko and Winter-Froemel (2011) introduced a distinction of loanwords as catachrestic and non-catachrestic innovations. Grounded in pragmatic theory, the distinction is supported by Levinson's theory of presumptive meanings (Levinson 2000) which divides pragmatic meanings into three types of generalised conversational implicatures -inferences of quantity, informativeness, and manner. In order to apply Levinson's pragmatic theory to the dichotomy between catachrestic and non-catachrestic loanwords, the authors take as relevant the implicatures of informativeness and manner. Expanding the notion of catachresis from the rhetorical tradition, they define loanwords as catachrestic if they introduce a new concept into a language, i.e. if they bear the implicature of informativeness and therefore appear as unmarked forms. On the other hand, loanwords that have a semantic equivalent in the recipient language are marked, linked to the implicature of manner and therefore defined as non-catachrestic. The notion of markedness makes the distinction clearer due to the fact that marked linguistic forms are usually longer, morphologically complex, less frequent and less neutral in register, which is best illustrated by lexical doublets (Levinson 2000: 138 in Onysko, Winter-Froemel 2011: 1555: (1) He was reading a book. (I-implicatures: He was reading an ordinary book.) (2) He was reading a tome. (M-implicatures: He was reading a voluminous book.) In order to test the proposed classification, they investigated the pragmatic functions of the 101 most frequent anglicisms in German. The results show that a third of the most frequent anglicisms from their corpus introduce a new concept into German (computer, internet, laptop, software, tennis, website...), and that the rest of the anglicisms share a semantic space with their German equivalents but can be differentiated in meaning by virtue of their different usages. Illustrated by the example of the anglicism shoppen ('shopping for fun, as a hobby') and its German equivalent einkaufen (Onysko, Winter-Froemel 2011: 1557-1558, their differentiation demonstrates the applicability of Levinson's pragmatic implicatures to the catachrestic/non-catachrestic dichotomy: non-catachrestic anglicisms do not introduce new concepts into a given language, but bearing M-implicatures they do take on specific meanings when compared to their RL (recipient language) equivalents. Although most anglicisms bear M-implicatures and are therefore pragmatically marked, the analysis reveals that this categorisation is not strict due to the fact that some anglicisms show certain deviations from their categories. Thus, the analysis shows that there are non-catachrestic anglicisms which, due to their diachronic development in the recipient language, bear I-implicatures (club, bar...), and appear as unmarked forms, just like their German equivalents (Verein, Lokal/Theke). Similarly, there are anglicisms that have German equivalents (computer -Rechner; homepage -Leitseite, Startseite ...) but function as catachrestic ones because of innovation in their meaning, as well as polysemous anglicisms which, depending on their meanings, can be categorised as both catachrestic and non-catachrestic (cockpit -catachrestic in the meaning of 'racing car', but non-catachrestic in the meaning of 'pilot's cabin'). As the authors point out (Onysko, Wiinter-Froemel 2011: 1563, these discrepancies show that pragmatic functions change over time with the changes of their language users' habits, and that the insight into these changes can only be possible with their further examination. Considering all the above, we believe that the proposed theoretical model of loanword classification set forth by Onysko and Winter-Froemel (2011) can offer a new perspective on loanwords that have been traditionally and unfairly treated as "necessary/unnecessary". This is the reason why we want to examine the pragmatic functions of anglicisms in Montenegrin as a completely unexplored field in this language. Following a pragmatic analysis of anglicisms in Montenegrin, conclusions will be drawn about the necessity of loanwords from a language user's point of view, which will determine whether their traditional classification as necessary or unnecessary is possible when pragmatic functions are taken into account.

The pragmatic functions of anglicisms in Montenegrin
On the basis of the theoretical framework for pragmatic classification of anglicisms in German (Onysko, Winter-Froemel 2011), we aim to investigate the pragmatic functions of anglicisms in Montenegrin. As the study that was conducted in German, our analysis has also confirmed that pragmatic classification of anglicisms is possible in the Montenegrin language. After describing the corpus and the methods of analysis in the following section, we will discuss results and draw conclusions on the classification of anglicisms in Montenegrin.

Corpus and methods of analysis
In spite of the fact that contemporary linguistic analyses rely on large and linguistically annotated, electronic corpora, there still is no such corpus of Montenegrin. This unfortunate shortcoming can primarily be attributed to the relatively young status of the Montenegrin language and the incomplete process of its standardisation -it was only after the declaration of independence of Montenegro in 2006 and the ratification of the new Constitution in 2007 that Montenegrin became the official language in Montenegro. In order to gather a large and representative corpus, we decided to combine the online portal of Vijesti, as the most widely read newspaper in Montenegro, and the corpus of the Montenegrin Academy of Sciences and Arts. Although we have not managed to overcome the shortcomings of the linguistically unannotated corpus that we had to resort to in our research, we have gathered a large multi-millionword corpus of newspaper, scientific and professional texts from various fields.
The news portal Vijesti had already served as the source for a collection of anglicisms that we needed for the purposes of a larger, unpublished study that we undertook in 2018. Thus we were able to gather 500 anglicisms from fields strongly dominated by English -fashion, economics and law, informatics and sports. From this number, we chose the 100 most frequently occurring anglicisms for this research. The extraction of the most frequent anglicisms was made possible with the electronic search engine (on the news portal Vijesti), and the File Seek program (for the corpus of the Montenegrin Academy of Sciences and Arts), which both allow to search for a certain word, count the number of its occurrences and provide all the contexts it occurs in. While the context offered in both corpora is fully valid, the results regarding the word frequency cannot be as accurate as they would have been if an electronically lemmatised language corpus had been used. Such a corpus would merge all grammatical variants with the base form of an anglicism, which would undoubtedly increase its frequency. In spite of being aware of this shortcoming, we are also aware of the fact that the higher token frequency that would have been offered by the lemmatised corpus does not exclude the validity of the results we can get by counting the occurrences of the base form of an anglicism. If a lemmatised corpus had been used in our research, the frequency of our anglicisms would certainly not have been any smaller, but only higher, which makes the data available to us less precise, but no less reliable.
In order to minimise the bias of the results and to arrive at a more valid classification, we followed some steps from the studies which we based our analysis on (Onysko 2007: 105-112;Onysko, Winter-Froemel 2011: 1556. From our previously compiled collection of anglicisms, we therefore excluded English proper nouns, and nouns derived from both English and Montenegrin elements (trenirati, implementacija...), as well as nouns that are generally thought of as anglicisms but are of uncertain etymology (agent, standard, rekord...). There are steps, on the other hand, that we were not able to follow in our study due to the different nature of the languages in contact in our research. Although they were not included in the aforementioned studies that served as a model, orthographically adapted anglicisms were included in our analysis because of the ideally phonemic orthography of Montenegrin, which requires the adaptation of foreign lexis according to its rules, and the avoidance of unadapted, ʻraw' loanwords. Processes of formal adaptation vary across languages, whereby the degree of integration is higher in languages with a smaller number of speakers than in those with a larger number because of language-related policies that seek to preserve the national identity of small societies, as well as because of the different genetic and typological characteristics of those languages compared to English (Hoffman 2008).
In order to classify anglicisms as catachrestic or non-catachrestic, we consulted different monolingual and bilingual dictionaries in order to check whether our anglicisms have Montenegrin equivalents. In the absence of such dictionaries for the Montenegrin language, we had to use dictionaries of Serbian and Serbo-Croatian (see References) and to rely on our native speakers' intuition. Agreeing with the view that the split-up of the former Serbo-Croatian language into Serbian, Croatian, Bosnian and Montenegrin was political and not linguistic in its nature, we consider these dictionaries to be completely valid sources for the detection of Montenegrin equivalents for the anglicisms in our analysis. Having detected potential equivalents in one of the listed sources, we checked them from a usage-based perspective, i.e. we checked whether they are really used in Montenegrin. For this purpose, we used what had previously served as the source for the extraction of our anglicisms -the web portal of the news magazine Vijesti and the corpus of the Montenegrin Academy of Sciences and Arts. With the semanto-pragmatic meaning as the main criterion for the classification of anglicisms as catachrestic or non-catachrestic, closer insight into the usage of Montenegrin equivalents has been very important, especially when it comes to anglicisms which have shown certain deviations from their categories.

Results and discussion
Analysis of the pragmatic functions of anglicisms in Montenegrin has allowed their classification as catachrestic or non-catachrestic. The classification is given in Table 1, and followed by a detailed discussion of the results. The table shows there are more non-catachrestic (60) than catachrestic anglicisms (46), but also a discrepancy in the number of total (100) and pragmatically indexed anglicisms (106). This is due to the presence of six polysemous anglicisms (indicated by an asterisk in the Table) whose different meanings have different pragmatic values, which is why we have classified them into both categories. This is not the only group of anglicisms that reflects the complexity of their classification as we have also recognised a number of catachrestic and non-catachrestic anglicisms which show deviations from their categories. That is why we have paid special attention to these anglicisms within the categories they belong to.

Catachrestic anglicisms
The results show that as many as 46 out of 100 anglicisms are catachrestic borrowings. This is not surprising, however, if we bear in mind that our anglicisms belong to a specialised vocabulary, i.e. come from the fields of fashion, informatics, sports and economics/law. The number of catachrestic anglicisms has to be large in the aforementioned fields in which the Anglo-American culture has had the greatest impact. New inventions and objects in these fields were extensively transferred from English to the non-English speaking world which, understandably, had no names for concepts it had not been familiar with. Catachrestic anglicisms in our analysis are, therefore, words that fill lexical gaps in Montenegrin.

Catachrestic anglicisms with Montenegrin hyperonyms
Unlike the catachrestic anglicisms we described, we came across a certain number of catachrestic anglicisms whose pragmatic functions, i.e. categorical affiliation was difficult to determine. As typical catachrestic anglicisms, they also bear I-implicatures, but there are Montenegrin words that can function as their equivalents. While the fact that they have Montenegrin equivalents brings them closer to the non-catachrestic category, the I-implicatures they bear, on the other hand, undoubtedly dissociate them from that category. A more careful insight into their meaning leads us to the conclusion that these anglicisms indeed bear I-implicatures because they have a more specific meaning than the meaning expressed by their Montenegrin substitutes. The relationship between these anglicisms and their Montenegrin substitutes is not, therefore, synonymic as in the case of non-catachrestic anglicisms and their Montenegrin substitutes, but rather hyponymic. That is why we have decided to classify them under a special subcategory of catachrestic anglicisms. We have recognised these anglicisms in the examples such as: kompjuter (computer), mini (miniskirt), laptop, tender, samit (summit), derbi (derby), displej (display), dres (dress)... As in German, where the anglicism computer has its close translational equivalent Rechner (Onysko, Winter-Froemel 2011), Montenegrin demonstrates the same relationship between kompjuter (computer) and računar. Apart from denoting an electronic computer and therefore merely functioning as a translational equivalent for the anglicism kompjuter, the word računar, however, has a broader meaning of a calculating device: ( Similarly, laptop has a narrow meaning in comparison to the Montenegrin word računar or the anglicism kompjuter. Denoting a special type of computer which is small and portable, laptop appears as a hyponym to its Montenegrin near equivalents računar and kompjuter. In the same way, the anglicism displej (display) denotes an electronic device only, or a part of such a device that presents visual information (of a computer, tablet etc.). Its Montenegrin nearequivalent ekran, on the other hand, is a more general term denoting a part of an electrical or electronic device that shows visual information (of TV, computer, etc.). One of the anglicisms belonging to this subcategory is the anglicism mini, which denotes a special type of skirt -a very short skirt/dress that characterised the fashion of the 1960s. Apart from a descriptive translation, there is no specific term for this type of skirt in Montenegrin, which is why the anglicism is the unmarked choice in Montenegrin. Interestingly, the adjectival function of this loanword (ʻwhich is smaller in size/capability/scope... than usualʼ), on the other hand, is marked, which means that the anglicism mini as an adjective would be classified as non-catachrestic. Within this subcategory we have also recognised anglicisms from the field of sports -dres (dress) as a special type of clothing (the sporting uniform of a local/national team), as well as derbi (derby) which, unlike the more general Montenegrin equivalent utakmica, refers to an important sports match between two local rival teams. When it comes to the register of economics and law, examples of this kind include samit (summit), which does not denote just any type of meeting, but an important meeting of highest-level officials, which is also considerably shorter than its descriptive translational equivalents in Montenegrin -sastanak šefova država i vlada; sastanak na vrhu. In a similar way, the Montenegrin word konkurs appears as a hypernym to its "correspondingˮ anglicism tender, which refers exclusively to calls/offers for the realisation of certain projects, delivery of goods or provision of services: (7) Klinički centar Crne Gore raspisao je krajem decembra javni konkurs za dodjelu pet specijalizacija. (Vijesti, 10 January 2019) (At the end of December, The Clinical Centre of Montenegro launched a call for applications for five medical specialties.) (8) Нијесу укључена средства која се издвајају за капитални буџет, као и инвестиције по основу расписивања Тендера за градњу ХЕ на Морачи и Комарници. (CANU, CG u XXI stoljeću -u eri kompetitivnosti) (Excluded are the funds intended for the capital budget, as well as investments based on the announcement of Tenders for the construction of HPPs on Morača and Komarnica)

Non-catachrestic anglicisms
Our classification shows the presence of 60 non-catachrestic anglicisms. As our anglicisms belong to different registers (fashion, sports, informatics, economics and law), non-catachrestic anglicisms primarily contain stylistic features in comparison to their unmarked Montenegrin synonyms that are used in everyday, non-specialised language. The usage of anglicisms in professional, specialised language, e.g. the registers of fashion, sports, informatics, economics and law, is motivated by their prestige and international status, because of which they are more associated with a higher level of professionalism, innovation, and progress than their RL synonyms.
Experts in certain areas resort to the use of non-catachrestic anglicisms also because they are shorter and more precise than their RL equivalents: plejmejker -vođa igre/tima/na terenu (playmaker); brend -robna marka/zaštitni znak (brand); bord -upravni odbor (board); startap -novoformirana firma (startup), holding -centralna firma (holding)... The conciseness of anglicisms and adherence to the language economy principle has helped them acquire an international, global status, which is why they are frequently used in advertising, too.
The prestigious status of English as a global language makes anglicisms sound more educated, fashionable and modern than their RL equivalents not only in advertising and other specialised languages. For this reason, a large number of non-catachrestic anglicisms are particularly used, for example, in the fields of leisure, fashion, beauty, music, entertainment, fashion, show business etc., as shown by examples from our analysis: kul (cool), fan (fan), parti (party), stajling (styling), luk (look), fešn (fashion), kasting (casting), imidž (image), glamur (glamour)...
Another difference between non-catachrestic anglicisms and their RL synonyms is a discrepancy in their collocational range. Used in the names of fashion shows, events, manifestations etc., the anglicism fešn (fashion), for example, dominantly appears in combination with other (usually unadapted) anglicisms -show, week, day, connection... Its Montenegrin equivalents moda/modni, on the other hand, are not exclusively used in the names of fashion manifestations and events, and are used with both, Montenegrin words and anglicisms: (Bearing in mind that two decades ago there were almost no fashion events in our country, and not as many designers as today, the audience had an opportunity to find out all about fashion events worldwide...) Finally, the corpus analysis has shown that non-catachrestic anglicisms are often used interchangeably with their RL synonyms to avoid repetition: (12) Uoči nastupa u Kotoru, tajlandska DJ-ka je za "Vijesti" pričala o underground sceni, žurkama u njenoj domovini, omiljenim destinacijama i objasnila zašto misli da je Crna Gora jedno od najboljih parti mjesta na svijetu. (Vijesti, 5 May 2017) (Shortly before her performance in Kotor, a Thai DJ told Vijesti about the underground scene, parties in her home country, her favourite destinations and explained why she thinks Montenegro is one of the best party places in the world.)

Non-catachrestic anglicisms bearing I-implicatures
As in the case of catachrestic anglicisms, we have recognised a number of non-catachrestic anglicisms that show certain deviations from their category. Despite having Montenegrin synonyms, these anglicisms do not bear implicatures typical of their category. Being formally (orthographically, phonologically and morphologically) unmarked and shorter in length from their RL equivalents, these anglicisms have been widely and frequently used in Montenegrin since the moment of their transfer. Such usage has influenced a change in their pragmatic functions over time, so they have, on one hand, partly lost their M-implicatures, and have acquired, on the other hand, I-implicatures. This is how they gradually came to share a similar semanto-pragmatic space with their Montenegrin synonyms, thus losing lexical markedness as a defining characteristic of non-catachrestic borrowings.
The fact that they have really lost their lexical markedness is best illustrated by their widespread parallel use with Montenegrin synonyms in a variety of fields. The anglicism trend, for example, is alternately used with the Montenegrin synonym moda in the field of fashion, whereas in the fields of law and economics, as well as in everyday language, the same anglicism is used interchangeably with two other Montenegrin synonyms -tendencija and/or pravac: (Bank's balance sheets displayed a tendency of growth reaching at the end of September an amount 5.5% higher than last year.) Corpus evidence (from the news portal Vijesti/from the corpus of the Montenegrin Academy of Sciences and Arts) shows different usage frequencies -the anglicism trend (1000+/148) is used much more frequently than its Montenegrin synonyms moda (473/2) and tendencija (454/49), which further confirms the unmarked nature of the anglicism.
In the end, the corpus analysis has indicated that the anglicism lider (leader) has also become close to its unmarked Montenegrin synonyms as it shares with them almost the same usage frequency (lider 1000+/32 -prvak 1000+/8; vođa 1000+/12) and semanto-pragmatic space: (21) Trofejni trener je osvojio pet titula prvaka NBA i vodio tim do plej-of plasmana svake sezone od 1998. godine. (Vijesti, 30 April 2019) (The trophy winning coach won five Championships in the NBA and has led the team to the playoffs every season since 1998.) (22) Želimo im na svaki način pomoći da postanu značajni partneri u svojoj državi i lideri u paraolimpijskom sportu. (Vijesti, 31 October 2018) (In every way possible, we want to help them become significant partners in their country and leaders in Paralympic sports.) In the sphere of politics, however, the anglicism seems to have a positive connotation in comparison with its Montenegrin equivalent vođa. In the context of local politics, vođa all too often has a negative connotation, as it has acquired the meaning of a ruler, reflecting thus the discontent of the Montenegrin opposition and its dissatisfaction with the ruling party and its 'ruler'. The word is often written with a capital letter -such use of the word is clearly sarcastic and indicative of the antipathy towards the party leader, who also happens to be the current president of Montenegro: (His hands are sweating, his heart is beating fast, his voice is trembling, because of the presence of one single man. The party leader -the overall leader.)

Polysemous anglicisms with both pragmatic functions
Apart from the illustrated anglicisms that show deviations from their respective categories, there are a small number of polysemous anglicisms which have also made pragmatic classification less straightforward. These anglicisms turn out to have both catachrestic and non-catachrestic functions, as their different meanings bear different implicatures. Due to the fact that they are both catachrestic and non-catachrestic at the same time, we have counted them in both categories. There are six such anglicisms in our analysis: čarter (charter), čip (chip), marketing (marketing), menadžment (management), procesor (processor) and set (set).
When denoting a written document granting privileges to or recognising rights of the people, the anglicism čarter (charter) has a non-catachrestic function, as there are Montenegrin equivalents ugovor, povelja, and povlastica. While the Montenegrin synonyms function as more general terms, corpus evidence shows that the anglicism bears M-implicatures for exclusively referring to documents of Rotary organisations granting privileges to or recognising rights of the Rotarians. When referring to a contract for the hire of a ship or an aircraft; an aircraft, boat, etc., available on charter; or a charter holiday or flight, the anglicism bears I-implicatures and there are no Montenegrin equivalents with which it can be substituted.
The anglicism čip (chip) has a catachrestic function when it refers to a microchip, as there is no Montenegrin word to denote this concept. However, when denoting a small disk representing money in poker and other games, čip is a non-catachrestic loanword, as it has a Montenegrin equivalent of French origin -žeton (jeton), which existed in the language long before the anglicism. Since the anglicism chip found its place in the analysis of anglicisms in German (Onysko, Winter-Froemel 2011), we take the opportunity to compare the semanto-pragmatic features of this loanword in German and in Montenegrin as different recipient languages. While being catachrestic in both languages in the sense of a microchip, the anglicism has different pragmatic functions in these two languages with its reference to the monetary token in casino games. Apart from being an unmarked term for game tokens generally in German, the anglicism is the only term used in this language for poker and other games of Anglo-American origin. Contrary to that, the same anglicism in Montenegrin does not show any preference to being used in the context of Anglo-American games rather than games of a different origin, and it is interchangeably used with its counterpart žeton. As a loanword of a newer date, it appears as a marked term and functions as a non-catachrestic anglicism: (26) Кockari mogu kupiti ili prodati čipove za "texas holdem poker". (Vijesti,12  The anglicism marketing has two meanings in Montenegrin: 1) the strategic functions and activities involved in appealing to consumers, such as advertising, branding, pricing, sales etc. and 2) radio and TV advertising. In the first one, the anglicism has its Montenegrin translational equivalents tržištenje, utrživanje or tržništvo, but they are not actually used in Montenegrin. To support this, there was not a single occurrence of any of these three words in our corpus. That is why we interpret the anglicism as catachrestic in this sense, in spite of the fact that it has Montenegrin synonyms. When denoting radio or TV advertising, the anglicism appears as a marked term as the preferred term is the Montenegrin equivalent reklame and the abbreviation EPP (ekonomsko-propagandni materijal: economic propaganda material), although the latter one is falling out of use.
In the meaning of 'conducting or supervising of something (such as a business)', as well as 'the collective body of those who manage or direct an enterprise' the anglicism menadžment (management) is a non-catachrestic loanword, as the Montenegrin equivalents upravljanje, rukovođenje, and uprava, rukovodstvo are used to denote the same concepts. While its Montenegrin synonyms have a broad field of usage, the anglicism is dominantly used in the field of economics. Apart from having a non-catachrestic function, the anglicism is also catachrestic when denoting the science of managing, as there is no Montenegrin word for that concept.
When it is used to denote the central processing unit of a computer, the anglicism procesor (processor) is catachrestic and it bears I-implicatures as there is no Montenegrin equivalent. The anglicism is non-catachrestic, conversely, when denoting a food processor, as there are Montenegrin equivalents multipraktik and mikser -both anglicisms of an earlier date than procesor, which is why they are unmarked choices in the context of food/kitchen equipment.
Finally, the anglicism set is catachrestic in the meaning 'a group of games constituting one division or unit of a match', as it introduces a new concept of a division in the duration of some sports games. In all its other meanings ('a group of things of the same kind; the movie stage; a session of music'), the anglicism is non-catachrestic, as it has its Montenegrin equivalents (skup, zbirka, garnitura ili komplet; kolekcija; scena). The length of the anglicism and its formal unmarkedness is probably the reason why our corpus evidence shows its high frequency usage (1000+).
The illustrated polysemous anglicisms bearing I-and M-implicatures have been counted in both the catachrestic and non-catachrestic categories. Apart from these, we have also encountered cases of polysemous anglicisms whose different references have the same pragmatic function, so we listed them in one category only, either as single instances of catachrestic or non-catachrestic anglicisms. Some of the examples include dizajn (design) as a catachrestic anglicism since it does not have Montenegrin equivalents for its transferred senses ('the arrangement of elements or details in a product or work of art; a decorative pattern; the creative art of executing aesthetic or functional designs'); rejting (rating) which is non-catachrestic, since all its three meanings ('relative estimate or evaluation: standing; an estimate of an individual's or business's credit and responsibility; an estimate of the percentage of the public listening to or viewing a particular radio or television program') can be expressed by Montenegrin equivalents (ugled; popularnost; pozicija), the anglicism skor (score) which has Montenegrin counterparts (rezultat; muzički komad; partitura) for all its senses ('the copy of a musical composition in written or printed notation; a musical composition; specifically : the music for a movie or theatrical production; a number that expresses accomplishment (as in a game or test) or excellence (as in quality) either absolutely in points gained or by comparison to a standard'), etc.

Conclusion
The research in this paper had as its starting point a problematic classification of loanwords into necessary and luxury ones. Taking a cognitive-linguistic approach that studies language from the usage-based perspective, we argue that the aforementioned distinction is inadequate because it only places value on the loanwords that fill lexical gaps while ignoring the functions loanwords have in their use. Our goal was, therefore, to examine the semanto-pragmatic functions of anglicisms in Montenegrin as an unexplored field in this language. As a theoretical model for our analysis, we have taken the pragmatic classification of anglicisms into catachrestic and non-catachrestic (Onysko, Winter-Fromel 2011) which has so far only been tested in German. Apart from its neutral and descriptive terminology, we have chosen this model as it is grounded in both semantics and pragmatics offering, therefore, a novel and comprehensive approach to loanword classification and loanwords in general.
In order to test the applicability of the described model in Montenegrin, this paper analyses the pragmatic functions of 100 highly frequent anglicisms. The results have shown that the classification of anglicisms into catachrestic and non-catachrestic is also possible in Montenegrin. As in the study of anglicisms in the German language, the majority of anglicisms in Montenegrin appear as non-catachrestic loanwords. While catachrestic anglicisms introduce new concepts into Montenegrin, non-catachrestic anglicisms share a similar semanto-pragmatic space with their Montenegrin equivalents. The illustrated examples show how these anglicisms, despite having Montenegrin synonyms, are often preferred terms in certain domains of usage because of their word length, preciseness, stylistic features, registers, connotations, collocations, prestige, etc. Such a result sheds new light on loanwords that have been traditionally named as luxury ones and interpreted as unnecessary as their different usages and pragmatic functions make them necessary from the language user's perspective. In this way, we are able to confirm our hypothesis -it is only when we take into account all the functions loanwords have that we can come to the conclusion that the prescriptive, puristic classification of words into necessary and unnecessary ones is untenable.
Although it has been possible to divide anglicisms into two categories based on their pragmatic functions, the classification has not been simple and straightforward. As in the study of anglicisms in German (Onysko, Winter-Froemel 2011), the results of our analysis have shown the presence of polysemous anglicisms whose different meanings have different pragmatic values, which is why we classified these anglicisms into both categories. In addition to that, within both the catachrestic and non-catachrestic categories we have recognised a number of anglicisms that show slight deviations from their respective categories but, unlike in the indicated study, we have classified them into special subcategories. Unlike typical catachrestic anglicisms that do not have a RL equivalent, there are a number of anglicisms that can formally be substituted by Montenegrin words. However, as these Montenegrin words are their superordinates or hypernyms rather than their synonyms (as is the case with non-catachrestic anglicisms), we have decided to classify them under a special subcategory of catachrestic anglicisms. There are also atypical noncatachrestic anglicisms which we have classified under a special subcategory. Due to being well-established and highly frequent in Montenegrin, these non-catachrestic anglicisms have partly lost their M-implicatures and started to share I-implicatures with their Montenegrin synonyms. The categorisation of anglicisms has therefore been slightly reframed in our study as, apart from catachrestic and non-catachrestic categories, it distinguishes three more groups of anglicisms -the category of polysemous anglicisms, as well as the subcategories of atypical catachrestic and noncatachrestic anglicisms.
Our analysis has, therefore, shown that the categorisation of anglicisms into catachrestic and non-catachrestic is also possible in the Montenegrin language. As in German, this distinction does not, however, always imply clear and strict boundaries between these two categories, since we have recognised a number of catachrestic and noncatachrestic anglicisms that do not have all the characteristics typical of their respective categories. Taking into account all the above, it can be concluded that taking a usage-based, cognitive-linguistic approach to borrowing was well-grounded, because our classification of anglicisms based on their pragmatic functions has shown overlapping of their categories and the unequal status of anglicisms as their members, which is actually characteristic of linguistic categorisation as seen from the perspective of cognitive linguistics (Taylor 1995).
Finally, we believe the significance of this research lies in the fact that it demonstrates the applicability of the proposed model of pragmatic classification (Onysko, Winter-Froemel 2011) in the Montenegrin language too. Not only does this confirm that the model is cross-linguistically relevant, but it also paves the way for similar research in other languages.