Who Is to Blame? : An Ecolinguistic Analysis of the Portrayal of Human and Non-Human Animals in the Initial Phase of the Corona Crisis

(1)

Linköping University | Department of Culture & Society | English

Linköpings universitet | Institutionen för kultur och samhälle | Engelska

Bachelor’s Programme in Language, Literature & Media | English Thesis, 15 credits

Kandidatprogram i Språk, litteratur och medier | Självständigt arbete i Engelska, 15 hp

Spring Term 2021

Vårterminen 2021

Who Is to Blame?

– An Ecolinguistic Analysis of the Portrayal of Human and

Non-Human Animals in the Initial Phase of the Corona Crisis

Vem ska beskyllas?

– en ekolingvistisk analys av skildringen av mänskliga och

icke-mänskliga djur i den inledande fasen av coronakrisen

Rebecca Wikström

Supervisor/Handledare: Nigel Musk Examiner/Examinator: Michael Smith

Linköping University

Linköpings universitet

SE-581 83 Linköping, Sweden 013-28 10 00, www.liu.se

(2)

2

1. Introduction

As I write this, we are in the middle of a global crisis. The corona virus has spread steadily and led to consequences on a larger scale than anyone could have imagined when it started over a year ago. It is not at all surprising that we want clear answers and that we even want to find someone to hold responsible. Who is to blame for this terrible situation that we have to live through?

Naturally, this question does not have one simple answer, but in the light of it, certain linguistic observations can be made. In news reporting, it is not rare to hear statements such as “bats spread Covid-19 to humans” or “infected animals must be put down”. At the same time, it seems it is not as common to explicitly talk about humans as agents in spreading the virus even though, through our actions, we arguably have a great responsibility in it. Many of us are aware of this, but the question is whether it is reflected in the language we consume on a daily basis. Could it be that humans are linguistically erased from the context, while animals are being blamed? These observations have awoken the urge and driving force to carry out a study that investigates linguistically whether they can be found throughout texts. This is important to uncover because the way we speak and write of elements of the world has effect upon it. As Fairclough phrases it:

[…] it is a feature of the social world that interpretations and explanations of it can have an effect upon it [and] can transform it in various ways. A critique of some area of social life must therefore be in part of a critique of interpretations and explanations of social life. (2010: 8)

With this claim in mind, language can be thought of as having the power to affect our ways of viewing the world, and thereby, the power to create what we think of as “reality”. Thus, it is crucial to uncover how we speak of humans and non-humans in the context of the corona crisis in language we consume regularly.

To perform the study, 15 articles from five different British newspapers have been processed through four different lenses: facticity, appraisal, erasure and salience. These categories have been developed primarily drawing on Arran Stibbe’s ecolinguistics (2021) and have the potential of meeting the aim of this study, namely to uncover linguistic blame by answering the following questions:

• How did the five newspapers linguistically portray non-human and human animals in the initial phase of the corona crisis?

• Based on the linguistic portrayal of non-human and human animals, who is constructed as being to blame for the corona crisis?

(4)

4

2. Theoretical Background

This chapter gives an overview of Critical Discourse Analysis and ecolinguistics, which are the primary theoretical approaches this study makes use of. It also presents previous linguistic studies set in the context of Covid-19, as well as studies on linguistic blame.

2.1 Critical Discourse Analysis

Critical Discourse Analysis (CDA) can be used across many different disciplines, but its general aim is to uncover power and ideology. While general linguistics and regular discourse analysis focus on

describing linguistic features, CDA aims to unravel “why and how these features are produced and

what possible ideological goals they might serve” (Machin and Mayer 2011: 4, emphasis in original). Norman Fairclough writes that

CDA is a form of critical social analysis. Critical social analysis shows how forms of social life can damage people unnecessarily, but also how they can be changed. CDA’s contribution is elucidating how discourse is related to other social elements (power, ideologies, institutions, etc.) and offering critique of discourse as a way into wider critique of social reality. (2018: 13)

This means that CDA is well suited to uncover how discourse(s) throughout texts is/are related to, affected by and has/have an effect upon the social context that exists beyond them. Wodak claims that in a changing world many new social issues appear that inquire understanding, and “Critical Discourse Analysis offers a program for research on such socially relevant phenomena” (1999: 185). She also identifies one very common misunderstanding of CDA, namely that critical would be synonymous to negative. She claims that ”[c]ritical does not mean detecting only the negative sides of social interaction and processes and painting a black and white picture of societies” (ibid.:186). On the contrary, it means searching for transparent and dichotomous explanations of the social world (ibid.).

According to Machin and Mayer, there is no such thing as an isolated, homogeneous form of CDA (2012: 4). Instead, there are many different ways of doing CDA, but what they all have in common is “the view of language as a means of social construction: language both shapes and is

shaped by society” (ibid., italics in original). Fairclough states that his goal with CDA is, and has

always been, to advance means of investigating how language is involved “in the workings of contemporary capitalist societies” (2010: 1). However, other researchers use CDA to uncover different phenomena than capitalism, for example racism and sexism. It is also applicable when investigating the relationship between human and non-human nature and animals – something that ecolinguistics has made use of.

(5)

5

2.2 Ecolinguistics

Ecolinguistics is a branch of linguistics that deals with the relationship between language and ecology, or “how language is involved in forming, maintaining, influencing or destroying relationships between humans, other life forms and the environment” (Stibbe and Alexander 2014: 1). Since the rise of CDA, many researchers in language have seen eco-criticism as an additional and further expansion of the field (Fill and Mühlhäusler 2001: 6); however, there is one fundamental difference between CDA and ecolinguistics. CDA focuses exclusively on criticising discourse, or parole – the pragmatics, syntax and the words – of both spoken and written language. Ecolinguistics does this too, but it also criticises the language system, or langue, “which in many instances favours an unecological fragmentation and a separation of humans from the rest of animate and inanimate beings” (ibid.). Ecolinguistics can define the anthropocentric way of depicting the world through language. That is,

using language with a “[p]rimary or exclusive focus on humanity; the view or belief that humanity is

the central or most important element of existence”, as opposed to the non-human world (Oxford English Dictionary 2021).

Although ecolinguistics is a quite young branch in the study of language, the journey to what defines it today started in 1912 when Edward Sapir made an early attempt to “establish a relationship between ‘Nature’ and language” (Fill and Mühlhäusler 2001: 2). In his text, Sapir talks about the “environment” as divided into social and physical environments. While physical environments consist of geographical aspects such as climate, flora and fauna, social environments consist of societal forces such as religion, politics and art, “that mold the life and thoughts of each individual” (2001 [1912]: 14). Sapir claims that “the physical environment is reflected in language only in so far as it has been influenced by social factors” (ibid.). That is, the way we describe the physical environment is affected by the social environment we live in. The latter could therefore be said to have a responsibility in how to linguistically construct the former.

Einar Haugen has often been labelled as one of the earliest researchers who defined what ecolinguistics is today. According to Fill and Mühlhäusler, “[w]hen Einar Haugen created the paradigm of 'the ecology of language' in 1970, he was referring to a new ecological study of the interrelations between languages in both the human mind and in multilingual communities” (2001: 1). In the decades to come, several branches of linguistics discovered and made use of this relationship to investigate a range of ecological parameters, and in the early 1990s, when ”all the different approaches which in some way link[ed] the study of language with ecology were brought together, […] a unified - though still diverse - branch of linguistics was established” (ibid.) – ecolinguistics.

(6)

6 According to Arran Stibbe, the initiator of The International Ecolinguistics Association and, arguably, one of the most prominent researchers in the field of ecolinguistics in contemporary times, “[e]colinguistics point out that mainstream linguistics, in its focus on the role of language in human-human interaction, has erased the interaction of human-humans with the larger ecosystems that support life” (2021: 161). Stibbe defines ecolinguistics as a branch of linguistics that “can explore the more general patterns of language that influence how people both think about, and treat, the world [as well as investigate] mental models which influence behaviour and lie at the heart of the ecological challenges we are facing” (ibid.: 1). He calls these mental models “stories we live by” (ibid.). With this description in mind, Stibbe’s ecolinguistics is fit to unravel how the portrayal of humans and non-humans influences how we conceptualise and see these entities in the context of the corona crisis, and how that influences who we blame for it. It is therefore to be considered the primary source of theoretical and methodological inspiration for this study.

Both Critical Discourse Analysis and ecolinguistics aim to uncover possible meaning embedded in language and both methods are affected by the analyst’s own ideological bias. Truth and reality are subjective, and what is considered positive and negative, beneficial and destructive portrayal can be very different to different analysts. Stibbe writes that

all critical language analysts have an ethical framework that they use for evaluating the language they are analysing, whether or not it is made explicit. An analysis of racist language, for instance, is likely to be conducted within a framework which sees racism as something negative that needs to be worked against rather than just an object for disinterested analysis of the technicalities of language. (2021: 11)

Although this study does not explicitly define an ethical framework as such, it does come with the assumption that anthropocentric language should be criticised. Human and non-human animals should be treated with the same amount of respect, the reason being that both are sentient beings who are directly and indirectly affected by human actions. On the issue of objectivity, Fairclough writes that, “the critical analyst, in producing different interpretations and explanations of the area of social life, is also producing discourse” (2010: 8). The only way of making the analysists own interpretations and explanations superior to the discourse they investigate and critique, is by “providing explanations which have greater explanatory power” (ibid.). Thus, neither CDA nor ecolinguistics claim to describe an objective truth or reality, and neither does this study. However, it does describe one possible reality by thoroughly explaining how the discourse(s) in the texts could convey how we portray and blame humans and non-humans in the context of the corona crisis.

(7)

7

2.3 Previous Studies

There are hitherto relatively few studies that have been carried out and published situated in the context of the corona crisis, the reason being that it still is a rather young phenomenon. However, in the linguistic field, some studies can be found. Khusnul Khotimah et. al. (2021) carry out a study based on Systemic Functional Linguistics and ecolinguistics to describe new lingual expressions related to ecology and health, or what they call “health ecolexicons”, in the context of Covid-19. They do this by looking at Indonesian media coverage and come to the conclusion that “[t]he lingual expression of the health ecolexicons has got the forms of synonyms, acronyms, foreign terms, and Indonesian and English abbreviated forms” (ibid.: 324). Through this study, they contribute with material situated in the corona crisis to the ecolinguistic field, but the study does not contain a critical investigation of either portrayal or blame.

In another study, Khaled Nasser Ali Al-Mwzaiji (2021) investigates the depiction of the presumed origin of the virus, in particular conspiracy theories about the virus originating from a laboratory, using CDA. In the study, speeches by U.S and Chinese politicians are analysed to examine how three scenarios are linguistically portrayed by politicians. Al-Mwzaiji finds that when the politicians refer to any other sources of origin than the laboratory scenario and the bat scenario, they use “two lexical choices – ‘believe’ and ‘strange’” and that “[b]oth of the words shift the semantic emphasis from scientific reasoning to unsubstantial assumptions” (ibid.: 248). Al-Mwzaiji concludes that the politicians base their speeches on groundless hypothetical conclusions from not yet finished investigations about the laboratory origin and that “the speculative assertions are inspired by implicit political and ideological interests of representation, domination and exercise of power” (ibid.). Thus, it investigates both blame and the crucial relation between language and power. However, it does not touch on any ecological aspects of this relationship, and thus it fails to comment on a possible ecological consequence of the political discourse.

Ryan Sowden et. al. carry out “a qualitative document analysis of all articles that described grief or bereavement after a death from COVID-19” in a corpus based on the British press (2021: 1). For this study, Critical Discourse Analysis and Terror Management Theory are used and they find that the texts are dominated by three main narratives: “sensationalist language” reflecting helplessness, fear and confusion; the “use of war metaphors” to deal with fear and uncertainty; and “euphemistic or glorifying language” that works to give grief a voice (ibid.: 13). By investigating this, the portrayal of one of the primary phenomena that the corona crisis has come to bear the stamp of is defined. However, since it does not describe the portrayal of blame – another prominent phenomenon throughout the corona crisis – it leaves a gap to fill.

(8)

8 On the subject of linguistic blame Prażmo (2020) executes a study which investigates how women (and sometimes men) are being blamed in so called “incel” fora (fora where men who are involuntary celibate write). Prażmo finds that the men writing in the fora linguistically blame their celibacy on women and other (incel) men by using different cognitive dehumanising (animate and non-animate) metaphors. By investigating this, the study describes the common and significant issue of hate-speech. However, it lacks a critical discussion on the relationship between human and non-human animals, because it presupposes that animals and non-humans are fundamentally different. Prażmo does not criticise the fact that animal metaphors are used to negatively appraise; thus, the study fails to comment on why being compared to an animal is consistently considered negative.

Another study on linguistic blame is performed by Wongnuch, Boonmongkon and Guadamuz (2019). They look at the discourse related to dengue fever and find that landless bamboo farmers in a village in Thailand are being blamed for the spread of it. The study “demonstrates that expert knowledge and practice were factors in naming and blaming the people even in periods without infection” (ibid.: 142). By investigating discourse generated by health authorities and government, it looks both at blame in a context of the spread of a virus and the relationship between language and power. However, it does not touch on the ecological aspects and consequences in blaming landless farmers instead of the landlords (which the farmers are economically dependent of) – something that an ecolinguistic approach could do.

To my knowledge, no ecolinguistic analysis of the linguistic portrayal and blame of human and non-human animals in the context of the corona crisis has hitherto been performed. Through this study, I wish to contribute to the field by filling this gap.

(9)

9

3. Data and Methodology

3.1 Sources of Data

Data from five different electronic newspapers were collected to create a corpus. Articles were collected from The Guardian and The Telegraph (which are broadsheet newspapers), The Daily Mail and The Daily Star (which are tabloid newspapers) and BBC News (which is a public service newspaper). These newspapers were chosen because they are well known, which means they have the power to influence a large quantity of people, and they represent different stances on the political spectrum.

3.2 Gathering of Data

The articles were gathered by five different searches on Google. The time of publication was set to 1 February – 31 May 2020 to narrow down the material to articles which were published in the initial phase of the corona crisis. The reason for choosing this specific timeframe was that the observations which inspired this study were made in that period of time. The texts were gathered by searching for the newspaper in question combined with the words “Covid-19” and “bats”. For example, the search could be phrased “The Guardian Covid-19 bats” or “The Daily Mail Covid-19 bats”. These particular search words were selected in order to limit the scope to articles that focus on the link between Covid-19 and bats, with the assumption that such articles contain relevant material to answer the research questions.

The first three articles in the search field published by the relevant newspaper, omitting results published by other sources, were then collected and put into a Microsoft Word document. Two exceptions were made: the third text from The Telegraph was omitted because it did not focus on the connection between animals and Covid-19, and the first text from The Daily Mail was omitted because it was too short to contribute to the analysis. Since this study aims to investigate linguistic elements of the texts, not other semiotic resources, all pictures were removed from the data. The articles were then closely read in search of language features of relevance for the analysis.

3.3 Methodology

The methodology of this paper is qualitative interpretive. To carry out the study, four categories were developed primarily drawing on four of the nine "stories we live by” (evaluations (in this study called appraisals), convictions (in this study called facticity), erasure and salience) that are the base of Arran Stibbe’s Ecolinguistics and the Stories We Live by (2021). To further define the framework, concepts and linguistic features were also drawn from other researchers: Andrew Goatly’s work on

(10)

10 ecolinguistics (2017), Norman Fairclough’s critical discourse analysis (2003), Theo van Leeuwen’s work on critical discourse analysis (2008), Greg Garrard’s work on ecocriticism (2012) and Mary Schleppegrell’s work on agency in environmental education (1997). This was done because Stibbe investigates material in slightly different contexts, and in order to perform an as exhaustive analysis as possible, the categories needed to be adapted to meet the requirements of the data. These specific categories were developed because, in combination, they have the potential of demonstrating how linguistic portrayal can construct blame. The categories can be summarised as follows:

• Facticity – patterns of language which portray something as true or false.

• Appraisal – patterns of language which portray someone or something as good or bad. • Erasure – patterns of language that background, exclude or entirely erase someone or

something.

• Salience – patterns of language that foreground someone or something.

To find these patterns in the texts, the data were processed by using each category as a lens. Segments of relevance were marked and analysed. The examples which were most demonstrative were then gathered in the results chapter. The categories will be exemplified and explained in more detail in the results.

(11)

11

4. Results

The first step of the analysis is to answer the first research question: how did the five newspapers linguistically portray non-human and human animals in the initial phase of the corona crisis? This is done to identify and define different elements that separately and collectively can work to attribute blame. The question is answered by presenting examples of linguistic portrayal in the data, according to each of the four categories. The next step of the analysis is to explicitly answer the second research question: based on the linguistic portrayal of non-human and human animals, who is constructed as being to blame for the corona crisis? This is done by presenting conclusive examples where the categories work in synergy. Through these, conclusions of how linguistic portrayal works to blame human and non-human animals in the corpus are presented.

4.1 Linguistic Portrayal

Facticity

In order to construct linguistic blame throughout texts, the author(s) can make use of and combine different means of linguistic portrayal; one of these is facticity. According to Stibbe, facticity in a text is conveyed by “patterns of language which represent descriptions of the world as true, uncertain or false” (2021: 17). By using linguistic items such as modal words, quantifying words, authority and different kinds of assumptions, a text can convince its reader of what is to be considered true and false. A text’s level of facticity can be designed by the use of certain adjectives and adverbs (Stibbe 2021: 119), and the level of modality is key. For example, Covid-19 might be described as “absolutely certain” to originate from bats. Such a statement does not open up to interpretation and is presented as true; it has a high level of facticity. However, if the virus is described as “rather likely” to have come from bats, the level of facticity is much lower, and the statement is therefore more open to interpretation.

Facticity can also be established through the use of presuppositions. Presuppositions can be described as “the representation of a proposition as an obvious, taken-for-granted, background fact about the world” (Stibbe 2021: 227). Norman Fairclough writes that “[w]hat is ‘said’ in a text is ‘said’ against a background of what is ‘unsaid’, but taken as given” (2003: 40). To exemplify, the question “when did the virus jump from bats to humans?” presupposes that “the virus comes from bats”. The use of presuppositions contributes to a high level of facticity.

Facticity throughout the text(s) can be created by pointing at authority or experts. Theo van Leeuwen writes that expert authority “may be stated explicitly, for instance, by mentioning credentials, but if the expert is well known in the given context, it may be taken for granted” (2008:

(12)

12 107). He claims that expert legitimisation is often present in what he calls “verbal process clauses” (X says Y) or “mental process clauses” (X believes Y), where the expert is the subject of the clause (ibid.). High facticity through authority can also be created by presenting the expert’s statements in quotations, as opposed to retelling their statements. By doing so, reliability and trust in what is claimed is established.

To demonstrate the facticity in the studied texts, and how it works to attribute blame, examples of how the texts make use of professional authority will first be presented. Then, it will be exemplified how facticity in the actual statements in the texts is created. Lastly, examples of how these work in synergy to establish over-all facticity patterns are displayed.

In the studied texts, levels of facticity through professional authority are created by a) using the epithets “professor” and “doctor” or nouns such as “researcher”, “scientist” and “author”, b) providing the reader with background information of the professional in question (academic background, previous research and place of work), c) placing the professional authority as the subject in a verbal or mental process clause, d) quantification of professionals (such as “some scientists” or “most scientists”), and e) quoting professionals.

Example 1 demonstrates how high facticity is established by professional authority. The example comes from an article about the origin of Covid-19. It discusses the idea that the virus was not a laboratory escape, but most likely originated in wild animals.

Example 1:

Prof Andrew Cunningham, from the Zoological Society of London, explained: ‘We've actually been expecting something like this to happen for a while. ‘These diseases are emerging more frequently in recent years as a result of human encroachment into wild habitat and increased contact and use of wild animals by people’. - BBC News, 4 May 2020

The epithet “prof” (short for professor) establishes trustworthiness and high facticity. So does the non-defining subordinate clause “from the Zoological Society of London”, which provides the reader with additional information of the professor. Moreover, the professor’s claim is presented in a quotation, which means it is presented in his own words, and not retold in someone else’s. This too serves as a means of establishing trust and high facticity. In addition, the professor is the subject in the verbal process clause “Prof Andrew Cunningham […] explained […]”, which also works to increase the facticity. Overall, the professional facticity in this example is very high, which leads to

(13)

13 distributing blame to “human encroachments” and “increased contact and use of wild animals by people”.

Example 2 shows how medium level facticity can be created by the use of professional authority. Although the authority does establish trust, it is much vaguer than the previous example. The article from which the example is taken is about the origin of Covid-19; it discusses and rules out the hypothesis that the virus may have been a laboratory escape.

Example 2:

Scientists think COVID-19 […] originated in bats and could have been transmitted to people via another mammal. - The Daily Mail, 25 May 2020

The fact that the noun “scientists” is the subject in the mental process clause “scientists think […]” creates high facticity. However, “scientists” is rather vague, and naming individual scientists would be more trustworthy. It is presupposed that the scientists have relevant competence to comment the matter, but no background of their experiences or place of work is provided. Therefore, the section may be described as carrying medium level of facticity. Although not as pronounced as in the previous example, the professional authority does establish blame, which is directed to bats and “another mammal”.

In example 3, the facticity is decreased further by using vague quantification as the premodifier to the head noun “researchers”. The example comes from an article that discusses human’s destruction of nature as a link to the spreading of viruses such as Covid-19.

Example 3:

But a number of researchers today think that it is actually humanity’s destruction of biodiversity that creates the conditions for new viruses and diseases such as Covid-19 […]. (emphasis mine)

- The Guardian, 18 May 2020

In the same way as “scientists”, the noun “researchers” is rather vague and not as trustworthy as the presentation of specific experts would be. However, it still establishes professional authority. So does the fact that the noun phrase “a number of researchers” is placed as the subject of a mental process clause. However, the premodifier “a number of” indicates that the theory is only shared amongst some

(14)

14 researchers; it does not reflect that it is shared by the majority of them or is even the most common theory. This lowers the reliability and thereby the facticity in the text. In addition, “a number of” does not refer to a specific amount; it could be referring to anything between 5 and 100. Overall, the professional authority is vague, and creates a relatively low level of facticity. The blame apportioned to “humanity’s destruction of biodiversity” is therefore not very strong.

Levels of facticity can also be conveyed in the statements (what is actually claimed, either by the author of the article or by the professional authority in question). The studied texts established this by a) modification or absence of modification in verb phrases and noun phrases, b) capitalisation, c) mathematical quantification and terminology, d) using hyperlinks to external sources, and e) presuppositions.

Capitalisation and lack of modification in the verb phrase can create high facticity, which example 4 demonstrates. The example comes from an article that is about research pointing at bats as the source of the virus. It discusses the virus’ DNA and its similarity to coronaviruses found in wild bats.

Example 4:

China's coronavirus DID come from bats. (emphasis in original) - The Daily Mail, 3 February 2020

The capitalisation of the auxiliary verb “did” emphasises the claim and contributes to a high level of facticity. In addition, there is no modification in the verb phrase. This combination leaves no openings to other hypotheses on the matter, and therefore contributes to establishing high facticity which works to heavily blame bats for the virus.

In example 5, modification in the verb phrase is used to achieve high facticity. It comes from the same article as mentioned above.

Example 5:

Dr Michael Skinner […] said: 'The discovery definitely places the origin of nCoV in bats in China’. (emphasis mine)

(15)

15 The adverb “definitely” modifies the verb “places”, contributing to increasing the facticity in the statement that the virus comes from bats in China. There are no openings for other hypotheses in the statement; the facticity is thereby high and distributes blame to bats in China.

Modification in the verb phrase can also be used to lower the facticity. This can be seen in example 6, which is taken from an article that is about human lifestyle and industrial animal farming and how that has come to allow viruses like Covid-19 to spread.

Example 6:

Since other bat coronaviruses have transited to humans via an intermediate animal host, it seems likely that this one did too. (emphasis mine)

- The Guardian, 28 March 2020

The verb “seems” is used to reflect the likelihood that the virus has been spreading from bats to humans via an intermediate animal. It is a vague verb and does not indicate certainty. The facticity is therefore rather low. Moreover, the adverb “likely” reflects that the scenario is not entirely certain; however, it is believable. Overall, the example reflects low facticity which means that blame is not heavily or actively apportioned to anyone, even though it is implicated that a non-human animal is to blame.

In example 7, mathematical quantification is used to establish high facticity in the statement. It also contains a presupposition. The example is from an article about bats’ way of living, their immune system and tolerance for viruses.

Example 7:

’We have found some 500 coronaviruses in China now that come from bats,’ said Dr Daszak, disease ecologist and president of research organisation EcoHealth Alliance.

- The Telegraph, 6 February 2020

By using the quantifier “500” as a premodifier for the head noun “coronaviruses”, high facticity is established. This is because mathematics, measurements and numbers have connotations to academic and trustworthy research. Moreover, by presenting the fact that they have found this many coronaviruses in bats, it is presupposed that Covid-19 also comes from bats. This too creates high facticity in the statement and works to blame bats.

(16)

16 The use of hyperlinks can work to establish high facticity, as can be seen in example 8. The example is from an article about investigating wild animals to find out where Covid-19 came from.

Example 8:

When scientists cracked the code of the new virus, taken from the body of a patient, bats in China were implicated. (emphasis in original)

- BBC News, 25 February

The noun phrase marked in bold is a hyperlink which leads to a document containing the genome code of the virus. The link is highlighted by being marked in bold, and its function is to offer the reader to learn more about the code through an external source. This strengthens the statement; the trustworthiness and facticity increases, which apportions blame to bats in China.

Example 9 demonstrates how professional authority and facticity of statements reflected in the texts work in synergy. The example is from an article about there being an intermediate host between bats and humans.

Example 9:

Prof Stephen Turner, head of the department of microbiology at Melbourne’s Monash University, says what’s most likely is that [the] virus originated in bats. But that’s where his certainty ends, he says. On the

hypothesis that the virus emerged at the Wuhan live animal market from an interaction between an animal and a human, Turner says: ‘I don’t think it’s conclusive by any means’. (emphasis mine)

- The Guardian,28 April 2020

In the sentence marked in bold, the epithet “prof” (short for professor) establishes trustworthiness, and so does the non-defining modifying subordinated clause “head of the department of microbiology”. By presenting this information, the reader will presuppose that this person has certain knowledge that can be trusted. The fact that the professor is the subject in the verbal process clause “Prof Stephen Turner […] says […]” strengthens the facticity. It is further strengthened by presenting the professor’s statement as a quotation. However, the modifying adverbial “most likely” indicates that the subject for discussion itself is not entirely certain, but has the likelihood of being so. In the two sentences marked in italics, the facticity shifts; a juxtaposition between the bold and the italic sections is indicated by the conjunction “but”. In the mental process clause, the professor “thinks”, a verb which carries much lower facticity than for example “claim”, that the hypothesis is not “conclusive by any means”. The fact that he claims this means that there is much more complexity to the matter than what has been explicitly brought up in his statement. However, the reader is not

(17)

17 supplied with any further explanations, which lower the facticity in the text. In conclusion, because of this juxtaposition between the two statements, mixed levels of facticity are present in the overarching claim. The blame is thereby not strongly assigned to anyone, but it is implicated that bats, in combination with others who are not explicitly brought up in the statement, are to blame for the virus.

Appraisal

Appraisals can contribute to establishing blame in texts. Stibbe claims that “[a]ppraisal patterns are clusters of linguistic features which come together to represent an area of life good or bad” (2021: 79). Appraisals can be created by the use of lexis which explicitly evaluates something, such as

good/bad and right/wrong. They can also be created by words which have positive or negative

connotations (ibid.). In this study, words such as spacious, bright and natural, is considered having positive connotations, and crowded, unnatural and dark is considered having negative connotations. Stibbe writes that “[w]hen positive words cluster together […], they form appraisal patterns” (2021: 80). Such clusters can create euphemisms (patterns of language that understate and downplay). In the same way, clustering negative words can form dysphemisms (patterns of language that overstate and exaggerate). By repetitively describing something or someone using understatements or overstatements, the reader will be manipulated to see them in a certain light. The use of such language has the power to downplay or exaggerate involvement, and it is therefore of great importance to take into consideration when examining linguistic blame.

Appraisals can also be established by metaphors (Stibbe 2021: 79). If a text for instance claims that “bats are culprits”, bats are negatively appraised by moving them to the context of crime, which could be considered inappropriate in this context.

Moreover, appraisals are relevant in terms of showing the importance of something (Fairclough 2003: 172). That is, it is not only the representation of something as positive or negative in itself that may be conveyed by using them, but they can also tell us what the text considers important or unimportant. For example, if a text presents an area of life using negative words, it conveys that it does not consider that area important. Thereby, any moral difficulties in blaming that area are downplayed or removed.

Through the examples taken from the studied texts, it will first be demonstrated how the texts use different means of appraisal in order to evaluate someone or something and, through that, distribute blame. Then, it will be displayed how these means can be used in synergy.

(18)

18 In the corpus, the appraisal patterns were found to be exclusively negative. The texts establish such appraisals by a) contrasting clauses carrying positive or negative appraisals, b) using lexis with negative connotations such as “dense” and “murky”, c) clustering lexis carrying negative connotations, and d) metaphorically reframing bats and pangolins by moving them from the domain of their natural way of living to the domain of crime and mystery.

Example 10 shows how two contrasting clauses can create negative appraisal. It is taken from an article which is about bats’ resilience to viruses.

Example 10:

Bats may be linked to good fortune in traditional Chinese culture, but as the most likely source of the fast-moving coronavirus outbreak sweeping the country and globe they have triggered havoc, not prosperity. (emphasis in original)

- The Telegraph, 6 February 2020

In this sentence, the two clauses have a contrasting function. Bats are in the first clause linked to “good fortune”, which is something positive. Both the premodifying adjective “good” and the head noun “fortune” work to explicitly appraise bats positively. However, the conjunction “but” indicates a contrast between this statement and what is to come in the next clause. In the second clause, bats are placed as the subject that “have triggered havoc”. The noun “havoc” can be defined as synonymous to “devastation” and “destruction”, which are extremely negative words. The outcome is that bats are appraised negatively, which leads to blaming them for the “coronavirus outbreak”.

Negative connotations can be used to appraise bats negatively, which example 11 demonstrates. It is from the same article as the above.

Example 11:

What's more, they are found on every continent except Antarctica, their ability to fly means they can transmit viruses far and wide, and they live in dense, crowded colonies. “So if you think about how they live their lives, you can understand why there are a lot of viruses circulating among bats,” said Jonathan Ball, professor of molecular virology at the University of Nottingham.

(19)

19 The adjectives “dense” and “crowded” are used to put bats in a bad light. Both words have negative connotations, as opposed to “roomy” and “spacious”. In this context, the words are used to link bats’ way of living to the spread of the virus. Arguably, it is a dysphemistic way of describing them and it does therefore appraise them negatively. Through this, bats are blamed for spreading the virus.

Example 12 demonstrates how bats are reframed by moving them to the context of mystery and crime. The section comes from an article which is about finding the source of Covid-19 in wild animals.

Example 12:

The second part of the puzzle, then, is the identity of the mystery animal that incubated the virus in its body and possibly ended up in the market at Wuhan. One suspect for the smoking gun is the pangolin. (emphasis mine) - BBC News 25 February 2020

The noun phrases “the puzzle” and “mystery animal” have connotations to mystery, and the noun phrases “one suspect” and “the smoking gun” have connotations to crime. Though this, pangolins are metaphorically reframed by moving them to the context of mysterious crime drama. These noun phrases work to cluster into a dysphemistic description of the animal, appraising it negatively. This results in blaming the animal for actively spreading the virus.

Different means of appraisal can work together to negatively appraise non-human animals, as is showed in example 13. In this example, bats are both metaphorically reframed in the context of mystery and described in terms that have negative connotations. The article from which the example is taken discusses the fact that people living nearby a cave where many bats live seemingly are immune to Covid-19.

Example 13:

People living near the mysterious cave in Yunnan, China, are immune to the deadly disease – even though thousands of critters inside are carrying the virus. (emphasis in original)

- The Daily Star, 12 February 2020

The premodifier “mysterious” to the head noun “cave” places the bats living in the cave in the context of mystery. The adjective “mysterious” can be defined as something that is impossible or hard to explain or understand, and which is of puzzling nature. Thus, the word has connotations to things which are beyond humans’ area of understanding. Since this cave is where bats live, the connotations spill over to them too. It leads the reader to think of bats as entities that are impossible to relate to and therefore are not worthy of consideration. Moreover, the noun “critters” could be said to carry

(20)

20

negative connotations and does therefore place bats in a light which appraises them negatively. This results in removing any moral difficulties in blaming the bats for the virus.

Erasure

An important aspect in establishing blame in texts is the absence, or erasure, of someone or something. In order to analyse texts in depth “linguists [must] give critical attention not only to participants which are explicitly represented in [them] but also to those which are supressed, backgrounded, excluded or erased […]” (Stibbe 2021: 139). The term “erasure” may be used to describe that something important and worthy of attention has been ignored, removed or manipulated. It works as an umbrella term for suppression, backgrounding and exclusion (ibid.: 140-141). Stibbe writes that this can be done by simply omitting to mention someone or something, but it can also be displayed “through linguistic devices such as passives, metonymy, nominalisations and hyponymy” (ibid.: 144).

Erasure can occur by using lexis related to other discourses than the relevant one, and thereby metaphorically reframe an entity by moving them to a different domain. For example, by using lexis from the discourse of food, the fact that animals are living and sentient beings is backgrounded or suppressed. This may lead to the idea that their only ability is serving as food for humans, and that they are not important in any other contexts. By establishing this idea, any moral difficulties in blaming the animals are removed.

Furthermore, participants in an action can be erased by manipulating semantic roles. Nominalisations do so by turning verbs into nouns, or processes into things. For instance, the noun “destruction” is a nominalisation of the verb “destroy”. Where the verb would reflect that X destroys Y, both the agent and the patient are absent in the nominalisation. Mary Schleppegrell writes that the use of nominalisations may work as non-human agents in texts, but it “implies a notion of force or energy which is not intentional rather than a notion of agency or responsibility” (1997: 55).Moreover, Andrew Goatly claims that “nominalisations often exclude reference to agents or external causes, suggesting […] a self-generated process” (2017: 53). This kind of abstraction may lead to the reader believing that the “destruction”, or what other nominalisation it may be, is inevitable or unchangeable, which leads to no one being blamed for it. Another way of erasing social actors and deflect blame is by using passive constructions. Since a passive construction does not include an agent, the process may be viewed as inevitable. An example of a passive construction is “habitats are being destroyed”, instead of “X destroy habitats”.

(21)

21 Through the examples taken from the corpus, different means of erasure will first be demonstrated. Then, examples of how different means of erasure can be used in synergy to apportion blame will be presented.

The studied texts were found to erase entities by a) exclusively focusing on humans, b) using nominalisations and passive constructions, c) using metonymies and hyponymies, and d) metaphorically reframe non-human animals by moving them from the domain of living beings to the domain of food and non-living items.

Relevant entities can be entirely erased by omitting to mention them. Example 14 demonstrates how this can be done. The example is from an article about pangolins carrying the virus.

Example 14:

Scientists say the sale of the animals in wildlife markets should be strictly prohibited to minimise the risk of future outbreaks.

- BBC News, 26 March 2020

The adverbial ”to minimise the risks of future outbreaks” provides the reader with the reason to why ”the sale of animals in wildlife markets should be strictly prohibited”. The extreme cruelty animals are put through at the markets is not presented as a reason for closing them down, and therefore the lives and feelings of the animals are entirely erased from the context. This may lead to the reader believing that they are not worthy of consideration. By exclusively presenting this explanation, and excluding other reasons which are highly important, the text generates anthropocentrism. As a result, any difficulties in blaming the animals for the virus is entirely removed.

In example 15, the lives of chickens are erased by the use of anthropocentric discourse. The section is from an article which is about the consequences of factory farming in the context of Covid-19.

Example 15:

This gives us an opportunity, he says, to question our lifestyle choices – because chicken isn’t cheap if it

costs a million lives – and vote for politicians who hold agribusiness to higher standards of ecological,

social and epidemiological sustainability. (emphasis mine) - The Guardian, 28 March 2020

By using the uncountable noun “chicken” instead of the countable “chickens”, the animals cannot be considered anything but food. This erases the fact that chickens are living and sentient individuals.

(22)

22 Moreover, it is explicitly stated that “chicken isn’t cheap” with the condition adverbial “if it costs a

million lives”. This condition refers exclusively to human lives, but by not explicitly saying so,

chickens are portrayed as not having lives at all. Using chickens as food always costs millions of lives because it presumes that we are killing them. The fact that chickens are living creatures is completely erased from the context. As a consequence, their lives are portrayed as unimportant, and thus the text implies that they, without any consequences, can be blamed for viruses such as Covid-19.

In example 16, bats are reframed by moving them from the domain of living beings to the domain of food, which works to erase their sentience. The example is from an article which discusses the fact that some locals who live nearby a bat cave are immune to Covid-19.

Example 16:

[M]arket traders in Indonesia have said sales of bats and bat curry continue. [emphasis mine] - The Daily Star, 12 February 2020

The noun phrases “bats” and “bat curry” are placed in a close relation to each other. In the noun phrase “bat curry”, “bat” is a premodifier to the head noun “curry”. Since “curry” is the main noun, bats are portrayed as a dish. Because of the close relation between the two noun phrases, the unmodified noun “bats” is placed in the same domain. In other words, the noun phrases are constructed as co-hyponyms under the superordinate category “dish” or “food”. By doing so, the fact that bats are sentient individuals is erased, which leads to removing any moral difficulties in blaming them.

In example 17, the fact that pangolins are living creatures is erased by placing them as co-hyponyms to objects. The example is from an article which discusses the origin of Covid-19 and how it came to spread.

Example 17:

As reported in Nature, pangolins were not listed on the inventory of items being sold in Wuhan, although this omission could be deliberate as it’s illegal to sell them.

- The Guardian, 28 April 2020

Pangolins are placed as co-hyponyms under the superordinate category “items being sold in Wuhan”. The noun “items” can only refer to non-living entities, and therefore, the hyponymic categorisation

(23)

23 turns pangolins into non-sentient objects. It erases the fact that they are living being, and it leads the reader to view them as unimportant.

Erasure of human actions can be performed by hiding agency in metonymies, as is demonstrated in example 18. It is taken from an article which is about US-granted research on bats in a laboratory in Wuhan.

Example 18:

The laboratory at the centre of scrutiny over the pandemic has been carrying out research on bats from the cave which scientists believe is the original source of the devastating outbreak.

- The Daily Mail,11 April 2020

The noun phrase “the laboratory” is a metonymy for the researchers working at the laboratory. By placing the metonymy as the agent in the clause, the actual humans performing the research on bats are hidden and thereby erased from the context. This may lead to the reader believing that there is no one to blame for carrying out the research on the bats.

Human agency can be erased by the use of nominalisations. Example 19, which is from an article about the link between viruses and factory farming, shows this.

Example 19:

‘Hopefully,’ he says, ‘this will change our notions about agricultural production, land use and conservation’. - The Guardian, 28 March 2020

The agents in the processes “X ‘produces’ agriculture”, “Y uses land” and “Z conserves X” are erased by turning the verbs in to the nouns “agricultural production”, “land use” and “conservations”. By doing so, the use of land and agricultural production are presented as natural or inevitable processes. This erases the fact that humans are behind the actions, and that we have a choice in carrying them out. Because of the absence of agency, the text does not fully blame humans for the actions.

Human involvement can be erased by using passive constructions, as is demonstrated in example 20. The example is from an article about US-granted research on bats in a laboratory in Wuhan.

(24)

24

Bat samplings were conducted ten times from April 2011 to October 2015 at different seasons in their natural habitat at a single location (cave) in Kunming, Yunnan Province, China. Bats were trapped and faecal swab samples

were collected. (emphasis mine)

- The Daily Mail, 11 April 2020

The fact that the bats were trapped by someone, and the faecal swab samples were performed by someone, is erased by using passive constructions in the verb phrases “were conducted”, “were trapped” and “were collected”. By doing so, the clause does not need an agent. This creates abstraction and supresses the fact that there are actual individuals behind the actions performed on bats. Instead, it is presented as a natural process. As a consequence, no one gets blamed for the actions.

Example 21 displays how different means of erasure can work in synergy. The example is from an article about the process of finding the source of Covid-19 in wild animals.

Example 21:

Conservationists are at pains to point out that although bats are thought to carry many viruses, they are also essential for ecosystems to function. ‘Insectivorous bats eat huge volumes of insects such as mosquitoes and agricultural pests, while fruit bats pollinate trees and spread their seeds,’ he says. ‘It is imperative that these species are not culled through misguided “disease control” measures’.

- BBC News, 25 February 2020

Bats are presented as essential for ecosystems to function because they eat pests, mosquitoes and agricultural pests. These functions in the “ecosystem” are exclusively based on human benefits, and therefore anthropocentric. The fact that bats “pollinate trees” may also be seen as anthropocentric since agricultural connotations have already been established. It brings to mind large-scale fruit production. The fact that they “spread [the trees’] seeds” does on the other hand not do the same thing. It is quite common that humans want to control where trees, crops and other plants grow, so bats’ spreading seeds arguably does not have any human benefit in the context of agriculture. However, the highly relevant fact that bats are essential for their own sake is omitted in the arguments about protecting them.

The passive verb construction “are not culled” erases the human agent. This creates abstraction since it makes the process seem inevitable. In addition, “to cull” means to select and kill wild animals and can be considered a euphemistic and abstract way of presenting the action of killing them. Furthermore, the noun phrase “disease control’ measures” is a nominalisation, which means that the agent performing these measures has been erased. This too is a euphemistic way of saying that

(25)

25 someone is killing the animals. Overall, this example erases human involvement and does thereby dilute human blame.

Salience

The opposite of erasure is salience. In order to convey linguistic blame, one must also look at what and whom is being highlighted in the texts; salience refers to something that stands out and is easily recognised (Chiarcos, et al., 2011: 2). That is, by foregrounding, drawing attention to and concretise entities in texts, they can be highlighted. Stibbe claims that there can be a direct building of salience in people’s minds through vivid and concrete depiction (2021: 160), and “[b]y analysing a range of linguistic features, including focus, vitality, levels of abstraction, transitivity, and metaphor, it is possible to reveal salience patterns which represent an area of life vividly and concretely” (ibid.: 161). For example, animals can be described using concrete, vivid or recognisable lexis such as being “curious” or having “homes” and “families”. According to Garrard (2012), this can be referred to as critical anthropomorphism. As opposed to crude anthropomorphism, which ridicules non-human animals, critical anthropomorphism works to understand non-humans in human terms (ibid.: 154-156). This language use may lead the reader to view animals in a new light. By doing so, animals may be seen as sentient beings worthy of consideration instead of non-sentient objects. In the next instance, this may work to remind the reader of the fact that there are consequences in blaming them. Theo van Leeuwen writes that attention can be brought to entities by activation; that is, by foregrounding them as agents in clauses (2008: 33). In this context, salience can foreground human responsibility by placing us as agents performing actions, and thereby apportion blame to humans. In the same way as erasure, salience works on a scale. That is, the salient patterns can be more or less overarching. Entities can, for example, be made highly salient by both placing them as agents in clauses and describing them vividly and concretely throughout texts. In contrast, they can be made salient by only being placed as agents and by being described in very abstract words. If so, the salience is not as overarching. Keeping this in mind, not every example in this section is to be considered a

model for salience patterns. In some examples, attempts to make an entity or action salient are

foreshadowed by combining them with erasure.

In the texts, salience of non-human animals and human activity is created by a) explicitly pointing at human involvement, b) placing humans as agents in clauses, c) critical anthropomorphism (using recognisable lexis, such as “home”, when referring to non-human animals), and c) placing animals as living beings in hyponymic levels.

(26)

26 In example 22, human involvement in the outbreak of new viruses is made salient by explicitly evaluating it as “true”. The example is from an article that is about the likelihood that industrial animal farming has contributed to the increased spreading of viruses like Covid-19.

Example 22:

It’s true, in other words, that an expanding human population pushing into previously undisturbed ecosystems has contributed to the increasing number of zoonoses […]. (emphasis mine)

By introducing the complex noun phrase “an expanding human population pushing into previously undisturbed ecosystems” with “[i]t’s true […] that” it is explicitly and concretely presented who the responsible entity is. This makes human involvement salient. However, one could argue that prominent human actions that has led to this are diluted by placing “humans” as a premodifier in the head noun “population”. Humans are part of a nominalisation and therefore there is nothing suggesting how we have performed these actions and that we are responsible for them. Nor is it stated what these actions have led to for non-human animals. Overall, there is an attempt in making human actions salient, but it is not a very strong one. As a result, blame apportioned to humans is not very pronounced.

In example 23, bats are depicted as having homes, a concept which could work to highlight the fact that they are living beings worthy of consideration. The example is from an article discussing bats’ tolerance to viruses.

Example 23:

Dr Daszak and his colleagues have spent years exploring vast networks of limestone caves that are home to

millions of bats in southern China. (emphasis mine)

- The Telegraph,6 February 2020

Bats are depicted as having “homes”. This is a concrete noun, and a concept which is familiar to the majority of humans. It has the potential of drawing the reader’s attention to bats’ similarities to us, more so than for example “habitat” would do. This could potentially influence the reader to seeing bats as worthy of consideration, and thereby remind the reader that there are consequences in blaming them for the virus. However, by placing the noun “home” in a modifying clause, and not in the main clause, the critical anthropomorphism is not made very prominent.

(27)

27 Placing humans as the agent performing an action works to highlight human involvement, as example 24 shows. The example is from an article about humans contributing to spreading viruses such as Covid-19 by expanding into previously undisturbed environments.

Example 24:

’We are creating habitats where viruses are transmitted more easily, and then we are surprised that we have new ones’. (emphasis mine)

By placing the pronoun “we” as both the subject and the agent in the clause, maximum blame is achieved. Through this construction, humans are explicitly linked to creating the conditions for Covid-19 to spread. Thereby, human involvement is made salient, and blame is distributed to humans. Moreover, the noun “habitats” is conventionally used to refer to non-human animals’ home. However, in this example, it is used to refer to places where non-human animals and humans live together, and therefore it could be said to place human and non-human animals as co-hyponym. By doing this, the reader might be reminded of our similarity, which may work to make animals salient. However, it is not an arresting attempt, and the noun “habitat” might as well be used in an arbitrary manner.

Example 25 demonstrates how different means of salience can work in synergy. Bats are portrayed as living entities by hyponymic categorisation, and humans are placed as agents performing the actions. The example is from the same article as the above.

Example 25:

‘We have created densely packed populations where alongside us are bats and rodents and birds, pets and other

living things […]’. (emphasis mine)

“We” is the agent as well as the subject in the clause. This construction gives humans maximum blame for the performed action; through this, human responsibility is made salient.

Bats are placed as co-hyponyms under the superordinate category “living things”. This is a collocation, commonly used to describe non-human animals. Nevertheless, regardless of it being a set expression, it explicitly turns the animals into things – something which semantically cannot be living. However, the head noun “things” is modified by the adjective “living”, which explicitly provides the things with the ability to live. The combination of the words forms an oxymoron, and because of this lexical contradiction, it forms a very weak form of salience. Since this is one of very

(28)

28 few examples to attempt at foregrounding animals, it serves as representative of how animals are portrayed throughout the analysed texts. In other words, animals are rarely made salient in these specific texts, and when they are, it is not very prominent. As a result, any moral consequences in blaming the animals are downplayed.

4.2 Linguistic Blame

This section of the results demonstrates how the combination of the four categories can work in synergy to establish blame. Through the examples, the second research question is explicitly answered: who is constructed as being to blame for the corona crisis?

In order to blame bats for spreading Covid-19, negative appraisal, erasure and facticity can work in synergy. The example, which is from an article about bats being tolerant to viruses, demonstrates this.

Example 26:

It’s not just coronaviruses that jump from bats to humans – they have also been identified as the natural reservoir for deadly diseases including rabies and the Ebola and Marburg viruses. This is perhaps unsurprising when you consider that bats make up roughly 20 per cent of all species of mammals – there are more than 1,200 types – and they have extraordinarily long lifespans, with some bats known to reach 40 years old. What's more, they are found on every continent except Antarctica, their ability to fly means they can transmit viruses far and wide, and they live in dense, crowded colonies. ‘So if you think about how they live their lives, you can understand why there are a lot of viruses circulating among bats,’ said Jonathan Ball, professor of molecular virology at the

University of Nottingham. (italics mine, underlined in original)

- The Telegraph, 6 February 2020,

Professional authority is established by placing “Jonathan Ball, professor of molecular virology at the University of Nottingham” as the subject in a verbal process clause. Not only does the epithet “professor” reflects high level of facticity, but so does the presentation of his place of work. In addition, the fact that what he claims is presented in quotations also contributes to high professional authority. By doing this, high facticity is established throughout the text. It is also high in the professor’s statements, and is established by presuppositions, mathematical quantification, the use of hyperlinks to external sources and by the absence of modification in the verb phrases. The professor claims that bats’ natural ability to fly and their natural way of living is the reason that many viruses circulate amongst them. In the next instance, it is presupposed that they have spread Covid-19. Since presuppositions lead to high facticity, it is presented as true that bats are to blame for the spread if the virus.

(29)

29 The professor also states that bats are “the natural reservoir for deadly diseases” – a noun phrase that both appraises bats negatively and portrays them as objects. The head noun “reservoir” is often used in medicine to describe a source of pathogens, but literally, it refers to a container, pool or other storing objects. Therefore, bats could be said to be portrayed as non-living objects, which erases the fact that they are sentient beings worthy of consideration. In the next instance, this removes any moral difficulties in blaming them. By also placing the prepositional phrase “for deadly diseases” as the postmodifier, bats are placed in a close relation to two words with negative connotations, which appraises them negatively. Through this passage then, bats are constructed as being to blame for the corona crisis.

Erasure and salience can work in synergy to blame humans for the corona crisis, as is showed in example 27. The example is from an article which is about decreasing biodiversity and human responsibility.

Example 27:

In 2008, [Kate] Jones [chair of ecology and biodiversity at UCL] and a team of researchers identified 335 diseases that emerged between 1960 and 2004, at least 60% of which came from animals. Increasingly, says Jones, these zoonotic diseases are linked to environmental change and human behaviour. The disruption of pristine forests driven by logging, mining, road building through remote places, rapid urbanisation and

population growth is bringing people into closer contact with animal species they may never have been near

before, she says. The resulting transmission of disease from wildlife to humans, she says, is now ‘a hidden cost of human economic development. There are just so many more of us, in every environment. We are going into

largely undisturbed places and being exposed more and more. We are creating habitats where viruses are

transmitted more easily, and then we are surprised that we have new ones’. (italics mine, underlined in original) - The Guardian, 18 March 2020

High level of facticity is established throughout the text by pointing at professional authority. Contributing to this is the fact that the claim of the professional is presented in quotations, that she is the subject in a verbal process clause (“says Jones”), and the noun phrase “a team of researchers”. Moreover, mathematical quantification, and lack of modality is verb phrases, establish high facticity in the statements. What is said, then, is presented as certain, and the person saying it is presented as trustworthy.

The claim that “zoonotic diseases are linked to environmental change and human behaviour”, suggests humans are involved in the emerging of viruses such as Covid-19. Nonetheless, it is presented through the normalisations “environmental change” and “human behaviour”, which erases the fact that there are actual human actions behind it. The change in environment and the behaviours