A Critical Discourse Analysis of Intergroup Interpretations and Representations of Ideologies and National Identities in Sino-US News Reports of Russia-Ukraine Affair

: This study applies critical discourse analysis (CDA) and linguistic intergroup bias (LIB) theory to conduct a comparison of reports related to the Russia-Ukraine incident by Chinese media CGTN and US media CNN from late February to late March 2022. By analyzing their linguistic features and the intergroup psychological characteristics they project, the study explores the national identity embodied in these features, as well as the underlying ideological meaning and the power relations between different countries implied behind it. The results of the analysis show that: 1) discursive strategies are adopted by the newspapers to construct national identity and intergroup relations, including the discursive categorization of group identity and intragroup identity, and the discursive re-categorization of intergroup relations, 2) the similarities and differences in discursive strategies made by the Chinese and American media, 3) different ideologies and power relations reflected by these discursive strategies and the phenomenon of linguistic intergroup bias.


Introduction
The outbreak of this Russian-Ukrainian Affair has had a profound impact on the entire world. Russian-Ukrainian Affair has continued to dominate the news. There was extensive media coverage of the incident on both sides as both countries are deeply affected by the dispute, although China and the United States do not appear to be the focus of the current situation.
As public theaters, the media are one of the most important institutional venues for each national community to convey its shared experiences and reveal its underlying cultural and ideological assumptions [1]. In times of national crisis and conflict, a national community's discursive bond comes to the fore in media coverage [2]. Concurrently, national identity as a group rises to the level of ideology, subsequently making the nation the greatest criterion of distinction between in-group and out-group. And as a result of the presence of in-and out-groups, the phenomenon of linguistic intergroup bias also emerges.
As a consequence of the function of language in the social dimension, news discourse carries hidden ideologies while reporting on events and in turn constructs its events themselves. Moreover, the phenomenon of linguistic intergroup bias that occurs in media coverage is not only a manifestation of ideology but also enhances the propaganda nature of ideology. Thus, by examining how ideologies are reproduced in the press through discourse and how the specific phenomenon of intergroup linguistic bias reflects ideology, a deeper understanding of how ideologies, especially those related to national identity, are reproduced in society can be gained.
Therefore, this paper attempts to compare news coverage of Russia-Ukraine events by the Chinese and American news agencies CGTN and CNN from the end of February to the end of March. This study focuses on the similarities and contrasts between these two groups' approaches to reporting on the Russia-Ukraine events and how their intergroup prejudice reflects the divergent political ideologies of China and the United States. Discourse analysis would reveal the ideological differences that China and the United States manifested in the Russia-Ukraine affair by using linguistic intergroup bias (LIB) and critical discourse analysis (CDA) as theoretical foundations. By analyzing the linguistic features of the relevant news discourse, it is possible to explore the ideological meanings embodied or hidden in the news discourse's language and elaborate on the power relations implied behind it. It also adopts the complementary perspectives of CDA and LIB to suggest that the phenomenon of LIB may exist between national groups in international news.

Studies of Linguistic Intergroup Bias Theory
The linguistic intergroup bias (LIB) model proposes a systematic bias in language usage [3]. Maass found intergroup bias in the study of language abstraction levels in social interactions, as evidenced by individuals using language with higher levels of abstraction to describe positive behaviors of ingroup members and negative behaviors of out-group members, and using language with lower levels of abstraction to describe negative behaviors of in-group members and positive behaviors of out-group members [4]. In other words, people tend to use the more general abstract expressions to characterize positive things about their ingroup and negative things about their outgroup and more limited specific expressions to describe the opposites. Combining journalism and communication with LIB, Paulsen call on researchers to incorporate communication theories and approaches, such as CDA [5]. Xiong recognized that LIB cannot fully analyze the discourse bias that appears in the text in a dynamic and incoherent form, nor can it analyze the bias implicit in the discourse [6]. CDA, as an instrumental theory for analyzing the power relations and ideologies hidden in discourse, can be integrated with LIB to analyze the bias phenomenon in discourse. Based on the combination of CDA and Face-negotiation theories, Shan and Liu, through textual analysis of the studied corpus, pointed out that the U.S. media coverage of the Wenchuan earthquake was dominated by negative reports and supplemented by positive reports, and found that the negative and positive meanings characterized by intergroup linguistic bias were generated in the interactive process of maintaining self-image and seeking the identity of the other [7].

Critical Discourse Analysis
Critical discourse analysis (CDA) focuses on the application of discourse analysis to the critique of social practices [8]. It brings the critique of social analysis to the analysis of language use and features and focuses on relations between discourse and other social elements (power relations, ideologies, institutions, social identities, and so forth) [9]. The objects under investigation can be of paramount international importance to relatively small-scale ones concerning individuals regardless of their positivity or negativity.
For further analysis, a few fundamental terminologies in CDA needs to be clarified. First of all, CDA defines discourse as "a form of social practice" [10], suggesting that CDA views discourses as the language used in speech and writing that serves social interaction [11].
The second is the word "ideology". As critical discourse analyzers, it's crucial to be aware that a group's members typically utilize discourse that is ideologically prejudiced [8]. An ideological perspective can also be concealed by the use of words, or disguised behind the discourse.
Thirdly, apart from discourse and ideology, the term "power" is also a central concern in CDA. According to the summary of Wodak, "power is about relations of difference, and particularly about the effects of differences in social structures. [12]" The language is entwined in social power in several ways and provides a finely articulated vehicle for differences in power in hierarchical social structures [12].
In conclusion, the study of CDA aims to investigate power relations and covert ideologies as they are expressed in discourse by delving into the discourse's linguistic characteristics and the socio-historical and cultural settings in which it is entrenched.
From the 1990s, critical discourse analysis emerged and has begun to adopt the critical theories of sociology and ideology. Due to its problem-oriented approaches, CDA emphasizes the need for interdisciplinary work [11].
Fairclough's theoretical and linguistic foundation is based on sociology, social semiotics, and systemic functional linguistics. He proposes a three-dimensional model (description, interpretation, and explanation), which is widely accepted as the foundation of critical news analysis [13].
Scholars such as Hart and Chilton point out that cognitive linguistics can provide a useful analytical tool for CDA to enter the human mind and reveal mental representations and thought processes in political discourse, and specifically analyze the role of conceptual metaphors in CDA and specifically analyze the role of conceptual metaphors in CDA [14], [15].

Studies of Contrastive Analysis of News Report from China and the US
The comparative analysis of news reports from China and the US from the CDA perspective has been mostly conducted by Chinese scholars. Most of the studies concentrate on analyses of stylistics and language features of news reports and editorials.
Among the studies on political themes, Xin Bin also analyzed the practiced coverage of the South China Sea arbitration case from two perspectives: different types of speech and generic intertextuality [16][17]. Liu Hua and Liu Junwei compare and contrast the characteristics and mechanisms of multi-modal metaphorical language in COVID-19 news reports to analyze the similarities and differences in the characteristics between the two countries [18]. While Deng and Yang use the evaluation system as a framework to analyze the national image constructed by China and the United States in the report [18].
The above literature studies have mostly analyzed the ideology of news reporting only from the theoretical background of CDA, but have not focused on the psychological mechanism behind it. In LIB studies, although there are analyses of news discourse, few scholars have integrated LIB with CDA to study discursive strategies in news discourse more comprehensively. Therefore, this paper combines LIB and CDA as a theoretical framework to investigate the similarities and differences in discursive strategies, as well as linguistic intergroup bias reflected by ideology, in the coverage of the Russia-Ukraine affair between China and the United States from the standpoint of national identity.

Methodology
The study explores the parallels and contrasts in discursive methods used by CGTN and CNN in their coverage of the Russia-Ukraine affair, as well as the intergroup linguistic bias as a reflection of ideology by using the linguistic category model (LCM) and transitivity analysis from systematic functional grammar (SFG) by Halliday as an analytical tool. Through corpus software UAM Corpus Tool 3.3, a corpus is built for analyzing processes of transitivity in two groups of news reports collected.

Research Data
To ensure its comparability, the current study chooses 6 news report samples from CGTN and 3 from CNN based on the following criteria. First off, both news organizations serve as essentially formal and authoritative representatives of the different governments' positions. These news articles were taken from their respective official websites and are in English. Second, these news articles are all published at the same time and are about comparable, if not identical, news occurrences or news objects. Furthermore, the data are expected to be comparable in both length and content given the contrastive nature of this essay. This way, the study guarantees that the groups of news discourse are close and comparable in length.

Linguistic Category Model
Linguistic Intergroup Bias is theoretically based on the linguistic category model proposed by Semin and Fiedler [20], [21], [22]. The linguistic category model, which separates four degrees of abstraction in interpersonal terms in its original version, has been taken as the idea of language abstraction. According to this model, there are four levels of language abstraction from low to high: the most concrete descriptive action verbs (DAVs), interpretive action verbs (IAVs), state verbs (SVS), and the most abstract term, adjectives (ADJs). When one advances from the concrete to the abstract pole of the continuum, the interpretative implications change progressively. The more abstract the statement, the more information about the protagonist is offered, the more implicit temporal and cross-situational stability is provided, and the more difficult it is to check the statement's veridicality and conceive disconfirming cases [20]. Table 1 is an example of avoiding empirical ambiguity. Refer to a general disposition that applies across situations, behaviors, and objects B is an outgoing person

Transitivity
Halliday's systematic functional grammar has been considered his main contribution to linguistic research and as the methodological and analytical tool of CDA. According to Halliday, language has three main metafunctions: interpersonal, ideational (subdivided into empirical and logical), and textual [23]. Transitivity belongs to the empirical metafunction. Transitivity is concerned with the interpretation of a specific area of our experience, as a configuration of the process, the participants involved in it, and the circumstances that follow. There are six different processes: material process, mental process, verbal process, behavioral process, relational process, and existential process.

Material Process
Material processes are processes of doing, which are generally indicated by verbs of doing. They express the notion that some entities do things -things that may be done to some other entities and represent what is happening in the external world [24].

Mental Process
Mental processes are processes of sensing, represented as perception, reaction, and recognition [24].
There are generally two participants in the mental process, one is the subject of mental activity "perceiver", and the other is the perceived phenomenon.

Verbal Process
Verbal processes are processes of saying. The speaker is not necessarily a person in the verbal process, and the addressee can also appear as a participant. The verbiage may be a message to be delivered to the addressee, or the addressee may be asked to do something [25].

Relational Process
Relational processes are processes of being and having, which can be divided into two modes: attributive and identifying. These two modes of relational processes can be further classified as intensive, circumstantial, and possessive [24].

The Statistical Analysis of the Data
According to Halliday's categorization of transitivity, the following table shows the distribution of process in the two groups of news reports on the Russia-Ukraine affair.   Table 6 shows the occurrence of six processes in the four sample reports. It is noticeable that there is a dominance of material processes in both two groups. Relational process and verbal process also hold a part respectively in each group. A small or even no proportion is held by the behavioral and existential process.
On one hand, Fairclough states that "the system of transitivity makes options available, and which process type is chosen to signify a real process may be of cultural, political or ideological significance" [26]. In other words, the selection of process types by the producer of a text can result in political and ideological implications.
On the other hand, although the transitivity divides what happens in the objective world, the people and things involved and the environmental factors such as time and place associated with them into six kinds of processes, the participant and process in the transitivity system, which is the focus of attention, correspond precisely to the subject and predicate in the sentence. And these subjects and predicates can reflect the differences in linguistic abstraction in the framework of LCM and LIB, which in turn reflect the phenomenon of linguistic intergroup bias.
Therefore, the top-four processes, material, mental, verbal, and relational processes, and the phenomenon of linguistic intergroup bias within these four processes are concerned in the following discussion to find out ideological discourse strategies that describe group identity and intergroup relations.

Discursive Categorization of Group and Intragroup Identity
A core aspect of national identity is the attitudinal orientation of people who are in the sphere of national decision-making [27]. They are important for the ingroup because in the process of categorization they provide members with a common narrative in which to relate themselves to the nation, set boundaries that define what the ingroup is and is not, and legitimize claims and actions. Since the Russian-Ukrainian affair is inextricably linked to the political claims of each country, the subjects mentioned by both newspapers in the process, such as actors, sensers, and sayers, can be classified into five categories: Russia-related, Ukraine-related, China-related, America-related, and NATO-related. Several examples are listed as follows.
1) Russia has devised its countermeasures against the sanctions and black-listings. (Sample 1  from CGTN) 2) On the other hand, China hopes to see negotiations among the EU, Russia, and the Americaled NATO to create a balanced, effective, and sustainable European security framework in the interest and lasting security of Europe. (Sample 4 from CGTN) 3) Biden is willing to engage with Putin, even as he prepares withering economic sanctions should another Russian invasion of Ukraine transpire. (Sample 3 from CNN) 4) China's foreign minister spoke with his Ukrainian counterpart and said that China was "deeply grieved to see the conflict" and that its "fundamental position on the Ukraine issue is open, transparent and consistent." (Sample 1 from CNN) The phenomenon of linguistic intergroup bias is reflected in the way Chinese and American media portray each other, as 5,6,7, and 8 demonstrate.
5) On the other hand, China hopes to see negotiations among the EU, Russia, and the Americaled NATO to create a balanced, effective, and sustainable European security framework in the interest and lasting security of Europe. (Sample 4 from CGTN) 6) Chinese banks and companies also fear secondary sanctions if they deal with their Russian counterparts. (Sample 1 from CNN) In terms of China-related sensers in the mental processes, CGTN as an ingroup describes itself with the state verb "hope", while CNN uses the state verb "fear" to describe the outgroup member, Chinese banks, and companies' behavior, which is in line with the ingroup's use of highly abstract terms such as state verb to describe the positive behavior of the ingroup and the negative behavior of the outgroup in the phenomenon of linguistic intergroup bias. 7) Second, China has declined to join unilateral sanctions against Russia as Beijing makes foreign policy decisions based on the principle of the indivisibility of security and accommodates the legitimate security concerns of stakeholders. (Sample 3 from CGTN) 8) "Recognizing that China's economy and industrial output have been under enormous pressure in recent months, Chinese policymakers will likely attempt to strike a delicate balance between supporting Russia rhetorically but without antagonizing Western regulators," he added. (Sample 1 from CNN) In terms of the behaviors in the China-related material processes, CGTN as an ingroup uses lowlevel abstract action verbs to describe its negative behavior, while using an adverbial clause to explain in detail the reasons for the Chinese government's rejection. While CNN as an outgroup also uses low-level abstract action verbs and modified with a series of attributes and clauses to describe the positive behavior of Chinese policymakers, which is in line with the ingroup's use of lower abstract terms such as descriptive action verbs and interpretive action verbs to describe the negative behavior of the ingroup and the positive behavior of the outgroup in the phenomenon of linguistic intergroup bias as well.
As to the categorization of intragroup identity, CNN adopts a wider range of sources including government spokesmen or civilians, etc. as actors or sayers (shown in examples 9, 10), while CGTN tends to use the country, the city, or an authority figure as sayers, especially in Ukrainerelated material processes and verbal processes. 9) A long line of men snakes out of an unassuming building in the western Ukrainian city of Lviv on an overcast day. (Sample 2 from CNN) 10) Many businesses have answered a call by the region's authorities to join the war effortnamely by making Molotov cocktails. (Sample 2 from CNN) Verbal processes are processes of saying. The frequent choice of verbal processes and various types of speakers indicates the objectivity and reliability of news reporting. As shown in Table 8, CNN's verbal process accounts for a much larger percentage than CGTN's. Besides, it can also be noticed from Table 7 that speakers of CNN are more diversified than that of CGTN, in other words, CNN adopts a wider range of sources to add credibility to their reports. While CGTN utilizes the authority of sayers to increase the credibility of the reports.

Discursive Re-categorization of Intergroup Relations
The strategy of re-categorization was adopted by the newspapers to shift the cognitive boundaries of different national identities to reconfigure the intergroup dynamics between the nations.
Chinese and American media have different attitudes towards the Russia-Ukraine incident. As examples illustrated below, in the material processes, CNN is more directly pointing out the Russian invasion, while CGTN prefers to describe the Russia-Ukraine as a crisis or conflict rather than directly characterizing it as an invasion.

11) For sure, what is going on between Russia and Ukraine is not what China and the European Union (EU) expected. (Sample 3 from CGTN)
12) As the armed conflict in Ukraine enters the third week, the international community has made multiple efforts to reach a ceasefire through the United Nations (UN) and other diplomatic channels. (Sample 3 from CGTN) 13) A few days ago, this would have been an unfathomable prospect, but no one here says they can put anything past Putin anymore, especially after he invaded Europe's second-biggest country last week. (Sample 2 from CNN) 14) Russia's brutal invasion of Ukraine has so far left Lviv unscathed. (sample 2 from CNN) As shown in examples 11 and 14, CGTN uses relatively less abstract language, such as a descriptive action verb, to describe the Russia-Ukraine incident as conflict, while CNN uses relatively more abstract language, like interpretive action verb, to describe the Russia-Ukraine affair as an invasion. The different level of abstraction of language also indicates that the U.S. side classifies Russia as an out-group and uses high abstraction terms to describe its negative behavior. While China uses low abstract terms to describe Russia's behavior, its intergroup relationship with Russia is closer than that of the U.S. side.
Furthermore, in examples 15 and 16, the US media reclassifies its outgroup by using the terms "the two nations (states)", "ties", and "friendship". Even though China had no involvement in the Russia-Ukraine affair, it was nevertheless introduced and categorized as an ally or friend by the US media.
15) The two nations have forged close ties in recent years, with Chinese leader Xi Jinping calling Russian President Vladimir Putin his "best and bosom friend" in 2019. (Sample 1 from CNN) 16) During Putin's visit to Beijing last month, the two states proclaimed that their friendship has "no limits." (Sample 1 from CNN) In summary, these subtle linguistic intergroup bias differences and the re-categorizing discourse strategy reflect the different ideologies of China and the United States.

Major Findings
Although CDA and LIB have many applications in the analysis of news discourse respectively, there is a lack of research combining the two and examining how the Russian-Ukrainian affair is covered by the Chinese and American press from the perspective of national identity. This paper focuses on the discursive construction of ideology and national identity in the context of political events and linguistic intergroup bias reflected by ideology. By analyzing the reports collected from CGTN and CNN, the findings can be summarized as follows: First of all, due to the core aspect of national identity, the subjects mentioned by both newspapers in all kinds of processes, such as actors, sensers, and sayers, can be classified into five categories: Russia-related, Ukraine-related, China-related, America-related, and NATO-related.
Second, the way Chinese and American media portray one another reflects the phenomenon of linguistic intergroup bias. Lower abstract terms are used to describe the negative behavior of the ingroup and the positive behavior of the outgroup, while higher abstract terms are used to describe the positive behavior of the ingroup.
Thirdly, the many classifications of intragroup identification show the various methods of giving the reports more authority. To increase the trustworthiness of its information, CNN draws from a wider variety of sources. While CGTN depends on the credibility of the sources to strengthen the stories.
The US media reclassified its out-group by combining China and Russia as its outgroup, while the intergroup relationship between China and Russia presented by the Chinese media is closer than that between the US and Russia presented by the US media. These subtle lingreclassifiedroup bias differences between reports on CGTN and CNN and the re-categorizing discourse strategy reflect the different ideologies of China and the United States.
Overall, the similarities and differences in discursive strategies used by the Chinese and American media, as well as the various ideologies reflected by these strategies, are provided through the analyses of discursive features and the phenomenon of linguistic intergroup bias in news reports in two groups.

Limitations and Suggestions for Future Studies
Fowler points out that newspaper discourse is so complicated that concentrating on one aspect inevitably brings neglect of the others [28]. Consequently, there are some limitations to this study.
Due to the limitations of the research and the incomplete news event, the study's data set is limited in breadth. The study would be more thorough and compelling if it was based on a wide range of facts. The relationship between discourse, ideology, and power can also be interpreted using other language aspects like modality, thematic choice, and intertextuality. Thirdly, there is no quantitative methodology used in the study of LIB, and there is no evidence to substantiate the results. Additionally, as CGTN is the official Chinese media channel for foreign propaganda and CNN's readership may be more domestically oriented in the United States, the styles and audience goals of the two newspapers may have an effect on the findings of the analysis.
With the limitations mentioned above, it is expected that further research will be conducted with more convincing and reliable data to encompass all necessary theories and methods to explore the discursive strategies, as well as linguistic intergroup bias reflected by ideology from the standpoint of national identity.