Abstract

The present study explores the inherent ideologies of newspaper discourse and the subjugation of the readers to the said ideologies. 30 articles reporting Black murders by White policemen published in The New York Times, USA Today and Washington Post between 2014-2021 are used for creating three corpora of 25,561 words in total. AntCon 3.5.9 is used to generate the wordlist, N-Grams and concords; and LancsBox 6.0 is used to annotate the corpora. The study uses Critical Discourse Analysis and the Ideological Square Model of Van Dijk as its framework. The findings reveal that the use of certain lexical items plays a significant role in the construction of a discourse that abets the manipulation of the news. The deliberate omission of the ethnic identity of the victims and linguistically naturalizing the murders in the Washington Post makes it a racist and partial reporter of crimes. NYT and USA Today support pro-black.

Key Words

Corpus-based Analysis, Ideology, Racism, CDA, Ideological Square

Introduction

News media serve as a source of information for the masses. They claim to be pursuers of neutral and objective reporting. However, Egelhofer and Lecheler (2019) revealed that lately there has been a substantial rise in circulation of fake news. Strömbäck et al. (2020) stated that the trust in news agencies has dropped to 32% in 2016 worldwide and reportedly, 12% of people have blind faith in news media. Van Dijk (2019) explored how ideology and racism could possibly be related to discourse and what sort of relationship they share with each other. He stated that racist ideologies are often expressed through discourse, and display the inherent ethnic prejudices. Van Dijk (2000) clarifies that racist ideologies are more commonly observed in the West. This ideology, as specified by Van Dijk, results in the generation of a discourse that divides the masses into a dichotomy of "US" and "THEM".

The study at hand evaluates the nature of news reporting by three American newspapers. It aims to identify the ideologies that are advocated by each newspaper and hopes to ascertain the lexical tools that are employed to achieve such purposes. For this, the paper will take into consideration 10 events of black murders and closely follow the discourse used by the newspapers for their coverage. An in-depth analysis of the text will enable the researcher to reach a sound conclusion regarding which side of the dichotomy, US and THEM, are the black victims closely associated with by these newspapers. 

Literature Review

This study aims to look into the role of print media in projecting racism by analyzing articles on black murders by American newspapers which are: The New York Times, Washington Post, and USA Today. It thoroughly examines how the black murders are portrayed by the newspapers, specifically when the perpetrators are white police officers. Tabassum, Shah, and Bilal (2013) evaluated the prominent left or right-wing ideologies in English-language newspapers in Pakistan. They viewed the editorials concerning the death of Osama Bin Laden and determined how they functioned in the construction of ideologies. Editorials, otherwise known to be representative of objective perspectives, were concluded in their study to be chiefly the proponents of the newspaper's ideologies. Therefore, Pakistani English newspapers were discovered to be actively engaged in advocating their personal ideologies rather than objective viewpoints. 

Media institutions promote ideologies that are strictly congruent with their political affiliations, as quoted by Shojaei, Youssefi, and Husseini (2013) in their study. They found that Western newspapers, chiefly British and American newspapers, use language as their foremost tool for dispersing ideologies. Secondly, their study culminated in the finding that using language in a preplanned manner aids the newspapers in highlighting or suppressing their desired sides. It further asserts the power that newspaper agencies possess by virtue of linguistic tools in the dispersal of ideologies. More on Western news media, Miranti (2014) studied the editorial section of the New York Times and Washington Post for their discourse on non-registered immigrants. The results showed that the New York Times has a proclivity for a liberal position whereas the Washington Post was more conservative, hence depicting the immigrants as "out-group". Moreover, Mahmood, Kausar, and Khan (2018) also reiterate the notion that newspapers have the capacity to emphasize or suppress any group of their liking. They studied the reporting of the 16th December attack on the Army Public School in Dawn News and The New York Times. It was discovered that Dawn News more prominently portrayed the “US” discourse, in comparison to The New York Times in which “THEM” discourse was used. It meant that the Pakistani newspaper chose to represent the attackers as militants i.e. US and the American newspapers represented a more impartial image by referring to the attackers as Taliban. 

Bartesaghi and Pantelides (2017) were of the view that Critical Discourse Analysis (CDA) is capable of analyzing myriads of varieties of data. They state that it may be used both as a theoretical or methodological framework. Xie (2018) believed that CDA enables the researcher to view newspaper discourse with a critical lens by analyzing the lexical choices and examining the practice level at which the newspaper discourse is executed. It means that it breaks down the newspaper text into different levels based on its apparent and hidden meanings. Xie further stated that CDA may be used to test the accuracy of a news text, find its implications, target audience, and the desired results of the news agency. 

Employing CDA as their theoretical framework, Tanvir, Khoula and Zahra (2018) explored the discourse of American newspapers on incidents of mass shootings. Their findings revealed that the newspapers deliberately omitted objective facts and focused on stressing their ethnicities whenever the shooters belonged to ethnic groups such as Afghan, Bosnian, Koreans, Saudis, and Vietnamese. However, the race of the perpetrator was never revealed in the case of white shooters and the discourse opted was comparatively objective. The subjective discourse employed in the case of the ethnic shooters comprised of adjectives with negative meanings thus serving as the broadcaster of racist discourse. These findings were in line with those of Sajid, Anwar, and Ashraf (2019). They revealed that the newspaper discourse is manipulated with the help of certain linguistic tools after analyzing English newspaper headlines in Pakistan. This discourse is shaped to suit a preferred ideology. 

These studies accentuate the subjectivity that is deliberately opted for by the various news agencies for projecting their ideologies onto the masses. The present study selected 10 incidents of black murders for gathering 30 articles, 10 from each newspaper. These articles will be analyzed with respect to their lexical word classes i.e. adjectives, nouns, and lexical verbs to find how these newspapers differ in their preferred ideologies. Yule (2020) defines adjectives as words that provide additional information about the nouns with which they are used. He refers to nouns as naming words that are used to define people, places, things, etc. Lastly, according to him, verbs may be used for three purposes: to describe an action, state, or course of event. These words will provide the researcher an insight into the nature of the discourse of the selected articles. The literature studied established a gap in the literature. To fill this, the present study will seek answers to the following research questions:

i. How do the lexical word choices differ in the three newspapers when reporting incidents of black murders?

ii. How do these words portray the ideology of the relative news agencies? 

Research Methodology

The research follows a mixed-method design. Creswell et al. (2003) defined mixed methods as one that may have any of the four underlying models as the acting design in any research. The one followed in this study is the mixed method accompanied by an explanatory approach. This design comprises two steps: quantitative collection of data followed by its analysis and secondly its explanation, using the data that has been gathered qualitatively. This study is based on corpus techniques because it uses deductive reasoning.

For this study, three corpora of a total of 25,561 words were compiled. Each newspaper i.e. The New York Times, USA Today, and Washington Post had its own corpus which comprised 10 articles per newspaper, making it 30 in total. This study utilized purposive sampling, in which the researcher selected 10 incidents of black murders from the years 2014-2021 based on their own judgment. The names of the black victims and the dates of their killings are given in Table 1.




 

Table 1

Black victims of white police shootings and the dates of their murder

Black Victim	Date of Murder
Eric Garner	17 July 2014
Michael Brown	9 August 2014
Tamir Rice	22 November 2014
Walter Scott	4 April 2015
Alton Sterling	5 July 2016
Philando Castile	6 July 2016
Stephon Clark	18 March 2018
Breonna Taylor	13 March 2020
George Floyd	25 May 2020
Daunte Wright	11 April 2021

        




These incidents were selected because all the victims were of black ethnicity and the perpetrator in each case was a white police officer. Furthermore, the manner of killings was similar as well i.e. choking or shooting. Each of the victims was killed based solely on the officer's suspicion. The events are also interlinked because they caught people's attention worldwide and resulted in various movements online, including "#BlackLivesMatter” and “#ICantBreathe". These incidents were closely followed by the news media and therefore had sufficient material to be researched. They all took place within the last decade and are hence fairly recent and relevant. During the pandemic, Americans came out on the street to protest against these murders. With George Floyd's

killing, all the previous victims were remembered. News agencies such as USA Today and the Washington Post made frequent mentions of the earlier killings and related them to the most recent murders of Floyd, Taylor, and Wright.

This study is based on Van Dijk’s Ideological Square Model. This model describes a dichotomy that is created through discourse and marks the distinction between "US" and "THEM". Via this model, Van Dijk elucidates how a group may be otherized by intentional highlighting of their bad characteristics, whereas another group may be treated as naturally an in-group by hiding their negative features and bringing forth their positive attributes. It may be represented as given in Figure 1.




 

Figure 1

Van Dijk’s Ideological Square Model



Lastly, Critical Discourse Analysis (CDA) serves as the theoretical framework for this research. It is commonly used for analyzing discourse from a wide array of contexts, including that of news. Xie (2018) pointed out that it is frequently used in research in order to uncover the true source of power manipulating the news discourse. He highlights the foremost step in Critical Discourse Analysis which is to determine the relationship between discourse and the basis of its production. Nartey and Mwinlaaru (2019), talked about the unison of CDA and Corpus Linguistics, stating that when they go hand-in-hand, they arrive at more systematic and accurate results.  

Accompanying the Critical Discourse Analysis theory is the corpus tool, AntConc 3.5.9. For the data analysis, three corpora were generated out of the 10 articles selected for each newspaper. These corpora were analyzed using the aforementioned software. This software has the features of concordances, N-Grams, collocates, wordlists, keywords, and concordance plots. However, this research will primarily focus on the words, concordances, and N-Grams. The wordlist provides the list of the words with respect to their frequency in the text, the concordances show the context in which a selected word is used and the N-Grams provide the nth word/s which appear in the proximity of a chosen word. The software LancsBox 6.0 was used to annotate the data with the help of its feature "Whelk". This marks all the different parts of speech, thus the verbs, nouns, and adjectives can be easily located. All these features combined will not only speed up the analysis but also yield far more accurate results as compared to manual analysis. 

Data Analysis

Tanvir, Khoula, and Zahra (2018) studied the representation of ethnic and white shooters in various American newspapers establishing that media is not entirely impartial in news reporting. This idea was further explored in the present research to examine how the American newspapers, USA Today, The New York Times, and Washington Post represent black victims. The corpora were analyzed to compare the discourse of the three newspapers, predominantly the action verbs and the ethnic adjectives such as black or white. The results were then compared to determine the inherent ideology being propagated by the respective newspapers.

The analyses were congruent with the findings of the above-stated study. The wordlist tool was used to generate the frequencies of the verbs kill, murder, and die for the corpora. The results are given in the table below.




 

Table 2

Frequencies of the verbs kill, murder, and die in the three corpora

Newspaper	Verb*	Frequency
New York Times	Kill	6
	murder	1
	Die	6
USA Today	Kill	8
	murder	2
	Die	12
Washington Post	Kill	19
	Murder	0
	Die	7

        




The asterisk with the verb means that the wordlist tool searched all forms of the mentioned verbs. The usage of these words either as verbs or nouns was identified using the Whelk feature in LancsBox. It is evident from the table that in a 25,561-word corpus, the frequencies of the verbs kill and murder are quite insignificant. The verb choice reveals how the incidents have been reported. The verb "die" naturalizes the event and makes it a case of natural demise. It softens the nature of the crime and does not accurately describe the atrocity associated with the criminal act. However, the verbs "kill" and ''murder'' convey impartiality, objectivity, and neutrality. Verbs are often the words that appear in the headlines of news articles reporting murders and are therefore among the first words with which a reader interacts. Therefore, the usage of verbs largely determines the approach that a newspaper is adopting.

Coming to the ethnic adjectives which were manually selected with the help of Whelk in LancsBox, they were looked up in the three corpora using wordlist in AntConc. Their frequencies are given in the table below.




 

Table 3

Ethnic adjectives and their relative frequencies in the three corpora

Newspaper	Ethnic Adjective	Frequency
New York Times	Black	16
	White	10
USA Today	Black	44
	White	14
Washington Post	Black	16
	White	4

 




The frequencies reveal how frequently the ethnic identity of the victims and the perpetrators were revealed in the articles. USA Today appears to have the highest frequencies of both the adjectives, black and white. The New York Times falls in second position whereas the Washington Post has made the least mention of the race of the killers. This is in line with the findings of Miranti (2014) where the Washington Post was revealed to be a conservative news agency. The mention of a person’s race in such killings is made to give him his due recognition. It is an undiscriminating strategy with the help of which their identification as “US” is ascribed to their race, specifically in the case of minorities. It is to reveal that they are accepted as members of the in-group distinction. Moreover, the highlighting of the killer’s race is done in recognition of the racist facet of the crime. The figure below shows the randomly selected concordances for blacks in USA Today.




 

Figure 2

KWIC (Key Word In Context) results for black in USA Today



In figure 2, the concordance lines 24, 25, 26, 27, 28, 29, 30, 31, 33, 34, 37 have used black as the race of the victims. Line 32 refers to a black perpetrator in the killing of a white woman. Line 35 refers to the oppressed black people. Lastly, line 36 refers to the community of black people. Line 37 is also one that has revealed the ethnicity of the white killer as well. Thus, USA Today has deliberately highlighted the race of the victims, conforming to the criteria of identifying a group as “US” as detailed in Van Dijk’s Ideological Square. The figure below illustrates the random KWIC of black in the New York Times. 













Figure 3

KWIC (Key Word In Context) results for black in the New York Times


The concordance lines 3, 4, 5, 6, 12, 13, 14, and 15 all show the ethnicity of the black victims. Line 7 refers to neighborhoods where black people live. Line 8 refers to the black American community. Line 9 refers to a jury that consists of two black jurors. Line 10 and Line 11 are among those concordance lines which have not made any mention of the victims. Wherever the race of the black victims is revealed, it is where the New York Times highlights them as in-group. Failure to observe the in-group discourse is witnessed in the lines with no mention of the black victims. Now lastly, the figure below shows the concordances of blacks in the Washington Post.
 


Figure 4
KWIC (Key Word In Context) results for blacks in the Washington Post 

In figure 4, concordance lines 1, 5, 6, 9, 10, 11, and 13 are the only lines that describe black people as victims of police shootings. Washington Post had 16 hits for blacks, however, not all concords reveal the race of the black victims. The victim's race is deliberately omitted in order to hide the racial bias of the crime and the perpetrator. It is also the newspaper with the least use of white in its discourse, again to hide the racist nature of crime and to make it just another incident of ordinary police shooting, with no involvement of ethnic minorities. This "otherization” of the black people restricts them to “out-group” in the Washington Post discourse. Van Dijk’s ideological square terms this distinction as US vs THEM. In this case, the police officers are highlighted as “US” and 
the black victims as “THEM”.
Furthering the examination of racial bias in the newspapers, the N-Grams, particularly the tri-grams and four-grams of the noun “death” were located using the Cluster/N-Grams feature of AntConc. "Death" is used for reporting the final living state of the victims. This word has a naturalizing impact on the discourse. It conceals the truth of events and discloses the incident as a completely natural matter of the victim's expiry. This advances the possibility of racial partiality and subjectivity. The figure below illustrates the most frequent N-grams of death in the corpora. Tri-grams are lexical clusters that consist of three words and four-grams consist of four words.
 


Figure 5
 N-Grams of death in the selected three corpora

The noun “death” is quite frequently adopted in the three corpora. New York Times has used it while talking about holding the perpetrator accountable for the death of the victim. It has also highlighted the shooting and referred to the investigation into the incident. In USA Today, a police officer is mentioned as one responsible for causing the death of the victim. It has also called the death "wrongful". Lastly, the USA Today corpus spoke of an incident where the victim bled to death, thus mentioning the manner of death as well. Quite contrasting to the first two corpora, the Washington Post has bluntly reported the killing incidents as events of dying. This manipulates the reality and projects racist disclosure of the crime. After this, the N-grams of the noun "murder" are illustrated in the figure below.
 












Figure 6
N-Grams of murder in the selected three corpora

Murder is the verb that reports the true nature of the crime. Murder involves two participants i.e. the murderer and the murderer who is undeniably the culprit. However, "death" completely rejects the involvement of a killer and is accredited to natural factors. When murder is used, it also reveals the intent of the malefactor to kill. Thus, it is recognized as voluntary manslaughter which is punishable by death or life sentence in most cases. Contrastive to this, the labeling of the incident as death automatically releases the killer of all charges. This difference in the implication of death and murder is of immense significance when the readers interact with them. They are readily subjugated to the covert ideologies propagated by the newspaper discourse. Montag (2017), when explaining Althusser’s Interpellation, clarifies that a person is only recognized as the subject of an ideology after he is exposed to that ideology, and when he acknowledges that the ideology was addressed to him without realizing that it is a discourse directed by the personal belief system of those in power. Hence, the employment of "death" for identification of the incident falls more under the realm of racist discourse, subjugating the masses to racist ideology, whereas "murder" conveys an impartial image and does not subject the readers to any hidden ideology. 
The corpora of the New York Times and USA Today incline more towards impartial reporting and objective stating of events, whereas contradictory to this, the Washington Post has employed the lexical tools in a manner that advocates ethnic prejudice. The word choice determines the course of the discourse. Words with more naturalizing impacts promote racist discourse and are observed in the corpus of the Washington Post with the frequent use of death and the almost negligible use of murder. Secondly, the Washington Post has been observed to have deliberately hidden the race of victims and killers in order to avoid projecting any racist aspects to the crimes. Finally, it has failed to report the crimes objectively and factually by deliberately avoiding the use of the “murder” verb to report the action of the killing. 

Conclusion

This study explored the role of newspaper discourse in disseminating racist ideologies in reporting black murders committed by white police officers in America. It examined the role of lexical items such as nouns, verbs, and adjectives in manipulating the nature of discourse. For this purpose, three corpora of ten articles each were generated from scratch for the top American newspapers, USA Today, New York Times, and Washington Post respectively. The articles were of 10 incidents of black murders which took place between the years 2014 – 2021. The corpora comprised of 25,561 words. The lexical classes, nouns, action verbs, and adjectives were comparatively analyzed for the three corpora in lieu of Van Dijk’s Ideological Square Model and Critical Discourse Analysis.  

The corpus tools AntConc 3.5.9 and LancsBox 6.0 were used for their features wordlist, concordances, N-Grams, and Whelk. The findings revealed that the newspapers, namely the New York Times and USA Today projected a pro-black discourse. On the other hand, the Washington Post advanced conservative discourse concerning black murders. It avoided the use of the verb "murder" to report the crime, hid the ethnicities of the victims and perpetrators, and normalized the black killing. It defined the incidents as deaths, thus toning down the harshness of the news report. It was found to be an active peddler of racist ideology and a reporter of subjective and biased news.  

References

• Bartesaghi, M., & Pantelides, K. (2017). Critical Discourse Analysis. The International Encyclopedia of Communication Research Methods, 1-10.
• Creswell, J. W., Plano Clark, V. L., Gutmann, M. L., & Hanson, W. E. (2003). An expanded typology for classifying mixed methods research into designs. A. Tashakkori y C. Teddlie, Handbook of mixed methods in social and behavioral research, 209-240. https://doi.org/10.1177/1558689812461179
• Egelhofer, J. L., & Lecheler, S. (2019). Fake news as a two- dimensional phenomenon: A framework and research agenda. Annals of the International Communication Association, 43(2), 97-116. https://doi.org/10.1080/23808985.2019.1602782
• Mahmood, T., Kausar, G., & Khan, G. Z. (2018). A Critical discourse analysis of the editorials of" Dawn" and" The New York Times" in the aftermath of Army Public School attack. The" Us" versus" Them" ideology. Journal of Research in Social Sciences, 6(2), 1-17. https://doi.org/10.1177/2158244020941471
• Miranti, I. (2014). Transitivity Analysis in the Construction of Newspaper Ideology: A Comparative Study on the New York Times and the Washington Times’ Editorials. Semantic Scholar. https://doi.org/10.5430/wjel.v13n6p107
• Montag, W. (2017). Althusser’s empty signifier: What is the meaning of the word ‘interpellation?’. Mediations: Journal of the Marxist Literary Group, 30(2), 63-68
• Nartey, M., & Mwinlaaru, I. N. (2019). Towards a decade of synergising corpus linguistics and critical discourse analysis: a meta-analysis. Corpora, 14(2), 203-235. https://doi.org/10.3366/cor.2019.0169
• Sajid, M. A., Anwar, B., & Ashraf, M. (2019). Politics, ideology and print media: A CDA of newspapers' headlines. Khazar Journal of Humanities and Social Sciences 22(3), 44-59. https://doi.org/10.5782/2223-2621.2019.22.3.44
• Shojaei, A., Youssefi, K., & Hosseini, H. S. (2013). A CDA Approach to the Biased Interpretation and Representation of Ideologically Conflicting Ideas in Western Printed Media. Journal of Language Teaching & Research, 4(4). https://doi.org/10.4304/jltr.4.4.858-868
• Strömbäck, J., Tsfati, Y., Boomgaarden, H., Damstra, A., Lindgren, E., Vliegenthart, R., & Lindholm, T. (2020). News media trust and its impact on media use: Toward a framework for future research. Annals of the International Communication Association, 44(2), 139- 156. https://doi.org/10.1080/23808985.2020.1755338
• Tabassum, M., Shah, S. K., & Bilal, M. (2013). A critical discourse analysis of the left and right wing ideologies in Pakistani English newspaper editorials. Journal of Education and Practice, 4(13), 72-78.
• Tanvir, O., & Zahra, S. T. (2018). Newspaper Racial Agenda: A Critical Discourse Analysis: A Critical Discourse Analysis. CORPORUM: Journal of Corpus Linguistics, 1(2), 01-16. https://doi.org/10.1590/S0102-44502006000200003
• Van Dijk, T. A. (2000). Ideology and discourse: A multidisciplinary introduction. Pompeu Fabra University, Barcelona, 1025-1034.
• Van Dijk, T. A. (2019). Ideologies, racism, discourse: Debates on immigration and ethnic issues. In Comparative perspectives on racism (pp. 91-115). Routledge. https://hdl.handle.net/11245/1.203736
• Xie, Q. (2018). Critical discourse analysis of news discourse. Theory and Practice in Language Studies, 8(4), 399-403. https://doi.org/10.17507/tpls.0804.06
• Yule, G. (2020). The study of language. Cambridge University Press.