A LINGUISTIC ANALYSIS OF FAKE NEWS IN SELECTED USER RESPONSES TO AZIMIO TV’S FACEBOOK PAGE POSTS

JOHN MOxxYI BXXONGO

A Proposal Submitted In Partial Fulfillment Of The Requirements For The Doctor Of Philosophy Degree In Applied Linguistics, Laikipia University

February 2024

1.1 Background to the Problem

The world is going through an unprecedented period of proliferation of fake news, especially on social media. The creation and spread of fake news can have dire consequences for individuals, communities, and even entire nations. For individuals, fake news can lead to misinformation that affects personal decisions, mental health, and trust in credible information sources. Communities may experience increased polarization, mistrust among neighbors, and social unrest due to misleading narratives. At the national level, the spread of fake news can undermine democratic processes, influence election outcomes, and incite violence.

For instance, the Capitol riot in the United States saw Trump's supporters violently storming the Capitol building, believing his lie that his presidential victory had been stolen. This event led to violence, destruction of property, and the deaths of at least seven people (Cameron, C. (2022, January 5)). Similarly, in Brazil, supporters of President Jair Bolsonaro attacked government federal buildings in an effort to overthrow President Luiz Inácio Lula da Silva, fueled by the fake news that Bolsonaro's election victory had been stolen (Machado, A. (2023, January 9)).

In Africa, the pervasive spread of fake news has precipitated substantial unrest and violence, deeply affecting societal stability and public trust. In Nigeria, false news reports about the activities of Boko Haram have frequently caused panic and fear among the population, leading to rushed decisions and, in some cases, deadly stampedes (Smith, J. (2021, January 15).

Regionally, the proliferation of fake news has significantly undermined responses to health crises. During the Ebola outbreak in West Africa, misinformation about the disease's origins, transmission, and cures hindered public health efforts and contributed to the virus's spread (World Health Organization, 2015). Similarly, in South Africa, fake news about COVID-19, including false cures and conspiracy theories, undermined trust in health authorities and compliance with safety measures (Reuters, 2020).

The impact of fake news has been profoundly disruptive in Kenya, leading to widespread social, political, and health-related turmoil. During the 2017 Kenyan elections, false reports and doctored images circulated on social media, stoking ethnic tensions and violence. The spread of misinformation contributed to post-election violence, resulting in deaths and widespread fear among communities (BBC News. (2017, November 12). During the 2022 general elections, misinformation campaigns were rampant, with false claims about candidates and election results circulating widely on social media platforms. These false narratives created confusion and mistrust in the electoral process, threatening the stability of the democratic process (Flamingo. 2019). The misinformation on extrajudicial arrests and killings has fueled public outrage and mistrust in law enforcement. Campaigns on social media have spread false reports of disappearances and unlawful killings by security forces, exacerbating tensions between the public and authorities (Smith, J. (2021, January 15) These false narratives not only create fear and panic but also undermine legitimate efforts to address crime and security issues in the country (Standard Media, 2022).

More recently, in Nairobi, during the COVID-19 pandemic, misinformation about the virus and its vaccines spread rapidly through social media and messaging apps. Centers for Disease Control and Prevention. (2023). False claims about vaccine side effects and conspiracy theories about the virus's origins led to vaccine hesitancy, undermining public health efforts. Additionally, false information about local incidents, such as fabricated reports of kidnappings and crimes, has caused panic among Nairobi residents, leading to unnecessary fear and mistrust within communities.

These illustrate how fake news can mobilize large groups of people to commit acts of violence and disrupt societal order, highlighting the urgent need to address and mitigate the spread of false information both globally and locally. For instance, in countries like Russia and China, the government decides what is true and what is false, irrespective of the facts (Bandurski, D. 2022, March 11).

As discussed in our literature review, several African governments have proposed or enacted media laws aimed at controlling the news available to citizens, often justifying these measures on the grounds of national security or public order. However, there is concern that these laws may not adequately consider the accuracy and reliability of the information being regulated (Mutung'u, 2018; ARTICLE 19, 2020).

McKay, G. (2022, June 22). Poll finds seven in 10 Kenyan Twitter users willing to share potentially false information for payment. “In the poll, one influencer told researchers that they worked for both sides of the political divide simultaneously. This influencer used one set of accounts to promote content favorable to the 4th President of Kenya Uhuru Kenyatta and another set reserved for content supporting Deputy President of Kenya, William Ruto now President."

Misinformation dissemination has become a prevalent issue within Kenya's digital landscape. According to a report by the (Newman, N. 2018). false rumors and manipulated images frequently circulate on social media platforms, influencing public discourse and perceptions. Reuters (2020) highlighted instances during political events where misinformation, such as false news stories and doctored images, played a significant role in shaping public opinion and contributing to social unrest. Research from the Aga Khan University (2021) underscored the widespread exposure of Kenyan social media users to misinformation related to health, politics, and societal issues, exacerbating misconceptions and fostering mistrust. Moreover, The Standard. (2022). Survey: A substantial portion of Kenyan internet users engage in sharing misinformation. Often due to a lack of fact-checking practices and susceptibility to viral content.

Finding consensus on fundamental facts is an essential principle of democracy. Voters and citizens should access accurate information to optimally make collective decisions like electing their preferred presidential candidates (Sugow, Abdulmalik, Isaac & Rutenberg, 2021). Although arriving at an exclusive 'objective truth' is impossible due to mediation in communication, Kenyans need to access and recognize the rudimentary facts underpinning political processes. The rising spread of inaccurate or false content denotes the solidifying of a fake news era where political rhetoric usually appeals to sentiment and emotion with minute respect to factual rebuttals (Prinanda, 2019). Feeding the public with misleading or false information entails deceiving and gaining a particular advantage. In a fake news environment, people re-engineer language to not only conceal but also sanitize deceptions and dishonesty (Mukhongo, 2020).

The proposed study aims to analyze the linguistic characteristics of fake news, in user responses, to factual Facebook posts in Kenya. Nelson & Taneja, (2018) note that the ‘fake news’ crisis has been one of the most discussed topics in both public and scientific discourse since the 2016 U.S. presidential campaign. The Oxford American Dictionary defines fake news as false information that is broadcast or published as news for fraudulent or politically motivated purposes.

For the proposed study, fake news will refer to the intentional misrepresentation of factual news as being untrue. It will analyze all the systematic linguistic devices employed to deceive other unsuspecting users into distrusting the factual information presented in the posts in favor of alternative untrue or unrelated narratives.

Truth holds a revered place in liberal democracies since society always turns to facts for redemption whenever democracy appears skewed or when politicians manipulate voters. However, the rise of fake news and concerted efforts to brand factual news as being fake is cause for consternation as the world plunges into the so-called post-truth era. The proliferation of fake news on social media platforms has become a pressing concern in modern societies, often influencing public opinion and potentially jeopardizing the integrity of democratic processes. In Kenya, a country where political discourse is vibrant and social media plays a significant role in shaping public perception, the emergence of fake news poses a unique challenge. A“disinformation industry” is blossoming in Kenya, where social media influencers and ordinary Kenyan citizens share false or misleading narratives intended to sway public opinion and influence political decisions. Chapter 6 of the Kenyan constitution provides guiding principles of leadership and integrity where among others, public officers are expected to offer selfless service based solely on the public interest, and demonstrated by honesty in the execution of public duties; accountability to the public for decisions and actions; and discipline and commitment in service to the people (Kenya, 2010). The employment of armies of bloggers or social media influencers to spread untrue information or exaggerate the achievements of the government while denigrating the opposition or vice versa goes against the stated principles of the Constitution.

"Subsequently, our literature review will delve into discussions by researchers such as Coady, D. (2021), Duffy, A., Tandoc, E., & Ling, R. (2019), Mutai, P., and Kimari, B., (2020), Mwita, J. (2020), and Maweu, J. M., (2019) have discussed the concept of fake news. However, a look at the existing research on this topic reveals a dearth or gap when it comes to employing linguistic tools to analyze fake news posts as reflected in Kenya’s social media landscape and using the findings to develop a linguistic tool for detecting fake news. The proposed study endeavors to bridge this gap.”

1.2 Problem Statement

Despite increasing efforts to combat misinformation, the proliferation of fake news persists, particularly evident in user responses to Azimio TV's Facebook page posts. This study aims to analyze the linguistic features and strategies employed in these responses to understand the dissemination and reception of fake news within this online community.

In the digital age, the proliferation of misinformation, commonly known as fake news, presents a significant challenge to public discourse and media integrity. Social media platforms, including Facebook, have become prominent arenas where misinformation spreads rapidly and influences public opinion. Azimio TV, a prominent media outlet, faces the challenge of managing user-generated content on its Facebook page, where responses to posts often include instances of fake news. This study seeks to conduct a comprehensive linguistic analysis of user responses to Azimio TV's Facebook page posts to uncover the linguistic features, rhetorical strategies, and discursive patterns utilized in the dissemination and reception of fake news. By examining how language is manipulated and interpreted in this context, the research aims to contribute to a deeper understanding of the mechanisms through which fake news propagates and its impact on media credibility and public perception. Ultimately, this analysis aims to inform strategies for mitigating the spread of misinformation on social media platforms.

1.3 Objectives

a) Analyze the linguistic characteristics of fake news posts on Facebook during anti-government protests in Kenya to uncover specific markers and patterns used in misleading information dissemination.

b) Study how language is manipulated in fake news posts to deceive or mislead, aiming to develop a linguistic model for detecting fake news that can assist in stemming its spread.

c) Propose practical strategies based on linguistic analysis findings to mitigate the impact of fake news during anti-government protests in Kenya, benefiting stakeholders interested in combating misinformation.

.

1.4 Research Questions

a) What specific linguistic features and patterns characterize fake news posts circulated on Facebook during anti-government protests in Kenya? How do these linguistic features contribute to the persuasive and deceptive nature of fake news during such socio-political events?

b) How do linguistic manipulation techniques in fake news posts during anti-government protests in Kenya contribute to their effectiveness in misleading or influencing public perception?

c) How can insights from the linguistic analysis of fake news during anti-government protests in Kenya be leveraged to develop effective strategies for detecting and mitigating the spread of misinformation on social media platforms like Facebook?

1.5 Literature Review

Egelhofer, J. L., & Lecheler, S. (2019) distinguish between the genre and label of fake news, focusing on its creation and instrumentalization. Coady, D. (2021) critiques the term 'fake news' as a tool for restricting speech, emphasizing the distinction between factual accuracy and free expression.

Egelhofer, J. L., & Lecheler, S. (2019) suggest that 'fake news' alludes to two dimensions of political communication: the fake news genre (i.e. the deliberate creation of pseudo journalistic disinformation) and the fake news label (i.e. the instrumentalization of the term to delegitimize news media). They note that while public worries about the use of the label by politicians are increasing, scholarly interest is heavily focused on the genre aspect of fake news. They connect the existing literature on fake news to related concepts from political communication and journalism research and present a theoretical framework to study fake news bringing clarity to the discourse about fake news and shifting scholarly attention to the neglected fake news label. Though their research helps draw a line between fake news as a genre and label, our study will focus on fake news as a genre with the sole focus on the analysis of its linguistic characteristics.

Coady, D. (2021) avers that governments and other powerful institutions around the world have adopted a variety of measures to restrict the reporting and dissemination of claims they deem to be fake news. He notes that many of these measures are clear breaches of fundamental rights, including freedom of speech and freedom of the press. He points out that contrary to common opinion, there is no new or growing problem of fake news but a new and growing problem caused by the term ‘fake news’. He argues that the function of the term 'fake news' is to restrict permissible public speech and opinion in ways that serve the interests of powerful people and institutions. While the proposed study emphasises the importance of uninhibited freedom of speech without consideration as to whether what people write or say is factual or false, our research will not delve into the pros and cons of freedom of expression; but will be limited rather to the objective analysis of fake news posts to identify their underlying linguistic elements.

Conroy-Krutz, J. (2020) discusses legislative measures against fake news in Tanzania, highlighting its impact on media freedom. Mwita, J. (2020) examines fake news in Kenyan elections, emphasizing its role in political polarization and media literacy gaps.

Conroy-Krutz, J. (2020, April) notes that Tanzania has seen a particularly large number of laws instituted in recent years that criminalize not only fake news per se but also what the state considers as being in opposition to the state's official narrative. The Statistics Act of 2015 criminalizes the publication not only of "false statistics," but also of any statistical information not authorized for release by the National Bureau of Statistics. The Cybercrimes Act of the same year bans the dissemination of "false, deceptive, misleading, or inaccurate information" and insulting or inflammatory speech. The writer notes that, in 2017, the government updated the broadcasting-services code to require that television programs include only such information as the government deems "interesting and relevant" while avoiding “inflammatory, defamatory and divisive matter.” Print journalists, meanwhile, cannot “breach peace” or “hurt the feelings of any person.” Common sense dictates that the only information or news that the Tanzanian government would find ‘interesting and relevant’ are those that are pro-government irrespective of their factual accuracy.

Mwita, J. (2020) notes that fake news has dominated the media debate the world over in recent times and points to its use in political discourses to portray opponents as inefficient, alienated, and outsiders. The researcher considers the Kenya election on August 8, 2017, as being dominated by fake news and propaganda with campaigns being highly polarised. The writer notes the lack of Media and Information Literacy (MIL) programs to stem the proliferation of fake news. The writer analyzed the use of fake news and propaganda in political campaigns leading to the August 8 elections of 2017 in Kenya and examined the effects of fake news in the Kenyan political landscape through collecting, viewing, and analyzing fake news and propaganda in political campaigns discourses leading to the August 8 general elections and recommended Media and Information Literacy as a remedy to combat fake news and negative propaganda. Our proposal hopes to supplement media and information literacy with a linguistic template for detecting fake news by examining the linguistic characteristics of fake news posts on Facebook

Duffy, A., Tandoc, E., & Ling, R. (2019) explore the impact of fake news on interpersonal relationships, particularly in social media contexts.

Duffy, A., Tandoc, E., & Ling, R. (2019) argue that while fake news has been widely reviled as an attack on democracy, less has been written about its threat to interpersonal relationships. They add that social networks have become increasingly popular for sharing news and as a result have also offered fertile ground for the spread of fake news. Their paper considers the impact of the latter on the former, particularly in circumstances where the sharer either does not know or does not suspect that the news they are sharing is fake. In their opinion, this distinction is important because while sharing information and news may be construed as a social good, sharing news that turns out to be fake might negatively impact relationships.

Their paper examines how people react when the news they have shared intending to foster social cohesion turns out to be fake, and as a result, damages that cohesion. Our proposed study is quite different because we hypothesize that the fake news posts we will analyze are aimed at primarily damaging social cohesion in Kenya. We assume that most of the actors engaging actively in posting fake news use fake social media handles hence do not expect that their friends or families will suspect that they are the ones disseminating fake news. While the issue of social cohesion or divide is beyond the confines of our proposed study, we hope that a provision of a linguistic template based on our study's findings will be beneficial to social media users, and other stakeholders, interested in detecting whether content that has been shared is factual or fake

Kessler, G., Rizzo, S., & Kelly, M. (2021) quantify Trump's false claims and their dissemination through social media, underscoring their global impact. Sousa-Silva (2022) assesses the prevalence of misinformation during the Trump administration and its implications for public discourse.

Donald Trump, as president of the United States states, may have played an oversized role in the promotion of fake news by his rejection of any news or facts that were not favorable to him by declaring them as fake. Trump might be the first US president to openly lie without

shame or fear of consequences. For example, Sousa-Silva (2022) asserts that approximately 70% of former president Trump's 'factual' avowals typically fall into 'false,' 'pants on fire,' and 'mostly false' falsehood classes. Kessler, G., Rizzo, S., & Kelly, M. (2021, January 24) note that during his four years as president, Trump made 30,573 false or misleading claims mostly on social media. When he was banned from major social media platforms for spreading lies

and propaganda that led to the storming of the US Capitol on January 6, 2021; he started his own media company where Trump and his followers post whatever they want without any

constraints. It is telling that for a man famous for spreading lies, conspiracy theories and falsehoods; he chose Truth Social as the name for his social media platform.

False or misleading claims on social media are not a problem just for the United States.

They were also common during the Brexit referendum as some politicians claimed that

European Union (EU) participation cost Britain 350 million pounds weekly (Sousa-Silva, 2022).

However, in highlighting this supposed ‘truth’, they conveniently failed to justify the money

received by the country from the EU in return. Such events are examples of the world

entering the uncharted era of post-truth politics, where politicians and their supporters

shape public opinion by appealing to emotion and personal beliefs, rather than to objective facts.

Mutai, P., & Kimari, B. (2020) analyze the use of social media and fake news in Kenyan elections, focusing on strategic misinformation tactics. Maweu, J. M. (2019) examines the role of social media in spreading disinformation during Kenyan elections, highlighting specific instances of cyber propaganda.

Mutai, P., and Kimari, B., (2020) have noted that similar to previous polls, the 2017 elections in Kenya were closely contested in all the seats and that during the 2017 polls, social networking sites were widely employed, including in campaigns where candidates not only set up websites but also employed bloggers and social media managers to manage their social media accounts and constantly post their campaign messages. The writers note that, while social media fostered access to important information on the elections, it was also used to spread fake news intended mainly to win over voters, create fear and alarm, and sometimes disparage some of the independent institutions that were managing the elections. Using data collected during and after the August 8, 2017, General Election and the October 26, 2017 repeat presidential contest, they examine how fake news was used to advance different political agendas. They examine the nature of the fake news during the 2017 elections and the implications of the spread of fake news in the 2017 elections. Our study hopes to go beyond examining the nature of fake news by presenting an applicable model for detecting fake news through easily identifiable linguistic characteristics of such news.

Maweu, J. M., (2019) examines how social media platforms (Twitter, Facebook, and WhatsApp) were used to spread disinformation and cyber propaganda during the 2017 general elections in Kenya. The writer notes that the 2017 general elections in Kenya were some of the most competitive, and tense, and elicited very active use of digital media by political leaders and citizens alike bringing with it increased spread of disinformation and fake news, which cast a shadow on the integrity of the election outcome. The writer notes that in March 2018, details emerged of the claims by Cambridge Analytica that it had engineered a digital campaign that painted Uhuru Kenyatta in positive light while smearing the image of his main rival, Raila Odinga. The writer argues that the contested results for the two presidential elections in August and October 2017 were largely dismissed as fake computer-generated results due to the alleged cyber propaganda and the extensive spread of disinformation. The writer’s focus is on how social media platforms were used to spread disinformation and cyber propaganda during the 2017 general elections in Kenya. The study will extend its focus to identify discernible linguistic indicators of fake news on social media, utilizing a case study approach centered on Facebook’s Azimio TV.

While existing literature extensively covers the political, social, and psychological dimensions of fake news globally and regionally, there remains a gap in understanding the linguistic characteristics of fake news specifically within the context of Kenyan social media, such as on platforms like Facebook. This study aims to address this gap by providing a linguistic template for detecting fake news posts and contributing to media and information literacy efforts in Kenya.

1.6 Theoretical Orientation

A linguistic analysis of fake news posts on social media can benefit from several linguistic theories and approaches, including discourse analysis, pragmatics, semantics, sociolinguistics, Corpus Linguistics, Critical Discourse Analysis, Computational Linguistics, Framing Theory, and forensic stylistics. For the proposed study, we will employ both Forensic Stylistics Theory and Computer-Mediated Discourse Analysis (CMDA).

1.6.1 Forensic Stylistics Theory

Forensic stylistics focuses on analyzing the linguistic features, patterns, and stylistic choices employed by individuals in their written communication (McMenamin, 2010). When applied to the study of fake news on social media, the theory of forensic stylistics can help identify and analyze linguistic characteristics used by different users. This will be particularly useful in our proposed study for answering the second research question, which seeks to identify discernible linguistic indicators of fake news.

Forensic stylistics provides a methodological approach to systematically analyze language use, examining elements such as syntax, vocabulary, and rhetorical strategies. This approach can uncover distinctive stylistic markers that differentiate genuine news from fake news, thereby offering valuable insights into the linguistic construction of misinformation. However, forensic stylistics primarily focuses on static texts and individual authorship, often lacking the tools to account for the dynamic, interactive, and multimodal nature of social media communication.

1.6.2 Computer-Mediated Discourse Analysis (CMDA)

To address the limitations of forensic stylistics, we incorporate Computer-Mediated Discourse Analysis (CMDA) which comes in as a compliment. The premise of CMDA is that the choice of words and expressions is potentially significant beyond the requirements of lexicon and grammar. It seeks to identify patterns in language structure and use that may have been produced unconsciously, yet shed light on broader phenomena such as decision-making, gender ideology, cultural identity, and the social construction of knowledge (Herring, S. (2019). CMDA is an approach rather than a theory, allowing diverse theories about discourse and computer-mediated communication to be entertained and tested. It applies methods adapted from language-focused disciplines such as linguistics, communication, and rhetoric to the analysis of Computer-Mediated Communication (Herring, 2004).

CMDA can involve qualitative or quantitative analysis, including coding and counting. At its core, it involves the analysis of logs of verbal interaction (for example, characters, words, utterances, messages, exchanges, threads, and archives). Herring (2019) argues that any analysis of online behavior grounded in empirical, textual observations constitutes CMDA. This approach aligns well with our proposed analysis of Facebook posts.

Regarding the implementation of the coding and counting approach to CMDA, Herring, S. C. (2019) outlines a five-step process:

1. Articulate research question(s)

2. Select a computer-mediated data sample

3. Operationalize key concepts in terms of discourse features

4. Apply method(s) of analysis to the data sample

5. Interpret results

Jones, R., & Hafner, C. (2020) also developed the Faceted Data Classification Scheme for CMD, which synthesizes and articulates aspects of technical and social context that influence discourse usage in CMC environments. Elements of this scheme include medium/technological features of CMC systems and social factors conditioning CMD.

Medium Factors

Medium factors are determined by messaging protocols, servers, and clients, as well as the associated hardware, software, and interfaces of users’ computers. Relevant factors include:

Table 1. Medium Factors

M1 Synchronicity

M2 Message Transmission (1-way vs 2-way)

M3 Persistence of transcript

M4 Size of message buffer

M5 Channels of communication

M6 Anonymous messaging

M7 Private Messaging

M8 filtering

M9 quoting

M10 Message format

Situation Factors

Social factors include information about participants, their relationships, their communication

purposes, the topics of communication, and the kind of language used. Relevant factors include:

Table 2. Situation Factors

In our proposed study, we intend to select medium and situational factors that are relevant and helpful to the analysis of the linguistic characteristics of fake news. Hence, medium factors such as M1, M3, M4, M7, and M8, and situational factors such as S1, S2, S3, S4, and S5 will be utilized. For example, In the study of fake news on social media, M1 (Synchronicity) could involve examining how real-time interactions (or lack thereof) influence the dissemination and reception of misinformation. M2 (Message Transmission) might focus on whether communication channels allow for interactive engagement or merely one-way dissemination of information. S1 (Participant Information) could analyze how the identities and affiliations of participants shape the content and credibility of fake news posts, while S2 (Purpose of Communication) might explore how differing intents—such as influencing political opinions or inciting social division—affect the linguistic strategies employed in disseminating fake news. These factors collectively provide insights into the contextual nuances and social dynamics influencing the spread of misinformation on platforms like Facebook's Azimio TV page.

The medium factors are integral to both Forensic Stylistics Theory and CMDA as they provide the context in which communication occurs while situation factors, such as participants' relationships, communication purposes, and contextual language use, are critical in analyzing the social dynamics and contextual influences on the discourse patterns of fake news on social media. Social factors that condition CMD include the information about the participants, their relationships to one another and their purposes for communication, what they are communicating about, and the kind of language they use to communicate (Baym, 1995; Hymes, 1974). These situation factors assume that context can shape communication in significant ways, without necessarily assuming that any one given factor is always influential

By integrating Forensic Stylistics Theory and CMDA, this study aims to provide a comprehensive framework for analyzing the linguistic and communicative strategies of fake news. This dual-theoretical orientation will enhance our understanding of how fake news is constructed and disseminated, offering valuable insights into the linguistic features that characterize misinformation on social media.

1.6.3 Conceptual Framework

To effectively analyze the linguistic characteristics of fake news on social media, the integration of Forensic Stylistics Theory and Computer-Mediated Discourse Analysis (CMDA) is essential. This conceptual framework outlines how these two theories will be combined to provide a comprehensive approach to our study.

Integration of Theoretical Orientations

Forensic Stylistics Theory

Focus: Detailed linguistic analysis of individual texts, examining elements such as syntax, vocabulary, and rhetorical strategies.

Strengths: Identifies distinctive stylistic markers that differentiate genuine news from fake news, providing insights into the linguistic construction of misinformation.

Limitations: Primarily focuses on static texts and individual authorship, lacking tools to account for the dynamic, interactive, and multimodal nature of social media communication.

Computer-Mediated Discourse Analysis (CMDA)

Focus: Identifies patterns in language structure and use in computer-mediated communication, considering broader phenomena such as decision-making, cultural identity, and social construction of knowledge.

Strengths: Addresses the dynamic, interactive, and multimodal nature of social media, providing a comprehensive analysis of online behavior grounded in empirical, textual observations.

Limitations: Requires detailed classification and operationalization of discourse features, which can be complex and time-consuming.

Conceptual Framework Diagram

Below is a conceptual framework diagram illustrating how Forensic Stylistics Theory and CMDA will work together in our study:

1. Articulate Research Questions

o Forensic Stylistics: Identifying linguistic markers of fake news.

o CMDA: Understanding how these markers function in the context of social media interactions.

2. Select Data Sample

o Collect fake news posts from Facebook’s Azimio TV page.

o Ensure a representative sample that captures the diversity of linguistic features and interaction patterns.

3. Operationalize Key Concepts

o Forensic Stylistics: Define and identify specific linguistic features (e.g., syntax, vocabulary, rhetoric).

o CMDA: Define and identify interaction patterns and contextual factors (e.g., synchronicity, participant relationships).

4. Apply Methods of Analysis

o Forensic Stylistics: Conduct detailed linguistic analysis of individual posts.

o CMDA: Analyze interaction patterns and contextual factors to understand how linguistic features are used in social media communication.

5. Interpret Results

o Synthesize findings from both approaches to provide a comprehensive understanding of fake news on social media.

o Identify linguistic cues that characterize fake news and explain how these cues function within the dynamic context of social media.

Conceptual Framework Diagram(s)

Diagram 1.

Diagram 2.

Implementation Steps

1 Articulate Research Questions

Forensic Stylistics: What are the specific linguistic markers that differentiate fake news from genuine news on social media?

CMDA: How do these linguistic markers function in the context of social media interactions, and what patterns emerge?

2 Select Data Sample

Gather a diverse and representative sample of fake news posts from Facebook’s Azimio TV page.

3 Operationalize Key Concepts

Forensic Stylistics: Define linguistic features such as syntax, vocabulary, and rhetorical strategies.

CMDA: Define interaction patterns and contextual factors such as synchronicity, participant relationships, and communication purposes.

4 Apply Methods of Analysis

Conduct a detailed linguistic analysis of individual posts using forensic stylistics.

Analyze interaction patterns and contextual factors using CMDA to understand the broader context of social media communication.

5 Interpret Results

Synthesize findings from both approaches to provide a comprehensive understanding of the linguistic characteristics of fake news on social media.

Identify and explain how specific linguistic cues are used in the context of social media interactions, offering insights into the construction and dissemination of misinformation.

By integrating Forensic Stylistics Theory and CMDA, this study will bridge the gap between the detailed linguistic analysis of individual texts and the dynamic, interactive context of social media. This dual-theoretical orientation will enhance our understanding of how fake news is constructed and disseminated, providing valuable insights into the linguistic features that characterize misinformation on social media.

1.7 Significance of the Study

Language is the universal tool of communication. We tell truths or lies using language. Those who create and disseminate fake news use language. The proposed study aims thus to provide valuable information on the characteristics of the linguistic choices employed by people

specifically, when they seek to spread fake news or claims (by rebranding factual news as being fake, or spreading falsehoods as being factual) on Facebook.

For the general internet consuming public, the proposed study will provide a rubric that will be helpful in discerning fake news. The study will also be helpful to stakeholders interested in stemming the spread of fake news on social media.

By employing both Forensic Stylistics Theory and Computer-Mediated Discourse Analysis (CMDA), this study will provide a detailed and nuanced understanding of the linguistic features and interaction patterns that characterize fake news on social media. This dual-theoretical approach will allow for a comprehensive analysis that bridges the gap between static text analysis and the dynamic nature of online communication enhancing linguistic understanding. This will greatly foster to the identification of specific linguistic markers and interaction patterns associated with fake news which will contribute to the development of practical tools and methods for detecting and combating misinformation on social media. These tools can be employed by social media platforms, fact-checking organizations, and policymakers to identify and address fake news more effectively.

The findings of this study will provide valuable insights for media literacy education programs. By understanding the linguistic characteristics of fake news, educators can better equip individuals with the skills needed to critically evaluate the information they encounter online, fostering a more informed and discerning public thus contributing to media literacy. The study's findings can inform policymakers and regulatory bodies in developing more effective strategies and policies to combat the spread of fake news. By providing a detailed analysis of the linguistic features and dissemination patterns of fake news, this research can support the creation of targeted interventions that address the root causes and mechanisms of misinformation. In the context of Kenya, where fake news has been shown to influence political discourse and social cohesion, this study will provide crucial insights into how misinformation is constructed and spread. By focusing on the linguistic aspects of fake news on Facebook’s Azimio TV page, this research will shed light on the specific ways in which language is used to manipulate public opinion and incite social division, thereby giving lucid guidelines on law making and policy shaping thus contributing to broader efforts to promote social harmony and political stability.

Academic Contribution: This study will add to the body of knowledge in the fields of linguistics, media studies, and political communication. Integrating forensic stylistics and CMDA, it will offer a novel methodological approach that can be applied to future research on fake news and other forms of online communication. The theoretical and empirical insights gained from this study will be valuable for scholars and researchers exploring the intersection of language, media, and misinformation.

The significance of this study lies in its potential to enhance our understanding of fake news through a comprehensive linguistic analysis, developing practical tools for its detection, informing media literacy and policy efforts, and contributing to academic research and social stability.

1.8 Methodology

This study adopts a sequential explanatory mixed-methods design to analyze the linguistic characteristics of fake news on social media, particularly focusing on Facebook's Azimio TV page (Creswell & Creswell, 2018). Building on insights from forensic stylistics theory and Computer-Mediated Discourse Analysis (CMDA) as discussed in the literature review, the sequential design involves an initial qualitative phase followed by a quantitative phase, allowing for a comprehensive exploration of the research questions. The first phase, qualitative analysis involves selecting and coding a sample of fake news posts based on identified linguistic features and patterns. Subsequently, a quantitative analysis will be conducted to quantify the prevalence and distribution of these linguistic characteristics across the dataset. Medium factors such as synchronicity, persistence of transcript, size of message buffer, and private messaging, along with situational factors including participant information and purpose of communication, will guide the selection and analysis of data. This methodological approach aims to provide a comprehensive understanding of how linguistic strategies contribute to the dissemination and reception of fake news on social media platforms.

The sampling procedure will involve purposive sampling of fake news posts from Facebook's Azimio TV page. Sampling techniques like Network Sampling involve identifying key nodes (users, pages, groups) within the social network (e.g., Facebook) that are known to propagate fake news. By selecting nodes strategically, researchers can capture a diverse range of fake news content and behaviors across the network (Wasserman & Faust, 1994). Time-series sampling involves capturing data over time to provide insights into how fake news evolves and spreads. Researchers can select specific time intervals (e.g., before and after significant events or policy changes) to observe changes in fake news patterns and behaviors (Wang & Kosinski, 2020). Initially, a qualitative sampling approach will select posts based on identified criteria from the literature review, focusing on linguistic features and patterns indicative of fake news (Flick, U. (2018). Following qualitative analysis, a quantitative sample will be drawn to statistically analyze the prevalence and distribution of these linguistic characteristics across a broader dataset. The primary data set will consist of fake news posts collected from Facebook's Azimio TV page during a specific period. The posts will be systematically archived and coded for linguistic analysis, incorporating identified medium and situational factors (Miles, Huberman, & Saldaña, 2019).

The study employs a systematic sampling approach to select fake news posts from Facebook's Azimio TV page, ensuring representation across different types and categories of fake news (Creswell & Creswell, 2018). This methodological choice allows for the systematic collection of data that is relevant to the linguistic analysis outlined in the literature review (Patton, 2015).

Data collection will be conducted using automated scraping tools, which systematically gather fake news posts based on predefined criteria, such as keywords and temporal parameters (Sloan & Quan-Haase, 2017). This approach aligns with systematic sampling principles, ensuring comprehensive coverage of relevant content over a specified period (Wang & Kosinski, 2020). To validate the reliability and validity of the sample, rigorous criteria will be applied to identify and exclude irrelevant or non-conforming posts. This process ensures that the selected data accurately reflects the linguistic characteristics and dissemination behaviors of fake news on social media (Zubiaga, Aker, & Bontcheva, 2018).). Ethical considerations will guide the handling and analysis of data, particularly concerning privacy and the potential impact of studying sensitive content. Transparency in data collection methods and adherence to ethical guidelines are paramount in maintaining the integrity of the research findings (Ess & AoIR Ethics Working Committee, 2019).

The study adopts a purposive sampling approach to select fake news posts from Facebook's Azimio TV page, guided by the need to capture diverse linguistic characteristics and dissemination patterns (Creswell & Creswell, 2018). This method allows for the targeted selection of posts that exhibit specific linguistic features identified in the literature review, ensuring relevance to the research objectives (Sloan & Quan-Haase, 2017)

.

Instruments to be used for the study include a Qualitative Coding Framework, which will be developed based on insights from forensic stylistics theory. The coding framework will identify linguistic features and patterns in fake news posts (Hoover & Walker, 2017). The quantitative Analysis Tools, are statistical software that will be used to analyze quantitative data, focusing on frequencies and correlations of linguistic characteristics across the dataset.

The Forensic Stylistics Theory will be applied in the qualitative phase to identify and analyze linguistic features, patterns, and stylistic choices in fake news posts (Hoover & Walker, 2017). This theory guides the initial qualitative coding process to uncover nuanced linguistic elements. Meanwhile, Computer-Mediated Discourse Analysis (CMDA) will be utilized in both qualitative and quantitative phases to examine language structure and use in digital communication contexts (Herring, 2018). CMDA informs the systematic analysis of medium factors (e.g., synchronicity, persistence of transcript) and situational factors (e.g., participant information, purpose of communication) relevant to the study of fake news on social media.

Justification:

The Research Design (The mixed-methods design) facilitates a comprehensive exploration of fake news linguistic characteristics, combining qualitative depth with quantitative breadth (Creswell & Creswell, 2018). The sampling Procedure (purposive sampling) ensures the selection of relevant data for in-depth linguistic analysis, aligning with theoretical frameworks and research objectives (Patton, 2015). The data sets (focus on Facebook's Azimio TV page) provide a specific context for studying fake news dissemination, enhancing the study's applicability and relevance (Miles et al., 2019). Finally, the Instruments used (qualitative coding frameworks and statistical analysis tools) supports rigorous data interpretation and validation of findings across phases of the study.

This methodology aims to contribute nuanced insights into the linguistic dimensions of fake news on social media, informed by theoretical frameworks and methodological rigor.

References

Bandurski, D. (2022, March 11). China and Russia are joining forces to spread disinformation. Brookings.

Conroy-Krutz, J. (2020, April). The squeeze on African media freedom. Journal of Democracy.

Coady, D. (2021). The fake news about fake news. In The Epistemology of Fake News (pp.

68–81). Oxford University PressOxford. http://dx.doi.org/10.1093/oso/9780198863977.003.0004

Coulthard, M., May, A., & Sousa-Silva, R. (2020). The Routledge Handbook of Forensic Linguistics. Routledge.

Duffy, A., Tandoc, E., & Ling, R. (2019). Too good to be true, too good not to share: The social utility of fake news. Information, Communication & Society, 23(13), 1965–1979. https://doi.org/10.1080/1369118x.2019.1623904

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

Giusti, S., & Piras, E. (2020). Democracy and fake news: Information manipulation and post-truth politics. Routledge.

Herring, S. C. (2001). Computer-mediated discourse. In D. Schiffrin, D. Tannen, & H.

Hamilton (Eds.), The Handbook of Discourse Analysis (pp. 612-634). Oxford: Blackwell Publishers. https://homes.luddy.indiana.edu/herring/cmd.pdf

Herring, S. C. (2004). Computer-Mediated Discourse Analysis: An Approach to Researching Online Behavior. In S. A. Barab, R. Kling, & J. H. Gray (Eds.), Designing for virtual

communities in the service of learning (pp. 338–376). Cambridge University Press. https://doi.org/10.1017/CBO9780511805080.016

Herring, S. C. (2007). A faceted classification scheme for computer-mediated discourse. Language@Internet. http://www.languageatinternet.org/articles/2007/761

Kessler, G., Rizzo, S., & Kelly, M. (2021, January 24). Trump’s false or misleading claims total 30,573 over 4 years. The Washington Post. https://www.washingtonpost.com/politics/2021/01/24/trumps-false-or-misleading-claims-tota l-30573-over-four-years/

Kozinets, R. V. (2009b). Netnography: Doing ethnographic research online. SAGE.

Kenya: The Constitution of Kenya [Kenya], 27 August 2010, available at: https://www.refworld.org/docid/4c8508822.html [accessed 3 February 2024]

Maweu, J. M., (2019) “Fake Elections”? Cyber Propaganda, Disinformation and the 2017 General Elections in Kenya, African Journalism Studies, 40:4, 62-76, DOI:10.1080/23743670.2020.1719858

McKay, G. (2022, June 22). Disinformation and democratic transition: A Kenyan case study

•Stimson Center. Stimson Center.

/

Mcmenamin, G. (2010). Forensic stylistics Theory and practice of forensic stylistics.

Moravec, P., Minas, R., & Dennis, A. R. (2018). Fake News on Social Media: People Believe What They Want to Believe When it Makes No Sense at All. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3269541

Mukhongo, L. L. (2020). Participatory Media Cultures: Virality, Humour, and Online Political Contestations in Kenya. Africa Spectrum, 55(2), 148-169. https://doi.org/10.1177/0002039720957014

Mutahi P., & Kimari B., (2020) Fake News and the 2017 Kenyan Elections, Communicatio, 46:4, 31-49, DOI: 10.1080/02500167.2020.1723662

Mwita, J. (2020). Fake News and Propaganda in Political Communication: Effects and Remedies. East African Journal of Interdisciplinary Studies, 2(1), 149-156. https://doi.org/10.37284/eajis.2.1.244

Nelson, J. L., & Taneja, H. (2018). The small, disloyal fake news audience: The role of

audience availability in fake news consumption. New Media & Society, 20(10), 3720–3737.

Santuraki, S. U. (2018). Anti-fake news legislations and free speech: A comparative evaluation of the trends in Germany, Malaysia, and Kenya. Proceedings of the International Law Conference. http://dx.doi.org/10.5220/0010049502960305

Smartt, U. (2020). Social media and fake news. In Media Law for Journalists (pp. 191–206). Routledge. http://dx.doi.org/10.4324/9780367855505-8

Sousa-Silva, R. (2022). Fighting the fake: A forensic linguistic analysis to fake news detection. International Journal for the Semiotics of Law-Revue internationale de Sémiotique juridique, 35(6), 2409-2433.

Sugow, Abdulmalik, and Isaac Rutenberg (2021). Securing Kenya Electoral Integrity: Regulating Truth. Elephant, (2021) https://www.theelephant.info/analysis/2021/10/22/securing-kenyas-electoral-integrity- regulating-truth/

Talwar, S., Dhir, A., Kaur, P., Zafar, N., & Alrasheedy, M. (2019). Why do people share fake news? Associations between the dark side of social media use and fake news sharing behavior. Journal of Retailing and Consumer Services, 51, 72–82. https://doi.org/10.1016/j.jretconser.2019.05.026

Talwar, S., Dhir, A., Singh, D., Virk, G. S., & Salo, J. (2020). Sharing of fake news on social media: Application of the honeycomb framework and the third-person effect hypothesis.

Journal of Retailing and Consumer Services, 57, 102197. https://doi.org/10.1016/j.jretconser.2020.102197