An Analysis of Attitude Resources on Coverage concerning COVID-19 in New York Times

This paper adopts the attitudinal subsystem of appraisal theory to conduct a discourse analysis on news reports in the New York Times concerning COVID-19 in the initial stage. The result reveals that Appreciation resources account for the highest proportion in related coverage, while Affect and Judgement come in second and third place, respectively, with similar percentages. With regard to Affect, New York Times tends to use In/security and Dis/inclination; Judgement, capacity and Appreciation, Composition. From the perspective of the polarity of attitude, it can be seen that writers in the New York Times use both explicit and implicit attitude resources to express negative feelings toward the Chinese government’s response to this pandemic.


Introduction
The COVID-19 pandemic is still raging around the world, and the World Health Organization has declared this coronavirus an international public health emergency for its wide range of infections, rapid transmission, and great difficulty in prevention and treatment. Since the outbreak of this pandemic, China, under the strong leadership of the Communist Party of China, has attached great importance to and actively responded to it. And the measures and responses of our country against this pandemic have attracted wide attention from international media. In particular, the mainstream media in the United States began to report on a large scale on the different situations faced by China and the United States in the fight against COVID-19. The message and information grew explosively, which makes it difficult to distinguish their authenticity. It has become the key to relevant research on how to focus and analyze valuable discourse categories in the torrent of discourse to highlight its function of social construction and guide people to probe into the connotation of these seemingly objective coverages (Li, 2020).
Lai & Xin (2012) argued that journalistic discourse has two basic functions: message passing and public opinion control. In other words, those news reports are not as impartial and unmediated as they are claimed to be. They not only convey information and keep readers informed of events and their progress but are also concerned with the stances they constructed explicitly or implicitly in the news for actual or potential readers to accept. Thus, discourse analysis of coverage concerning the COVID-19 pandemic from American mainstream media is conducive to disclosing the ideology infused in the language. This paper adopts the Attitude subsystem of the Appraisal Theory as the theoretical framework to examine the attitudinal resources presented in the New York Times news reports on the COVID-19 epidemic in the initial stage to reveal its attitude towards this emergency event and China's response to it.

A Brief Introduction of Attitude System in Appraisal Theory
Appraisal theory is located within the framework of Systemic Functional Linguistics (SFL). It refers to the linguistic resources by which writers/speakers come to express, negotiate and naturalize particular inter-subjective and ultimately ideological positions. And this theory divides evaluative resources into three broad semantic domains: Attitude, Engagement, and Graduation. The Attitude subsystem adopted in this paper is concerned with our feelings, including emotional reactions, judgments of behaviour, and evaluation of things by which speakers pass judgments and associate emotional/affectual responses with participants and processes. Attitude is itself divided into three semantic regions of feeling, Affect, Judgement, and Appreciation (Martin & White, 2005).
Affect is concerned with expressing positive and negative feelings towards people, people's behaviour, and phenomenon. It can be further divided into four major sets having to do with dis/inclination, un/happiness, in/security, and dis/satisfaction. Judgement deals with attitudes towards behavior, which we admire or criticize, praise or condemn. In general terms, judgement can be divided into social esteem and social sanction. The former has to do with normality, capacity, and tenacity, while social sanction has to do with veracity and propriety. With appreciation, we can turn to meanings construing our feelings of things we make and performances we give; natural phenomena could also be included. This term can be divided into our "reaction" to things or phenomena, their "composition", and their "valuation". Each of these Attitude resources contains positive and negative evaluative meanings and always involves grading the depth of feeling.
The detailed scheme of the Attitude system adopted as the theoretical framework of this paper is outlined in Figure 2 Appraisal theory was first applied as a linguistic tool in journalistic discourse. Besides journalistic discourse, other scholars also utilize Appraisal Theory to analyze other discourse genres such as legal discourse ( Korner, 2020), secondary school history texts ( Based on the Appraisal Theory, this paper is aimed at analyzing the attitudinal resources implied in the news reports of the New York Times concerning COVID-19 in the initial stage to see how writers use linguistic strategies to state their stance and balance the relationship between power and solidarity by persuading readers to align with writers.

Research Methodology
All the detailed information on where and how to collect the data, how to process the data, and the corpus tool are presented in this part. A combination of quantitative and qualitative methods will be employed to conduct the analysis of attitudinal resources.

Research Instrument
The corpus tool for annotation employed by this paper is the UAM Corpus Tool (version 3.3), developed by computational linguist Mick O'Donnell. It can be used in the linguistic field to annotate different corpora by adding layers and getting statistics needed by a scheme.

Data Collection
The coverage displayed in this research is from the official website of The New York Times. The time span of this research would be chosen from January 1, 2020, to January 31, 2020, the first period or initial stage of the COVID-19 outbreak in China. In all, 29 pieces of news reports are selected. This small corpus includes 31778 tokens, 4693 types, and 4097 lemmas.

Data Processing Procedures
For this study, four steps are involved in annotating the corpus and getting the final results with the help of the UAM Corpus Tool (version 3.3).
First, collect data in TXT files because the file format is the only compatible form to this tool and put them into this instrument to establish a new project; second, edit a theoretical scheme for annotation. Either manual or automatic layer is available according to the research need; third, add an annotation. Manually annotate the file within the framework of Attitude sentence by sentence, choose the right layer for semantic units according to the context to ensure accurate annotation of all attitudinal resources in the news discourse; fourth, click the "statistic" button to describe the file and choose the final result based on need. The detailed scheme used for annotation is displayed in Figure 2.1. Note to passing that the annotating process was repeated and discussed multiple times by two coders.

Results and Discussion
Both an overview and detailed analysis of attitudinal resources employed by the New York Times are presented in this section.

An Overview of Attitudinal Resources in The New York Times
By the convenience of the UAM Corpus Tool (version 3.3), an overview of all attitudinal resources in news reports concerning COVID-19 from the New York Times in January is presented in Table 4.1, and the distribution of attitude polarity is also presented in Table 4.2. Besides the affect, judgement, and appreciation, the attitude of news discourses can also be realized through attitude polarity that mainly includes positive and negative attitudes. But sometimes, it can be neutral or ambiguous when it is not clear whether the attitude presented in the text is positive or negative. The distribution of attitude polarity in news reports on COVID-19 from The New York Times is shown in Table 4.2. According to Table 4.2, it can be found that the number of negative attitudes far outweighs positive and neutral attitudes, with 468 items taking up 70.58% is more than triple from 150 items of positive attitudes occupying 22.59%. Only 47 items of neutral attitude applied in the news discourse, whose proportion is too small to be discussed here. Therefore, it can be concluded that an overwhelmingly negative attitude pervaded the New York Times coverage concerning COVID-19 in January.

Detailed analysis of Affect, Judgement and Appreciation in The New York Times
With the assistance of the UAM Corpus Tool (version 3.3), the detailed results of frequencies of three regions of feelings, affect, judgement, and appreciation, on coverage concerning COVID-19 in the initial stage from the New York Times will be concentrated in this section.

Analysis of Affect Resources
The system of Attitude involves three semantic regions covering affect, judgement, and appreciation. Affect, as the core component of Attitude, consists of four major sets: dis/inclination, un/happiness, in/security, and dis/satisfaction. Dis/inclination refers to feelings involving intention. Un/happiness involves the mood of feeling happy or sad. In/security covers people's feelings of peace and anxiety in relation to their environments. Dis/satisfaction deals with feelings of achievement and frustration in relation to the activities in which we are engaged, including our roles as both participants and spectators.
The detailed distribution of affect resources on COVID-19-related coverage in The New York Times is shown in Table 4.3.  Table 4.3, in/security is the most frequently employed among the four types of affect resources, with a number of 87 items and a frequency of 46.28%. The application of dis/inclination ranks second in these news reports, occurring 77 times and accounting for 40.96%. 15 items of un/happiness occupy 7.98%, while dis/satisfaction occupies the least proportion with the number of 9 items and the frequency of 4.79%.
With the assistance of the "Search" function in the UAM Corpus Tool (version 3.3), all lexical and clausal items annotated under specific layers and their concordances are listed clearly.
In terms of in/security, the most frequently used expression include "concern", "anxiety," and "trust". The former two vocabularies illustrate that writers tend to portray the panic of Chinese people, for they don't know how to respond to this novel coronavirus.
In the initial stage of this pandemic, people had no inkling of its source, its symptoms, whether it would be more severe than SARS 17 years ago, or how this epidemic would develop. The latter "trust" seems to be a positive word at the level of lexical meaning alone. However, when it comes to the context, two major negative attitudes can be summed up: first, to express the distrust of Chinese people and the international community on the Chinese government's response to this mysterious coronavirus compared to its performance in the fight against SARS by presenting words of a few people who told the reporter they were at the epicenter of this pandemic; second, reporters of The New York Times tend to express people's distrust to the authority of the World Health Organization for it praised China for its rapid response to the pandemic and its transparency and promptness in sharing information. The author condemns the WHO for not fulfilling its responsibility to conduct a thorough investigation of the outbreak of COVID-19 and for blindly believing the information provided by the Chinese government. In a word, writers of The New York Times tend to adopt in/security resources to express people's worry about the impact of the pandemic on their life and health and the government's inability to control this pandemic.
As for dis/inclination, the most frequently employed expressions are "fear" and "panic". Both of them indicate people's negative attitudes towards the future evolution of COVID-19 for the lack of knowledge of this novel coronavirus in the initial stage, and the government's misjudgement of the outbreak threw people into a panic.
With regard to the remaining two affect resources, also the least applied in COVID-19-related new reports, un/happiness resources are used to express hatred of Chinese or Asian people by foreigners and anger toward the government's "cruel" prevention and control measures against the pandemic. Dis/satisfaction resources are employed to express frustration.
To conclude, because of the insufficient knowledge of this novel coronavirus in the initial stage, it is inevitable for Chinese people and the international community to be anxious about this mysterious pneumonia. The Chinese government would also ineluctably take a lot of time to find effective measures to curb the spread of the outbreak. However, the application of affect resources in coverage of The New York Times demonstrates that the writers prefer to magnify people's fear and anxiety, the messy society, and the incompetence of governmental agencies.

Analysis of Judgement Resources
The data of judgement resources in news reports from The New York Times is shown in Table 4.4. All these judgement resources can be concluded into two categories: one for the behavior of governments and the other for judging some individuals.
As is shown in the table, capacity resource accounts for the highest proportion among the five types of judgement resources with the number of 77 items, taking 51.7%, followed by propriety and veracity with 26 and 25 items, occupying 17.43% and 16.78%, respectively. The proportion of the remaining two, normality and tenacity, is 6.71% and 7.38%, respectively. The last two judgement resources won't be analyzed in this paper for their small proportion. Capacity is mainly related to how capable someone can be. Its attitude polarity is presented as below: As neutral/ambiguous accounts for a very small proportion and it makes no difference to the analysis, it will not be discussed here. It is unexpected that a negative attitude toward capacity is not overwhelmingly dominant. However, combined with specific contexts, all these positive capacity resources can be summarized as four categories: 1) the positive evaluation of Chinese scientists; 2) compared with the behavior of the Chinese government 17 years ago, it did do a better job to respond to such a public health emergency this time; 3) the Chinese government was more capable of concealing the real pandemic situation and controlling the people's freedom of speech online than the pandemic; 4) the WHO's confidence and praise to China's capacity to control the outbreak.
And almost all the negative capacity is the criticism of the Chinese people and the international community for the Chinese government's incompetence in handling the outbreak in its initial stage.
Veracity is concerned with how truthful, or honest someone is. This resource applied by The New York Times is all used to condemn that the Chinese government withheld critical information about this pandemic, which depicts the government as a liar. And then, combined with propriety resources (how ethical someone is), the Chinese government is portrayed as a dictatorial monster, and China is a country with no freedom of speech in the initial stage.
All in all, reporters of The New York Times prefer to use judgement resources to criticize or condemn the Chinese government's inability to deal with the outbreak and prevent it from spreading abroad and covering up the related information, including underreporting cases of illness and death and whether it can be transmitted human-to-human. Reporters prefer to cite statements online from individuals who declared to be at the epicenter of the outbreak to increase the authenticity of their news.

Analysis on Appreciation Resources
The overall distribution of appreciation resources in The New York Times on COVID-19 is displayed in Table 4.6. According to Table 4.6, we can find the appreciation resources in the news reports in January are ample. Composition is the most frequently used among the three types of judgement resources, appearing in 122 items, taking up 37.2%. It is mainly connected with the balance and complexity of entities and man-made phenomena. Only the negative polarity of composition will be discussed in this section because most of the positive and neutral/ambiguous polarity of composition have no direct relevance to the outbreak.
In The New York Times news reports, the composition is mainly used to evaluate the characteristics of this pneumonia and measures to control it. In the initial stage of this outbreak, both government and media lacked sufficient knowledge of this novel coronavirus; the most frequently employed words include "pneumonia-like" and "deadly". Is this transmitted through animal-to-human or human-to-human? How fast and widespread would it spread? What's the condition of the pandemic at home and abroad, severe or mild? The composition resources can answer these questions. Besides the outbreak, this judgement resource is also used to evaluate measures to prevent and control this pandemic.
Reaction refers to things, whether they can attract people's attention or please someone. Obviously, people won't be pleased by this deadly pandemic, the positive polarity of reaction resources is of a small number. Similar to composition, writers use the reaction to describe the qualities of COVID-19. The difference between them is that reaction focus on the negative impact of this pandemic on human health and social and economic development, such as "harmful", "contagious," and "serious".
With regard to the valuation of resources, it is related to evaluating whether something is worthwhile. In the initial stage of this pandemic, this pneumonia-like illness is new, so "new", "novel," and "first known" are expressions used to emphasize its distinctness from other pneumonia.
Note that phrases such as "China virus" or "Chinese virus" appearing altogether 10 times in the selected news discourses are used to indicate that China is the source and transmission of the virus without any scientific research and evidence. The New York Times has demonized China by insinuating "China responsibility" and "China threat" and constantly emphasizes that the spread of COVID-19 around the world is directly related to China.

Conclusion
Under the guidance of the Attitude subsystem in the Appraisal Theory and with the assistance of the UAM Corpus Tool (version 3.3), this research makes an analysis of the distribution and frequency of attitudinal resources in news coverage concerning COVID-19 in the initial stage from The New York Times by means of the quantitative and qualitative analysis to investigate what is the attitude of the US mainstream media towards the COVID-19 pandemic and the Chinese government's response to it.
According to statistical analysis, negative attitudinal resources are suffused in the news coverage about COVID-19 in the initial stage from The New York Times. In/security, capacity and composition are the most employed resources in Affect, Judgment, and Appreciation, respectively. It can be concluded that western media, represented by The New York Times, fall over themselves to defame the Chinese government by making use of this pandemic. By focusing on individual extreme cases, it can make a general conclusion to criticize and condemn the Chinese government. Their seemingly objective news discourse is often laced with various biases and anti-Chinese ideologies. In the American news reports, the Chinese government is described as an incompetent and irresponsible authority. The central government is depicted as a tyrannic monster that would curb the spread of this coronavirus at all costs, while the local officials only care about their political achievement, regardless of people's well-being. Sometimes, their evidence just comes from the statement of only one person.