An explanatory combinatorial dictionary of English conflict lexis: a case study of modern political discourse

Размещено на http://www.allbest.ru/

An explanatory combinatorial dictionary of English conflict lexis:

Olga Solopova and Tamara Khomutova

Abstract

Though political discourse is in the mainstream of modern studies, scholars haven't so far paid much attention to compiling political discourse-oriented dictionaries. The need to further develop lexicographic theory and practice for specific purposes and advance new methods to dictionary making is a challenge that linguists are facing today. The aim of the case study is twofold: to work out the principles for making an Explanatory Combinatorial Dictionary (ECD) of English political conflict lexis and the microstructure of an ECD entry. The source of the data is the NOW corpus; the material is current American political discourse (2022). The ECD is generally consistent with Mel'cuk's Meaning-Text theory (MTT). The authors describe a process of collecting and processing the data: corpus search and analysis, automatic and manual text processing, glossary compilation with the use of lexicographic, semasiological, and etymological methods and present an example of an ECD entry consisting of semantic, phonological, and cooccurrence zones. The findings prove that the use of electronic text corpora offers an effective way for compiling a specialized discourse- based dictionary. The research illustrates the validity of MTT: though based on the data of “language in context”, the dictionary is synthesis-oriented: it aims at speech production. The paper is the first result of a bigger project sketching the overall framework of the discursive ECD of political conflict lexis, which subsequent studies will hopefully develop with more precision and detail. The dictionary will be helpful for scholars in linguistics, discourse analysis, media and communication, political science, and conflict studies.

Key words: lexicography, political discourse, explanatory combinatorial dictionary, conflict- provoking lexis, corpus, dictionary entry

Аннотация

Принципы создания англоязычного толково-комбинаторного словаря конфликтогенной лексики: на материале современного политического дискурса

О.А. СОЛОПОВА, Т.Н. ХОМУТОВА

Необходимость разработки теории и практики специальной лексикографии и применения новых дискурсивных методов к составлению словарей для специальных целей представляет собой важную задачу, которая стоит перед современной лингвистикой. Целью настоящей работы является: 1) разработка принципов создания англоязычного толково-комбинаторного словаря (ТКС) политической конфликтогенной лексики на материале корпуса текстов политического дискурса; 2) разработка микроструктуры словарной статьи. В качестве методологических принципов построения ТКС политической конфликтогенной лексики использованы принципы теории «Смысл-Текст» И.А. Мельчука. Источником материала является корпус News on the Web. Материал исследования представлен современным американским политическим дискурсом (2022). В статье рассматривается процесс сбора и обработки данных, включающий четыре этапа: формирование и анализ корпуса текстов, автоматическую обработку текстов, ручную обработку текстов, создание глоссария с использованием методов лексикографического, семасиологического и этимологического анализа. Приводится пример словарной статьи, микроструктура которой включает семантическую, фонологическую и сочетаемостную зоны. Результаты проведенного исследования подтверждают, что использование электронных корпусов текстов позволяет эффективно создавать специальные дискурсивные словари, «встраивая» лексикографические данные в социальный, политический и геополитический контекст. В статье представлены первые результаты большого проекта, которые определяют общую методологию создания англоязычного ТКС политической конфликтогенной лексики. Принципы теории «Смысл-Текст», ориентированные, прежде всего, на языковой синтез и производство речи, не исключают возможности его дальнейшего использования для анализа текста и дискурса. Словарь будет полезен специалистам в области лингвистики, дискурс-анализа, массмедиа, политологии и конфликтологии.

Ключевые слова: лексикография, политический дискурс, толково-комбинаторный словарь, конфликтогенная лексика, корпус, словарная статья

Many people have contributed, both directly and indirectly, to the ideas and analyses of this project. But most of all we would like to thank Igor A. Mel'cuk and Alexander K. Zholkovsky for the inspiration we gained from their Explanatory Combinatorial Dictionary. The insights of Mel'cuk's Meaning-Text theory turned out to be a source of challenge and motivation: his contribution to the project has been invaluable. Using the metalanguage of lexical functions, adopted in the Meaning-Text theory, to characterize Igor Mel'cuk as a lexicographer, we must state that he is genuine (Ver), influential (Magn), and well-versed (Bon) in the system of language.

Introduction

Recent years have seen a great surge of interest in lexicography, which is caused by the active development of science, and growing political, cultural and research contacts. Lexicography is generally understood as the theory and practice of compiling dictionaries, which are closely connected and interrelated. The focus of lexicography is the ways of organizing a dictionary entry, the structure of dictionaries and the methods of their compilation. We share the view advanced by Bergenholtz and Gouws (2012) that lexicographic theory is “a discipline not only directed at compiling dictionaries, but in a more general way at producing information tools”, which “can ensure enhanced information retrieval” (Bergenholtz & Gouws 2012: 40). Modern lexicography has significantly expanded and upgraded its tools with computer technologies for compiling and using dictionaries. At the same time, studies that organize lexicographic knowledge and introduce new lexicographic theories are rather scarce.

Within the context of novel lexicographic studies those of discourse lexicography should be highlighted, as the classical definition of lexicography is being transformed and modified in modern discourse studies. At the current stage of development, lexicography is interpreted more as the theory and practice of knowledge representations about the world, since, as Kozyrev and Chernyak (2015) state, dictionaries “respond to all changes in life and society, demonstrating a picture of dynamic language processes in a static form” (Kozyrev & Chernyak 2015: 6). Lexicographic practice has reached a new level and is becoming more anthropocentric and sociocentric. Lexicography functions within the framework of lexicographic discourse, which is interpreted as a discursive practice, as Issers (2015) puts it, “the dynamic organization of those communication systems within society, which, on the one hand, reflects the speech behavior and thinking mode of a social community, and on the other hand, it creates new forms of communication in a given socio-cultural reality” (Issers 2015: 65). In the case of considering lexicographic discourse as a discursive practice, it is possible to “refer to the sociocultural, cognitive, communicative and pragmatic features of the dictionary” (Plotnikova 2014: 23). The study of sociocultural, cognitive and communicative- pragmatic influence on dictionary compilation and interpretation determines the appeal to advance new discursive technologies in dictionary making.

Keen interest in the anthropocentric perspective of lexicographic research can be traced in dictionaries of political discourse, since political discourse, to a great extent, sets the linguistic worldview of people. At the same time, political lexis functions both as a unit of language and as a discourse formation that implements not only dictionary but wider meanings due to the existence of a certain system in the social and cognitive-discursive space. However, to date the number of political discourse-oriented dictionaries is relatively limited, one that we could think of is the Dictionary of Modern Political Labels (Skovorodnikov & Kopnina 2021), which is definitely not enough. With this gap in mind, we propose a discourse approach to compiling a dictionary of political conflict lexis. Modern political discourse is abundant in such lexis, which presses for its lexicographic study.

Our aim is to work out the principles for building an explanatory combinatorial dictionary (ECD) of English political conflict lexis based on a large corpus of political discourse texts that will test theoretical and computational methodologies, but, above all, to provide a tool for linguists, discourse analysts, translators, interpreters, and linguistic expertise specialists. As such, our project draws on previous work that shows the inadequacies of current dictionaries to meet the needs of the specialized user.

Theoretical framework. Discourse lexicography

Fesenko (2015) introduces the concept of a discourse dictionary which has the following characteristics: 1) it describes not only the denotative, but also the connotative meaning of the entry (appearing in the context); 2) the dictionary is not prescriptive, but descriptive; 3) the dictionary contains socio-cultural, cognitive and communicative-pragmatic information; 4) the entry is presented in all genre and stylistic diversity (Fesenko 2015: 52). Compiling discourse dictionaries is a laborintensive process that requires enormous efforts on the part of both theorists and practitioners of lexicographic science. Currently, linguists have done much work on the way of reforming lexicography and turning it from a theoretical construct into such a system that, as Sandomirskaya (2001) states, “refracts the general narrative of the language into the field of practical action, and captures the structure of this general narrative in its own settings” (Sandomirskaya 2001: 219).

Discourse dictionaries include a number of products, one of them is the electronic terminographic dictionary of phraseological units which actively function in Ukranian and Russian linguistics of the last century (Krasnobaeva-

Chernaya 2020). Active use of popular science discourse in mass media accounts for inclusion of terminology in the dictionary of neologisms that reflects the discursive vector of modern lexicography (Gromenko 2020). Another innovative lexicographic product is a discourse dialect dictionary based on data from Internet search engines, where “contexts act as a means of illustrating semantics” (Golev 2019: 114). A new multilingual dictionary of metaphors correlates with the discursive and lingua-cultural vector of development of lexicography, which is aimed at “solving the problem of lexicographic presentation of cultural connotation, assigned to the semantics of figurative words and expressions” (Gerasimova 2020: 95). Obviously, the lexicographic product serves to preserve the cultural heritage of a certain lingua-cultural community. In this regard, dictionaries of concepts act as an effective tool to gain insight into a certain culture, for example, on the basis of Russian vocabulary, phraseology and paroemiology (Con 2010), dialectology (CDDLP 2006-2012), lingua-cultural logic (ML 2005, 2009, 2014), concepts of Russian culture (Radbil & Saigin 2019), etc. The lingua-cultural aspect in dictionary compilation is clearly seen in the study of a dialect linguistic personality and its lexicographic fixation (Zemicheva 2017), gender parameters of dialect speech (Demeshkina & Tolstova 2017, VD 1998-2002, MDSD 2009-2010). A broad overview of discursive dictionaries is provided by Koshkarova and Solopova (2021) (Koshkarova & Solopova 2021).

Our brief analysis shows that present-day lexicography develops in the mainstream of the discursive-anthropocentric paradigm, when the dictionary reflects the functioning of the language not after the fact, but at the moment of its existence in the language community, which brings about the problem of specialized lexicography.

Specialized lexicography

Relative to discourse lexicography, as well as lexicography in general, is the problem of general and specialized lexicography. According to Bergenholtz and Gouws (2012) there are three branches of lexicography with respect to its object and functions: general lexicography, specialized lexicography, and general and specialized lexicography. The three branches of lexicography describe language for general purposes, language for specific purposes (LSP), and both languages simultaneously (Bergenholtz & Gouws 2012).

As stated above, discourse lexicography is anthropocentric and sociocentric, which means that discourse dictionaries describe the actual functioning of discourse at the moment of its generation in specific settings: social, cultural, temporal, spatial, etc. This brings about the idea that discourse lexicography is specialized per se. Nielsen (2013) points out that the two key fundamentals of specialized lexicography are its object, the dictionary, and its objective (Nielsen 2013). The dictionary is a tool providing specific types of help concerning one or more subject fields and their related LSP to specific types of users in specific types of situations. A specialized dictionary contains the necessary lexicographic data, arranged and presented using appropriate lexicographic structures. Dictionary functions thus become the point of departure for any discussion of dictionaries and the data they contain must be specifically adapted to user needs and competences. By identifying the key components of specialized lexicography lexicographers can draw up guidelines that can help them design, evaluate, make and use “ideal” and specific types of objects in LSP lexicography (Nielsen 2013: 24).

Our aim, as noted earlier, is the ECD of English political conflict lexis of the XXI century (second decade). It means that problems of political lexicography, as well as relations between lexicography and ideology, power and politics are of great importance for our research.

Political lexicography

Lexicography and dictionaries are never value-free, apolitical or asocial. Instead, they are subject to ideology, power and politics (Chen 2019: 362). Ideology and power are aspects of a dictionary that a lexicographer and a dictionary user have to encounter in any serious lexicographic enterprise (Kachru 1995: lxv). Chen (2019) states that “a discourse approach to lexicography is needed to unmask the power relations behind, and the ways in which language serves to sustain or disrupt the relation in the global context, before a change in the status quo is possible, as such an approach - with emphasis on social context - enables a practitioner to look at the lexicographic event as a social event and the dictionary as discourse with its own rules and principles” (Chen 2019: 364). Thus, the social context and power relations surrounding the production of a dictionary should not be overlooked.

This approach is taken by Benson (2002) who seeks to reveal the ethnocentric representation in the English dictionaries published in Britain/USA and to show how knowledge in the English dictionaries is filtered through Anglo-American perspectives on English in the world (Benson 2002, Preface). He showcases that the Oxford English Dictionary was more or less explicitly a project of British imperialism concerned with the consolidation of English as the dominant language of the world. It means that the dictionary does not simply replicate its source or just `transport' meaning; rather, it creates meaning; it rewrites and represents things in new ways (Chen 2015). We share Chen's idea (2019) that “a researcher in lexicographic discourse analysis should embed the lexicographic data in the social context, taking a political stance explicitly and focus on self-reflection as a scholar” (Chen 2019: 368).

Chen states that discourses project meaning; therefore, they evoke ideologies (Chen 2019: 370). Wodak and Meyer (2015) define ideology as a “collectively shared coherent and relatively stable set of beliefs or values” (Wodak & Meyer 2015: 30) while van Dijk (1998) presents a multidisciplinary approach to ideology involving cognitive and social psychology, sociology and discourse analysis (van Dijk 1998). That is, as Chen (2019) puts it, “socially, ideologies sustain the interests of groups; cognitively, they serve to organize the social representations (attitudes, knowledge) of the group, and thus indirectly monitor the group-related social practices, and hence also the discourse of its members” (Chen 2019: 370).

A dictionary is a representation of the world/society seen through the lexicographer's perspective. Lexicographers are not immune from ideology. Ideologies as ways of representing and constructing society reproduce unequal relations of power, relations of domination and exploitation. For example, we may investigate how lexicographers create a dictionary, within which the historical depth and geographical breadth of the language and the world could be exposed for examination from different perspectives. To take an example, western lexicographers can define “Russian special military operation” as: a war waged by Russia against Ukraine (2022) following Ukraine's attempt to join the EU. This definition attributes the Russian aggression to Ukraine's act of `national-liberation movement', which contradicts Russia's official view. Russian lexicographers will attribute it to the special military operation of Ukraine's demilitarization and denazification which Russia was forced to engage in. The operation was triggered by US and European elites in search of making Ukraine an enclave of NATO, which threatens Russia and Russian people in Ukraine. Lexicographic discourse studies may also be used to examine how the dictionary influences social understanding of a language and what the world is like. All this brings about the need for a special branch of lexicography, that of political lexicography.

The term “political lexicography” has been introduced by the Russian linguist S. A. Manik (Manik 2019). The novelty of her approach is determined by a broad lingua-cultural and discursive-pragmatic context, including the involvement of users in the process of dictionary compiling. The researcher gives a broad overview of existing political terminology dictionaries (Manik 2019). However, none of them are classified as discourse dictionaries. The author explores the discursive- communicative and cultural properties of English political terminology but fails to attribute the corresponding dictionaries to those of political discourse. Since our aim is to build an ECD of English political conflict lexis, we went in search of political conflict dictionaries.

Among the existing dictionaries we managed to find some special dictionaries of conflict, or conflict-provoking terms, which are rather few (Davletchina 2005, Dmitriev 2012, Merriman & Barrach-Yousefi 2021, Miller & King 2003, The Law Dictionary, Shipilov & Antsupov 2020). Conflict-provoking terms are understood as words that cause negative emotions and trigger conflicts. As for special political conflict dictionaries we failed to find any. That is why our research will be based on large corpora of English political conflict discourse and dictionaries of conflict terms available. One more issue worth discussing in connection with compiling the dictionary of political conflict lexis is that of combinatorial lexicography.

Combinatorial lexicography

Many of the discourse dictionaries mentioned are in fact combinatorial dictionaries, i.e., dictionaries of collocations. DeCessaris (2013) points out that “most words in the general vocabulary of a language do not have a clearly identifiable meaning out of context, rather they have a meaning potential that becomes specified once the context of use is established” (DeCessaris 2013: 16). Firth (1957) states that meaning only exists in context and through interpretation of that context, so that words can only be understood through the company they keep (Firth 1957: 11), thus, dictionaries can only ever show meaning potentials. Meaning lies in the present, it is about how people use words now: words do not have meanings, meanings have words. Once these premises are accepted, we can look for meaning potentials in corpora (Williams 2013: 26). Hence, we can term combinatorial dictionaries the forerunners of discourse lexicography which paved the way to discourse lexicography proper. Early examples of research into collocations appeared in Shakhmatov (1898), Ushakov (1935-40), Ozhegov (1949), but current understanding of a collocation as a fundamental factor in word meaning started in Russian lexicography only with DMLR (1948-1965), and in foreign studies with “Les mots et les idees. Dictionnaire des termes cadrant avec les idees” (Lacroix 1956), and the BBICDE (1990).

By the end of the XX century, the ideas of lexicographic reflection of lexis combinatorial properties continued to develop and improve. Russian lexicography started to explore new approaches to the interpretation of syntagmatic connections of words, e.g., Apresyan (1974), Zholkovsky & Mel'cuk (1967), Morkovkin (1977), etc. The description of the combinatorial properties of words was provided in the Dictionary of Combinability of the Russian language (DCRL 2002); Dictionary of Russian and English Lexical Intensifiers (DRELI 2007); ECD of the Russian Language (Mel'cuk & Zholkovsky 1984, 2016); New Explanatory Dictionary of Russian Synonyms (NEDRS 2000); Russian Associative Dictionary (Karaulov 2002) and others. A comprehensive overview of different trends in the development of Russian and foreign combinatorial lexicography is offered in (Vlavatskaya 2013).

In this paper we will try to apply the principles of compiling combinatorial dictionaries put forward by Mel'cuk and Zolkovsky (Mel'cuk & Zholkovsky 1984, 2016). We share the view advanced by M. V. Vlavatskaya (2013) that up to date this is the most original and effective dictionary project describing combinatorial- syntagmatic potential of the word (Vlavatskaya 2013). Mel'cuk and Zholkovsky's ECD is a theoretical dictionary, anchored in a linguistic theory - namely, Meaning- Text theory (MTT) - and making use of its conceptual tools. The general concept of the dictionary claims: “Natural language is a system that establishes correspondences between any given meaning and all texts expressing it; accordingly, the linguistic description of a certain language should be a set of rules that put all the texts of this language in line with any meaning” (Mel'cuk, 1995: 4). In MTT, meaning is no more than the invariant of synonymous paraphrases and it is thought to be directly accessible to speakers, being a part of their intuitive language competence.

The ECD has three general characteristics: 1) it is active, oriented towards speech production; 2) it is semantic (explanatory); 3) it is combinatorial, describing syntactic and lexical cooccurrence in a systematic way. Mel'cuk and Milicevic advance the following principles for compiling an ECD: 1) the formality principle which stands for the formal description of lexical units; 2) descriptive coherence principles which mean that both internal and external coherence should be achieved; 3) uniform treatment principles which demand that all linguistic units of the same class and vocables of the same lexical field must be described in a similar way; 4) the internal exhaustiveness principle which means that the description of a lexical unit must contain all the necessary information to use language correctly and find any other lexical units semantically related to this particular lexeme (Mel'cuk & Milicevic 2020: 203-209).

The best-known feature of the ECD and its basic notion is a lexical function that serves “a formal tool used to describe all types of lexical relations in a systematic way” (Mel'cuk & Milicevic 2020: 142, 161). This dependency associates with a lexical unit (its argument or keyword) and a set of synonymous lexical units that express a specific meaning associated with the former. Lexical functions can be paradigmatic and syntagmatic: paradigmatic lexical functions are derivative, related to the selection of an adequate value of the keyword, while syntagmatic lexical functions are combinatorial, encoding the collocational potential of the keyword. Any function meets specific requirements: it must embrace a relatively large number of pairs of words; it can be expressed in various ways, which implies the existence of a phraseological relation between the arguments and their values (for detailed treatment see Mel'cuk & Milicevic 2020).

The proposed dictionary, as well as ECDs in general, lexicographically provides the performance of the Meaning-Text model (MTT) and is similar to explanatory dictionaries, where lexical combinability and syntactic properties of words are reflected. This concerns general language dictionaries. However, discourse, which is language in real life context, is as multifaceted as life itself, and compiling such a dictionary would imply much time and effort on the part of lexicographers. Here arise the notions of domain-specific knowledge and specialized lexicography which is intended to help users build LSP discourse. As stated above, we refer discourse dictionaries to specialized lexicography. In our project we will try not merely copy Mel'cuk's model but introduce some novel principles to modern lexicographic practice, which will allow us to describe the conflict-provoking system of political discourse as a mobile and dynamic formation that is influenced by extra-linguistic factors.

Materials and methods

The case study, presented in the paper, addresses the specific issue of building an ECD of English Political Conflict Lexis by the in-depth analysis of the principles of compiling the dictionary and the microstructure of an ECD entry through both qualitative and quantitative methods. The principles of a research corpus compilation are of utmost importance for the results to be obtained. Collecting and processing the data for the research was a four-fold process, involving a) corpus search and quantitative analysis, b) automatic text processing, c) manual text processing, and d) glossary compilation with the use of lexicographic, semasiological, and etymological methods.

Research corpus compilation and corpus search

The source of the material was the News on the Web Corpus from English- Corpora.org (NOW). Nowadays text corpora are considered the most important tool for research in both computational and other branches of modern linguistics (Soloviev et al. 2022). Currently, the corpus contains web-based periodicals (2010- present), covering newspapers and magazines from more than twenty Englishspeaking countries. As the ECD of English Political Conflict Lexis is planned to be an experimental dictionary based on a large corpus of political discourse texts, it determines the data the source of the material and the research corpus are supposed to comprise. The central topic of discourse must concern a controversial or divisive political issue that centers on conflicting international priorities: this year one of the highest rated concerns in global political news has been the outbreak of a military clash between Ukraine and Russia, turning the latter into the most sanctioned nation in the world. As US officials frame America's role in the war in ambitious terms and their strategic thinking in relation to Russia and Ukraine is quite transparent, it is the American discourse that was chosen as the material for analysis.

Thus, the search query for compiling Research Corpus 1 was American articles, matching the keyword “Russia” within the date range: 24 February 2022- present, with the result being 1194347 matching strings. Normally, the NOW corpus has a query system that allows for effective searching by word form, lemma, or part of speech, including frequency lists and collocates sorted by time period, creating n-grams lists, generating concordance lines, comparing one section of the corpus to another. As the authors had a restricted license and a limited access to the options of the corpus, including only the search by a keyword and a time period (because of the sanctions context mentioned above), they had to compile Research Corpus 2, cutting it down to 500 texts (580544 words), sorted by relevance (Fig. 1) for the data to be processed with modern technologies using the data processing software, similar to the one the Now corpus is normally equipped with.

Figure 1. Research Corpus 2 

Corpus 2 comprised breaking news and editorials from top US news agencies and periodicals (APN, CNBC, CNN, Chicago Tribune, Daily Beast, Forbes, Fox News, Newsday, New York Post, Politico, the Atlantic, the New York Times, the Washington Examiner, the Wall Street Journal, USA Today, and others) to be further processed by LanA-Key program.

LanA-Key automatic processing

LanA-Key program is an easy-to-use application that requires neither special programming skills nor a preliminary stage of a corpus tagging, which serves as “an instrument for the automatic extraction of multi-component grammatically correct typed lexical phrases, featuring intelligent output and computationally attractive properties” (Sheremetyeva 2017) for various languages (Russian, English, German, French, Spanish, Chinese) and specific subject areas. The program was tuned to the extraction of nominal terminology from the English language political domain.

Processing the input string (Corpus 2), the program automatically lemmatized any word that can be unambiguously linked to a single headword The total number is 23485 lemmas, including numbers and functional parts of speech. (Fig. 2) and generated a frequency list of lemmas, excluding numbers and functional parts of speech (Fig. 3. NP Frequency).

Figure 2. Fragment of LanA-Key processing

Figure 3. Fragments of 1) NP Frequency, 2) NP ABC, 3) NP Reverse ABC

At the second stage the program sorted nouns and noun phrases (NPs) with lengths in 1-4 words alphabetically (Fig. 3. NP ABC list) and in reverse order (Fig. 3. NP Reverse ABC list), the latter being relevant for fixing attributive clusters (an attributive cluster is understood here as a group of NPs with a key noun and different attributive components modifying it). The iterative process allowed for classifying meanings within the ontology and revealing new words that will be included in the dictionary because of their environmental relevance rather than their simple frequency.

Manual processing

The NP Frequency list with the lemmas `Russia' (with 1212 entries, ranking second out of the total 23485 NPs) and `Russian' (with 187 entries, ranking twenty first) shows that Corpus 2 is quite representative for further analysis. The process of compiling an initial glossary The glossary presented in the paper is but the first step in compiling a more or less complete list of terms that will be included in the ECD of English Political Conflict Lexis. for the dictionary is a sequence of several data reduction phases that require manual processing and linguistic analysis with the help of etymological, lexicographic, and semasiological methods. Each procedure takes the input (a list of words) from the previous one and feeds its output (a new list of words) to the next phase.

The procedures used were as follows:

1) List 1, comprising 859 NPs with `Russia', `Russian' used as attributes, was extracted from the NP ABC.

2) The data were compared with the NP Reverse ABC; the NPs that had no attributive clusters were excluded from List 2, which was reduced to 423 NPs. In addition to NPs with `Russia' and `Russian', used attributively, List 2 contained NPs with nouns modified by `Kremlin' (134 entries), `Moscow' (287), `Putin' (527) as metonymic names of the country in political discourse, with some dictionaries considering them synonyms of Russia (Power Thesaurus).

3) To check the meaning of each key noun from List 2 six specialized dictionaries (both English and Russian) accessible via the Internet were used (Davletchina 2005, Dmitriev 2013, Merriman & Barrach-Yousefi 2021, Miller & King 2003, The Law Dictionary, Shipilov & Antsupov 2020). The available dictionaries present collections of words related to the concept of conflict and provide concise definitions of basic and specific terms, with one of the dictionaries (Dmitriev 2013) having a special mark of a `conflictogene” (conflict-provoking concept / conflict trigger) for over a hundred of entries. Still, a number of important limitations to this pilot study need to be considered. First, the dictionaries have a different number of conflict terms: two small-sized dictionaries (Merriman & Barrach-Yousefi 2021, Miller & King 2003) contain fewer than 200 entries while, for instance, the Law Dictionary includes over 15,000 words. The second important limitation lies in the fact that, in our opinion, some words, registered in the dictionaries, neither have conflict-provoking connotations nor belong to the concept of conflict; for this reason, further lexicographic data collection is required to determine a final list of terms either with precise `conflict' meaning in some uses or peculiar to the field of conflict studies. Third, when using Russian dictionaries of conflict studies, we focused mainly on a targeted search to choose between terms suggested by bilingual translation aids with the special subject area `politics' (Multitran Dictionary, ABBYY Lingvo 12, Cambridge Dictionary). In order to validate the findings further study needs to be carried out through a focus group method that offers a more effective way of assessing the quality of the resulting dictionary. However, as it has already been mentioned in 2.3, few dictionaries have been able to draw on any systematic description of conflict terms, with none of them focusing on political conflict studies, which points to an urgent need for a renewed discussion of compiling a dictionary of the kind. Thus, notwithstanding the limitations, the six dictionaries were used for lexicographic analysis.

The criterion for including a term in the initial glossary was its having a particular meaning related to the field of conflict studies, registered in at least two of the dictionaries (Table 1; the asterisk (*) specifies the meaning / form / use of the word).

Table 1. Fragment of lexicographic analysis

The ECD of English Political Conflict Lexis is generally compiled on the lines of the methodological principles of MTT, but it is much simplified if compared with the ECDs, fully consistent with MTT. As the proposed dictionary is meant not only “to boost the science of language” but also to satisfy practical needs, serving a particular public (linguists, discourse analysts, speechwriters, linguistic expertise specialists, etc.), it is “adapted to a particular level of understanding of its prospective users” (Mel'cuk 2006: 6). The aim of the authors is to test “the theoretical lexicon of a language”, successfully applied to compiling the ECDs of Russian and French (Mel'cuk & Zolkovsky 1984, Mel'cuk et al. 1992), on the data of the English language, to be more particular, of American political conflict discourse.

MTT postulates that any act of linguistic communication involves three major entities: meaning (a content to be communicated by linguistic signals), text (a complex signal to be used to communicate the content), natural language (a mapping between meanings and texts) (Mel'cuk & Zolkovsky 1984, Mel'cuk & Wierzbicka 2018). The sets of infinite meanings and infinite texts are contrasted to the finite number of correspondences between them. The postulates of MTT are applied to compiling the ECD, believed to be “a research tool, providing a standardized framework for the description of the lexical stock of any language” (Mel'cuk 2006, Mel'cuk & MiliCeviC 2020: 203). The ECD deals with strictly linguistic meaning: the literal meaning of lexemes, which can be arrived at solely on the basis of linguistic knowledge, without any reference to the extralinguistic context or common sense (the data for the present study are collected automatically and the information is presented exclusively from the viewpoint of text synthesis, enabling the user to pass from a given meaning to the corresponding text, which is of primary importance when analyzing conflict discourse). Any ECD entry presents a full description of a lexeme, comprising three major divisions that correspond to the triple nature of the linguistic sign, with the signified and the signifier taken in their Saussurean sense, and with syntactics denoting “a set of properties that control its cooccurrence with other signs” (Mel'cuk 2006: 20). As MTT considers meanings and texts to be formal objects that can be described by means of formal languages and specified by formal devices, it puts forward a system of special techniques, symbols, abbreviations, writing conventions (for detailed treatment see Mel'cuk & Milicevic 2020: 21-27), used for a rigorous and formalized description of the proposed ECD entry.

Thus, the microstructure of an ECD entry consists of three core zones, described in conformity with each other. The Semantic Zone contains the definition of the lexeme, its semantic label and connotations. The Phonological / Graphematic Zone specifies phonological features of the lexeme (its pronunciation, syllabification, and non-standard prosodic properties (if any)). The Cooccurrence Zone, divided into several sub-zones, presents combinatorial properties of the lexeme: 1) morphological (covering its inflectional paradigm), 2) syntactic (describing active and passive syntactic valence), 3) lexical (based on the notion of a lexical function, associated with semantic derivations of the lexeme paradigmatically and with its collocational potential syntagmatically), and

4) stylistic (including usage labels). The list of lexical functions includes those describing standard lexical paradigmatic relations: synonymy (Syn), antonymy (Anti), and conversion (Conv); derivational relations (S, A); syntagmatic lexical relations (adjectival functions (Magn, Ver, Bon), support verbs (Oper), phasal verbs (Incep, Fin, Cont), causative verbs (Caus, Liqu), some semantic derivations and several examples of complex lexical functions.

The source of the material for filling up the semantic and phonological zones as well as for stating the lexeme's morphological cooccurrence, stylistic cooccurrence and paradigmatic lexical functions is the dictionaries, while the lexeme's syntactic cooccurrence and syntagmatic lexical functions are studied using the corpus data of conflict political discourse. The use of the corpus data implies 1) checking the frequency of words in the corpus, 2) listing the collocations of the word, conventionally used together in conflict political discourse of the analyzed time span: N + the keyword, the keyword + N, Adj + the keyword, V+ the keyword (the output is a list of wordforms with significance scores for their cooccurrence with the keyword, ordered from high to low significance) (Fig. 5), 3) comparing the collocates of two words to see how they differ in meaning and usage (Fig. 6). The procedure helps the lexicographer “to sharpen the focus of definitions, highlighting salient facts and omitting remote possibilities, and to formulate explicit rules for choosing among near synonyms” (Kruyt 1995).

Figure 5. Fragment of the collocations list "SANCTION +Noun"

Figure 6. Fragment of comparing the keyword and its synonyms

An example of the simplified and user-oriented entry “SANCTION” as it is meant to appear in the ECD of English Political Conflict Lexis is presented below. The word has been chosen for the illustration as it is registered in all the dictionaries of conflict studies used for lexicographic analysis (Table 2). Though the Illustration Zone (exemplifying typical uses of the lexeme) is at times considered redundant, it is quite appropriate in the ECD of English Political Conflict Lexis as it makes it easier for the prospective user to understand a lexicographic description and substantiate the claims about conflict-provoking connotations of the lexeme; thus, the use of the entry lexeme in the corresponding zone (syntactic cooccurrence and syntagmatic lexical functions) is illustrated by actual sentences from the corpus.

SANCTION The semantic zone

Definition: sanctions of X against Y concerning Z for W - coercive action/s concerning Z taken by country/ies X against the country Y which is considered to have violated international law in W to end the violation Semantic label: coercive action

Connotation: punishment, ban, boycott, enforcement (negative, conflict- provoking)

The phonological zone

US /'sspkjbn/ UK /'sspkjbn/

The cooccurrence zone

(a) Morphological cooccurrence: Noun [C, usually plural], Politics;

(b) Stylistic cooccurrence (usage label): a conflict-provoking term used in political discourse, international law, and diplomatic discourse;

(c) Syntactic cooccurrence:

Active valence syntactic cooccurrence (Government Pattern) Though basically applying notations proposed by Mel'cuk, in the case of Government Pattern we use standard notations for parts of speech, their grammatical categories and syntactic functions accepted in Modern English Grammar (N - noun, com - common case, gen - genitive case, NP - noun phrase, NPabbr - abbreviated noun phrase, Pm - pronoun, Prnposs - possessive pronoun, A - adjective, o - zero degree of comparison, comp - comparative degree, superl - superlative degree, V - verb, Ved - participle II, Ving - participle I, gerund, VP - verb phrase, Vtrans - transitive verb, prep - preposition, etc. L stands for the entry lexeme).