The paper deals with the analysis of the communicative category of mitigation in German political discourse. The author provides a summarized overview of the theoretical background of the issue under study. The article is the second part of a complex research aimed at comparing ethno-cultural features of communicative mitigation in British and German political discourse. The relevance of the subject is due to the increasing politicization of the society, the growing scientific interest in the issues related to the interaction of language, culture and society, as well as in the study of communication and mitigation. This study aims to identify the means of realization of propositional mitigation in German political discourse and to determine the key principles of mitigation in German political sublanguage. The research was conducted on the basis of the Bundestag debates minutes posted on the official website of the German Bundestag. The material under study covers the period from November 8 to December 15, 2023. The total sample size is about 8.5 million characters, comprising 1 815 pages. The research is based on the following methods: contextual and linguocultural analysis, method of linguistic descriptions formalization, synchronous analysis of contexts and categories, as well as segmentation of certain verbal phenomena with their subsequent generalization, abstraction and characterization of ethnomental entities underlying them. At the first stage of the study, we compiled a matrix of propositional mitigation and identified lexical and grammatical indices of this type of mitigation. At the second stage we analyzed the material containing about 4 550 units of propositional mitigation collected with the help of continuous sampling. The paper provides instances of propositional mitigation for each of the means of propositional mitigation identified in the matrix. Most common appeared to be the mitigation expressed by the categorical mitigation modus: 2 525 units, 55 % of the total number, followed by the mitigation expressed through the assumption modus: 1 273 units, 28 %, and the least common proved to be the one expressed by means of supposition modus: 752 units, 17 %. The obtained results confirm the generally accepted characteristics of the political communicative style of Germans: suggestiveness, pragmatic categorization of political reality, and preventive approach to conflict resolution.

In his first play, Oscar Wilde turns to the image of a selflessly loving Russian girl Vera. Years later, in The Remarkable Rocket the author associates the theme of true love with a Russian girl’s image again. The purpose of the research is to reveal the resemblance of Vera Sabouroff and the Russian Princess at the level of their common semantic features in the following categories: physical appearance, attribute objects, elements of nature, colour and zoomorphic (ornithomorphic) symbolism. It has been revealed that Vera and the Russian Princess are similar in their essence. Both are compared to cold snow, have a resemblance to Snegurochka (the Snow Maiden) and, like her, are transformed by love. The two female images are characterized by Oscar Wilde with a combination of three colours: white, red, and gold. In both cases, the white colour (the implied snow-white one in Vera’s case) symbolizes the theme of innocence and the frozen state of human soul. The red colour contextually means self-sacrifice for the sake of love in Vera’s image and the soul’s awakening to love (as a feeling unknown before) in the Russian Princess’s case. The gold colour symbolizes the soul’s revival (as part of the morning sun metaphor) in Vera’s image and precious fidelity (the golden swan as a metaphor) in the Russian Princess’s case. While examining ornithomorphic symbols, conceptually close personal qualities of the two female images have been revealed: capacity for self-sacrifice (Vera as “nightingale”), fidelity and devotion (the Russian Princess’s swan). The plot and chronotope of the two works have a number of common features. It has been revealed that in both cases the implied moral message is identical: immortality of love and symbolic communion with true love as the sacrament of life. Thanks to the identification with Vera Sabouroff (who selflessly takes care of prisoners exiled to Siberia’s mines), the Russian Princess’s name and implied mental nature have been decoded. Russian Princess might have been a reference to the Decembrist’s wife and therefore may symbolize an ideal of unconditional female devotion. Closeness and sympathy of Oscar Wilde’s Irish mentality to the Russian mentality has been examined as the reason for his repeated appeal to the latter. The results of the research deepen our understanding of the two rarely studied female images in Oscar Wilde’s work and widen our understanding of the Russian culture influence on the world’s literature.

The article provides a systematic review of the application of eye-tracking in mental lexicon research, with a particular focus on the study of phraseological units and multiword expressions. The main eye-movement metrics used in experimental linguistics are examined: fixation duration, first fixation duration, fixation count, total reading time, regression rate, skipping probability, and other parameters that allow analyzing cognitive processing of lexical units in real time. Key effects of phraseological processing identified in foreign and domestic studies are analyzed: frequency effect, influence of conventionality, role of semantic transparency and compositionality, significance of contextual predictability. Special attention is paid to the comparison of conventional and occasionalized phraseological processing, as well as factors affecting the recognition speed of fixed expressions in different languages. A review of the mathematical apparatus of eye-tracking studies is presented, including methods of regression and correlation analysis, analysis of variance, Student’s t-test, non-parametric criteria, and logistic regression. Methodological issues of working with large eye-movement corpora, problems of metric reliability, and accounting for contextual dependence of eye-movement reactions are considered. It is shown that eye-tracking is a highly informative method for studying mental lexicon, providing access to unconscious processes of linguistic information processing and allowing verification of psycholinguistic models of lexical access and multiword expression representation.

The article addresses the problem of co-speech gesture alignment in the multimodal construal of first-person expository discourse. Adopting a cognitive semiotic framework, the study reveals the key semiotic dimensions of first-person expositive displaying its metaphoricity, entrenchment, indexicality, pragmaticity, agency, and transitivity in referent and event construal. It hypothesizes that despite the variance in co-speech gesture distribution mediated by different semiotic dimensions, there exist stable alignment patterns of complementary and redundant nature acting as their regulations. The research data are 20 recorded expository monologues containing 5603 co-speech gestures, with 725 first-person perspective utterances containing 1959 cases of co-speech gesture use in four groups, deictic, representational, pragmatic and adapters. The utterances were further explored to identify the complementary and redundant alignment patterns in the distribution of co-speech gestures aligned with metaphors, spatial deictic words and discourse markers, agentive and transitive verbs within the first-person expository discourse profile. The results show that while gesture iconicity displayed in representational and pragmatic gestures in co-speech gesture alignment is displayed in both complementary and redundant patterns, gesture indexicality displayed in deictic gestures is predominantly complementary. Additionally, the explored semiotic dimensions mediate the overall distribution of gestures within the utterance despite the amalgams of these dimensions, which alleviates the negative effects of their identification as synchronized with single cues in speech. Overall, the study paves the way to establishing the common cognitive semiotic grounds of multimodal alignment in discourse and to prognosticating this alignment as mediating by the distribution of semiotic dimensions.

The current article deals with stylistic functioning of the legal terms and related to law situations in Marcel Proust’s theoretical reasonings on “men-women”. In these reasonings, Marcel Proust resorts to the terms of the criminal law, criminal trial, to those of the financial law as well as to the figurative descriptions from the life of some representatives of legal professions, namely of an advocate or a notary. The methods used in the paper are those of semantical, lexical, syntactical and linguostylistic types of analysis. We have studied nine of such contexts selected from the first part of the fourth volume of the novel, numbering about thirty pages. The relations between two persons of the same sex, in accordance with the dominant views in Marcel Proust’s time, are often represented as a crime, so the vocabulary prevailing in these contexts is related to the criminal law and criminal trial. Drawing a parallel with a financial law situation, namely with the budget planning process, embodied in a medical metaphor, according to which a certain amount of liquid inside the human being has got to come out sometime, allows to represent men-women’s feelings as an objective reality in quantitative terms. In these fragments, legal terms are also often combined with different cultural allusions, for example to the ancient Greek myth of Pygmalion and Galatea, to one of the stories of the “Golden Legend” Christian collection, comparison of the men-women’s life with that of the poet as well as with the tragic destiny of Oscar Wilde; different kinds of arts are also mentioned, such as music schools, competing with each other not only due to their different views on music, but because of their opposite political views; here also belong judicial rhetoric and eloquence. Men-women’s societies are compared with different societies of art amateurs, as well as with some objectively existing social communities, for instance those of people suffering from a malady. Therefore, Marcel Proust, through the wide range of artistic expression means, shows that men-women’s predisposition is an objective inclination, given to them by the nature; they are guilty of nothing, they suffer from the society’s unfair attitude towards them just like an innocent person suffers because of a trial error.

The article is devoted to the study of stylistic features of the Lake School poets. The research methodology included several stages. Initially, a corpus of texts was formed, consisting of 99 poetic works by three authors belonging to this literary trend. At this stage, preliminary text processing and extraction of relevant features were carried out. Further, a model of text vectors space was applied to represent poetic texts. In order to analyze and identify the unique stylistic characteristics of each poet, a feature matrix was applied, analyzed using five common machine learning algorithms, including the support vector machine (SVM) method and Random Forest. To solve the problem of data imbalance, SMOTE (Synthetic Minority Oversampling Technique) was applied, which improved classification accuracy. In particular, the modified SVM model achieved an F1 measure exceeding 90 % in the task of recognizing the authorship of poetic texts. At the final stage of the study, the complex stylistic characteristics of the poetry by the three authors were compared, which revealed significant differences in their individual styles. The results of the study provide valuable information for further research in the field of poetry stylistic features.

The article proposes a solution to the issue of creating a system of critical questions applicable to the analysis of the argumentative component of texts. Critical questions underlie the assessment of the acceptability of the argumentation presented by the author and are aimed at checking all structural components of the argument. The proposed typology of critical questions and the method for their automatic generation using large language models are intended to expand the ontology implemented within the framework of D. Walton’s approach. This ontology contains a finite fixed set of typical argumentation schemes, which makes it possible to construct a complete finite set of critical questions. Each argumentation scheme in D. Walton’s compendium is presented as a set of premises (from 1 to 4), from which a conclusion is derived. Based on this, the typology of critical questions allows for three lines of attack: (1) challenging each of the premises of the scheme; (2) challenging the conclusion based on external data; (3) challenging the legitimacy of deriving a conclusion from the premises or the correctness of the description of the inference rule. To automatically generate critical questions, an experiment was conducted with the Claude 3.7 Sonnet model, which included two approaches to composing instructions (prompts): a) presenting substantive requirements for questions in a general form, which allows obtaining substantive specific and concretized questions for each scheme; b) creating a template for each type of question, resulting in the model that generates unified questions. The results of the experiments showed that the second approach provides a more accurate solution (77.6 % of correct questions), but the first strategy allows generating questions that are more specific for a given scheme. As a result, a complete set of critical questions has been created: 237 questions for 46 schemes. Based on the critical questions, attacking statements have been generated that can be generalized in the form of schemes allowing validation of an argument, which will solve the problem of argumentative analysis of polemics in dialogic and polylogical texts.

The paper proposes and tests a method for detecting stylistic change points in English-language fiction texts based on statistical analysis of a binary sequence taking into account the words belonging to the author’s invariant. A set of function words was considered as the author’s invariant; texts were displayed as sequences of zeros and ones used to study the stylistic features of texts. An empirical bridge was constructed based on this sequence. The maximum deviation of the empirical bridge served as statistical norm to detect possible change points. The method was tested on a corpus of 100 novels by British and American authors of the 19th–21st centuries, as well as on 9,000 pairwise combinations of concatenated texts. Threshold statistical values are identified that make it possible to distinguish homogeneous texts and texts with a possible stylistic change point. An approximation of the Kolmogorov distribution is used to assess statistical significance. P-values are calculated on the basis of the limiting Kolmogorov distribution. The results of the experiments show a stable ability of the method to identify a stylistic change point between the novels by different authors; the efficiency of the algorithm in distinguishing one text by one author and a combination of two texts by one author; the efficiency of the algorithm in distinguishing one text by one author and a pair of texts by different authors; independence of the method of the text length, that is, high and low values of statistics and p-values are observed for different text lengths. The analysis of empirical bridges and the behavior of individual function words show that stylistic features of the text can manifest themselves both on the scale of the entire work and within its individual parts. The method effectively records such changes. The developed approach can be used in the framework of corpus linguistics, applied stylistics, natural language processing, as well as in the creation of intelligent text analysis systems.

Automatic speech recognition (ASR) technologies have become a fundamental component of the backend infrastructure for voice and cognitive assistants, providing users with access to commercial services and government platforms. However, due to the uneven distribution of data on which ASR models are trained, the problem of model bias when processing atypical speech has become acute. The aim of this study is to assess the recognition quality and compare the level of inclusiveness of ASR models available on the Russian market while processing the speech of patients with aphasia. The material for testing was the new RuAphasiaBank dataset containing audio recordings of Russian-speaking patients with aphasia and neurotypical individuals. The dataset includes recordings from patients with different types of aphasia across three severity levels of speech impairment, as well as samples of different types of speech activity: dialogues, monologues, retellings, and reading aloud. For the evaluation of ASR systems, transcripts from 44 recordings of patients with mild and moderate comprehensive motor aphasia and 24 recordings from neurotypical informants were used. The standard WER and CER metrics of demo versions from four commercial ASR systems supporting Russian (Yandex SpeechKit, SaluteSpeech, Shopot, T-bank) were compared. All models demonstrated the highest number of errors when recognizing reading and retelling tasks. Monologues were recognized with the highest accuracy: Yandex SpeechKit showed the best performance (median WER 16–55%), while SaluteSpeech showed the worst (34–72%). The T-Bank model demonstrated the greatest stability in recognizing spontaneous speech in the control group (WER 28– 38%). The observed high WER with low CER indicates a lexical nature of recognition errors. The recognition quality of neurotypical and aphasic speech turned out to be comparable – the models generally struggle with processing natural conversational speech. For atypical speech, no clear leader could be identified, and the overall level of inclusiveness of ASR systems on the Russian market is assessed as low.

The article aims at describing the process of linguistic reconstruction of the collective memory of the Spanish conquest in the Latin American discursive and narrative space. The study is relevant due to the currency of the issue of the collective memory linguistic construction in the interests of the ruling elite, as well as the ongoing debates regarding the historical role of the Spanish conquest in the formation of a Latin American collective identity. The novelty of the paper lies in the socio-psychological approach used to identify linguistic means of reconstructing collective memory against the background of cultural trauma caused by the Spanish conquest through semiotic-narrative analysis. The purpose of the study is to identify linguistic means of reconstructing collective memory of the Spanish conquest as a traumatic event in the Latin American discursive and narrative space. Since collective memory is manifested and reconstructed exclusively through language interaction of people in discourse and narrative, which, in turn, are a space for transmission of common event experience, the study has two stages: 1) analysis of the reconstruction of collective memory in the Latin American media discourse; 2) analysis of the reconstruction of collective memory in the Latin American narrative space. Thus, it has been found that the socio-memorial framework of the collective memory of the conquest is formed by core memories of the unflattering facts of the Spanish conquest history, including violence against indigenous peoples, their discrimination, and theft of natural resources. On the periphery are facts of antisocial behaviour of indigenous tribes before the arrival of the conquistadors (cannibalism, violence against women and children, lack of adequate medical care and access to education), which contradict modern value systems and are subject to controlled collective oblivion in the process of linguistic reconstruction in the media discursive space. Journalists also strive to push the positive results of colonisation (development of education, medicine, culture, language) to the periphery of collective memory. However, the analysis of memory reconstruction in the narrative space suggests that Latin Americans remember and openly talk about the benefits of the conquest, which became the impetus for the development of an economically stable Latin American society. A comprehensive interdisciplinary approach to empirical evidence analysis makes it possible to examine the process of linguistic reconstruction of the collective memory of the Spanish conquest through memory preservation agents: 1) media and government structures that evaluate the importance of an event and determine the fund of the society’s memories; 2) members of the society who narrate a traumatic event. The prospect of the research is to consider the extra-linguistic objectivisation of the collective memory of the conquest, as well as to analyse the reconstruction of collective memory in the Spanish-speaking discursive and narrative space against the background of other traumatic events.

The expert essay offers a research response to the key provisions of the eco-cognitive model of simultaneous interpretation described in the monograph by A. V. Kolmogorova and E. V. Chistova, “Ecocognitive Approach to Language and Communication”. In the format of mirror critical reflections, the author presents a problematic overview of the innovative concept basic ideas in the perspective of their possible creative applications, considering the view on the translation process as a harmonious professional activity that can be studied using digital methods.

The review analyses the collection of the materials of the IV International Scientific Conference “Current Issues of the Mongolian and Altaic Studies” dedicated to the 85th anniversary of professor V. I. Rassadin, the 35th anniversary of the creation of the Tofalar script and the 25th anniversary of the Soyot script, held on October 23–26, 2024 at Kalmyk State University named after B.B. Gorodobikov (KalmGU), Elista. The publication was prepared by the Scientific Centre for Mongolian and Altaic Studies of KalmSU and is addressed to specialists in the field of Mongolian studies, Turkology and Altaic studies in general.

Since 1985, Iraida Yakovlevna Selyutina has successfully led a team of researchers who study the sound systems of languages and territorial dialects of Siberia and neighboring regions, as well as develop issues of phonetic typology, applying and improving the scientific ideas of her teacher, Vladimir Mikhailovich Nadelyаev. Her many years of work as a phonetician and a teacher have made an invaluable contribution to Russian humanities, and her tireless efforts as head of the main experimental phonetics laboratory have earned her the right to be regarded as one of the leading experts in her field, not only in Russian but also in international experimental phonetics. Iraida Yakovlevna continues to make a significant and invaluable personal contribution to the documentation, preservation, and research of the indigenous languages of Siberia and their sound systems.