Normative Needs in the Study of Elections and Campaigns[1]
Heesoo Jang, Department of Journalism, University of Massachusetts Amherst
Shannon C. McGregor, Hussman School of Journalism & Media and Center for Information, Technology, and Public Life – University of North Carolina, Chapel Hill
http://dx.doi.org/10.17169/refubium-43532; PDF
Challenges to democracies from the far right, authoritarians, and fascists are not new—but they are on the rise globally. Yet, because many of our theories and models for researching communication were developed on the basis of “stable” democracies in the West (as so eloquently addressed in the previous issue of the Political Communication Report), our field as a whole lacks a cohesive democratic normative framework with which to interpret these threats to democracy. Nowhere is this more apparent than in the study of elections and campaigns. Adopting a normative democratic basis means that we must re-think how we study the myriad of communication as it relates to elections and campaigns. In this piece, we discuss this across the press, campaigns, technology, and specifically AI.
The Press
In a democracy, the role of the press transcends mere reporting to encompass a robust assessment of how well the democratic process is being upheld. Journalism, with its foundational premise of informing the public, is crucial in evaluating and critiquing the conduct of elections and the behaviors of those who contest them. This democratic assessment involves not only tracking the accuracy and fairness of the electoral process but also ensuring that the portrayal of candidates and campaigns aligns with democratic norms, thus safeguarding the public’s right to factual, unbiased, and constructive information.
According to a report by The Washington Post (Blanco, Wolfe, and Gardner, 2022), during the 2022 U.S. midterm elections, 291 Republican candidates at state and federal levels challenged or doubted the outcome of the 2020 presidential election. Out of these, 179 candidates were successful in their electoral bids. Such actions pose a significant risk to the foundations of democracy. A functional democracy hinges on the ability to conduct credible, competitive elections—if a political leader or party cannot accept defeat, it compromises the peaceful transition of power.
The situation in the United States is one among several recent global examples where competitive elections have faced threats, marking a critical and notable instance. Over the last ten years, democratic systems worldwide have been increasingly vulnerable. In 2020, autocratic behaviors affected 25 countries, impacting 34 % of the global population. In contrast, only 16 countries, representing a mere 4 % of the global populace, made strides toward democratization (Hellmeier et al., 2021).
The emergence of anti-democratic candidates in the U.S. and many other countries globally presents a profound challenge to journalists and newsrooms committed to democratic ideals. For instance, comments by former president Donald Trump regarding using the military to conduct mass deportations were labeled as “extreme” by The New York Times and PBS, with further descriptions like “radical” by CNN and “boundary-pushing right-wing policies” by TIME. Reuters highlighted the use of “dehumanizing terminology,” noting an echo of xenophobic and Nazi rhetoric. This type of language tests journalistic norms in covering campaign communication and requires a careful approach to prevent the normalization of threats to democratic values.
To effectively cover anti-democratic candidates and threats to democracy, journalists must possess a deep understanding that is both quantitative and qualitative. This involves not merely counting instances of biased or inflammatory language, but also interpreting the broader impact of such rhetoric on public opinion and democratic health. This sophisticated approach allows the media to more accurately reflect the seriousness of the threats and the nuances of how they are presented to the public. In the case of Trump’s comments, this might mean pointing out how many people living in the U.S. might be deported and contextualizing this within current immigration-related deportations in the U.S. Coverage in this vein would also point out the ways in which such an action would violate not only international human rights treaties, but also laws and democratic norms within the U.S.
Normative research, which establishes clear standards for democratic journalism, is essential in our times faced by challenges to democratic norms. Without such standards, evaluating the effectiveness and integrity of journalism in the face of democratic threats becomes nearly impossible. By defining what responsible, democracy-preserving journalism should look like—especially when confronting anti-democratic rhetoric—researchers can and should provide a critical framework that guides and supports journalists in maintaining their role as defenders of democracy.
For example, our proposed framework for democracy-framed election coverage seeks to redefine journalistic practices to better protect democratic processes. In this framework, suggested by Heesoo Jang, Daniel Kreiss, Shannon C. McGregor, and Erik Peterson, we define democracy-framed electoral coverage as that which foregrounds fairly contested elections as both an established norm and a political ideal (Jang & Kreiss, 2024; Peterson, McGregor & Block, 2023). This frame of coverage goes beyond pointing out that claims of widespread voter fraud are false and not substantiated (if, indeed, there is no evidence that irregularities occurred) – it also positions election denial as a violation of democratic norms with deleterious implications for democracy. It treats election denial—or ex ante assertions that a candidate will not accept the result of an upcoming election—as fundamentally different from other campaign issues. It insists on a proactive role for journalists, not just as reporters of events but as active participants in preserving democratic integrity. By embedding democratic values at the core of election coverage, this framework aims to ensure that media outlets not only report on but also actively counteract anti-democratic narratives, thereby contributing to a more informed and resilient democratic society.
What makes democracy-framed journalism essential? It transcends the reporting of facts and figures to actively engage with the implications of those facts on democratic health. For instance, when candidates spout rhetoric that challenges the outcome of free and fair elections or threatens anti-democratic deportations, democracy-framed journalism doesn’t just quote these statements; it contextualizes them against the fabric of democratic standards, highlighting the risks they pose to the peaceful transfer of power, the integrity of electoral institutions, and the health of democratic institutions and the populace.
Campaign Communication
As our studies of electoral journalism needs a normative frame, so too does our study of campaign communication. In the past, we’ve urged scholars of campaigns to move beyond conceptualizing campaign communication in terms of issues or tone, to centering the role of political identity in campaign communication (Kreiss, Lawrence & McGregor, 2020). But even these expanded conceptualizations do little to lay out the norm-breaking of patently illiberal and anti-democratic campaign communication. Thankfully, work on populist campaign communication (e.g., Aalberg, Esser, Reinemann, Strömbäck, & de Vreese, 2017), and those with a particular emphasis on dehumanizing language (e.g., Hameleers, 2023), helps us to move closer to a normative democratic framework for analyzing such campaign communication. But we lack a cohesive framework to interrogate the democratic—and illiberal—ideals communicated by politicians.
The rise of anti-democratic candidates and politicians poses a challenge to researchers in similar ways that they pose a challenge to journalists, how do we best conceptualize communication that threatens democracy? Returning to the example of Trump’s promise, in his TIME interview, to deport millions of people living in the U.S., this is certainly communication about an issue (immigration), but that conceptualization, though important, fails to capture the anti-democratic nature of this statement. It’s also a populist appeal, rooted in white political identity, with a clear outgroup threat appeal. But none of these alone capture the democratic—and physical—threat represented by this communication.
As Danielle Brown argues in her contribution to Media and January 6th, it’s important that we use precise language to characterize the events of January 6th, 2021, at the U.S. Capitol. She argues that “by associating those events with protests, we threaten all legitimate protest activity” (Brown, 2024: 28). So too is the case with our conceptualizations of campaign communication: if we characterize Trump’s statements as issue-based or identity-based, we risk undermining legitimate and democratic issue-based or identity-based appeals.
Technology
Like journalists and researchers, social media platforms must also play a key role in placing democratic bounds around political communication, and speech about elections in particular. A U.S. working group recently issued a report calling on platforms to develop a framework to assess—and swiftly act on— “threats to the peaceful conduct of elections and the holding and transfer of power” (Eisenstat, Hendrix & Kreiss, 2024). These include developing and enforcing policies around threats to elections, being transparent about content moderation decisions in this vein, and ending exceptions to content moderation for high-value (read: high-power) users. In 2024, over 50 democracies around the world have had or will have elections. While social media platforms are by no means solely responsible for threats to elections, they are key conduits of communication from elites and between individuals and groups that seek to cast doubt on the integrity of elections, foment violent anti-democratic behavior, and prevent the peaceful transfer of power. One of the things we can do—beyond call on platforms to enact the type of policy recommendations from the working group—is to continue to develop and refine our conceptualizations of the role of technology in political communication within a normative democratic framework.
This also means that we should not, in our own research, assume that the online and digital participation we are so fond of measuring is pro-democratic. As Silvio Waisbord reminds us, “nothing about citizens expressing and organizing online necessarily leads to virtuous outcomes … Citizens may participate to contribute to the public good or to impose their will by force, to promote emancipation and critical reasoning, or to spread hate and violence” (2024: 87; see also Jackson & Kreiss, 2023). The same is true for politicians and other political figures active in campaigns. And we would do well to remember this when we consider the role of technology in campaigns and elections.
Failures in the aforementioned areas raise the specter of AI harms, though not necessarily in the same vein as commentators have wrung their hands about. Artificial intelligence, particularly in the form of large language models like ChatGPT, depends significantly on the data it is fed. Using multiple layers of neural network architecture based on transformers, this training data primarily comes from existing internet and media sources, which often lack a comprehensive democratic framework (Jang & Kreiss, 2024). Consequently, if the input data is skewed or lacks democratic underpinnings, the output from AI will mirror these deficiencies, unable to independently correct or recognize these biases due to its inherent reliance on the given data.
It is critical to focus not just on the integration of AI in electoral processes but on how AI models and their outputs align with or contradict democratic values. The lack of a standard framework for evaluating AI outputs against democratic principles means that these technologies could propagate undemocratic narratives unchecked, affecting the broader political discourse and potentially reinforcing harmful biases. The role of media in covering anti-democratic candidates provides a reflection of the broader challenges AI faces in political communication. AI technologies have the capacity to amplify certain political narratives while suppressing others, significantly influencing public perception and media coverage. A deep understanding of AI’s role and its impact on these processes is crucial for a critical analysis of how media reports on such candidates and their implications for democracy.
The press coverage of elections and campaigns serves as a primary source of training data for AI models like ChatGPT and Claude. This content shapes output. Putting aside the legal and ethical questions around tech companies’ use of news coverage as training data, the ways these AI models are currently trained makes them reflective of the dearth of democratic norms. If the press coverage is biased or fails to adequately address the complexities of democratic processes, the AI’s outputs will likely perpetuate these flaws. This underscores the need for our field—including those focused on campaigns and elections—to conduct audit studies of LLM models.
Several AI models, including ChatGPT, often employ a tactic known as “false balance” when generating responses to political queries (Vincent, 2023). This approach presents issues as being more evenly balanced between opposing viewpoints, even when one side clearly contradicts democratic principles. Integrating a democracy-framed election coverage framework could provide a systematic method to assess and guide AI outputs, benefiting researchers, journalists, and the public. Additionally, recognizing the potential harms perpetuated by AI models that prioritize Silicon Valley’s capitalist logics over democratic values is crucial. This recognition underscores the urgent need for a normative framework tailored to uphold and reinforce democratic standards in a new political landscape where we will see AI-generated content in both elections and campaigns.
Conclusion
Taken together, we argue that as we adopt new approaches to studying elections and campaigns, we must center normative democratic commitments. This essay underscores the necessity for a democracy-centered approach in studying elections and campaigns—how they are covered by the press, how they are conducted by campaigns, the role of technology and social media, and how we should assess AI technology governance. It is evident that without a clear normative framework that centers democratic principles, the press, campaign communication, and technology can inadvertently or deliberately influence electoral outcomes and public perceptions in ways that may not align with the ideals of a fair and transparent democratic process. A continuous and rigorous assessment of how journalism, campaigns, and technology serve or hinder democratic health is crucial.
References
Aalberg, T., Esser, F., Reinemann, C., Strömbäck, J., & Vreese, C. H. (Eds.). (2017). Populist political communication in Europe.
Blanco, A. Wolfe, D., & Gardner, A. (2022). Tracking Which 2020 Election Deniers Are Winning, Losing in the Midterms. The Washington Post. https://www.washingtonpost.com/politics/interactive/2022/election-deniers-midterms/
Brown, D. K. (2024). What January 6th Was Not. In Media and January 6th. Eds: Costley White, K., Kreiss, D., McGregor, S.C., & Tromble, R. Oxford University Press.
Eisenstat, Y., Hendrix, J. & Kreiss, D. (2024). Preventing Tech-fueled Violence: What online platforms can do to ensure they do not contribute to election-related violence. The Bulletin of Technology & Public Life. https://doi.org/10.21428/bfcb0bff.eeb0a669
Hameleers, M. (2023). Debasing Language Expressed by Two Radical Right-Wing Populist Leaders in the Netherlands: Geert Wilders and Thierry Baudet. In Debasing Political Rhetoric: Dissing Opponents, Journalists, and Minorities in Populist Leadership Communication (pp. 55-71). Singapore: Springer Nature Singapore.
Hellmeier, S. et al. (2021). State of the World 2020: Autocratization Turns Viral. Democratization 28(6): 1053–74.
Jackson, S. J., & Kreiss, D. (2023). Recentering power: conceptualizing counterpublics and defensive publics. Communication Theory, 33(2-3), 102-111.
Jang, H., & Kreiss, D. (2024). Safeguarding the Peaceful Transfer of Power: Pro-Democracy Electoral Frames and Journalist Coverage of Election Deniers During the 2022 US Midterm Elections. The International Journal of Press/Politics.
Kreiss, D., Lawrence, R. G., & McGregor, S. C. (2020). Political Identity Ownership: Symbolic Contests to Represent Members of the Public. Social Media+ Society, 6(2).
Peterson, E., McGregor, S.C., & Block, R. (2023). Election Denial as a News Coverage Dilemma: A Survey Experiment with Local Journalists. (Presented to the Political Communication Section, annual meeting of the American Political Science Association, Los Angeles, CA).
Vincent, J. (February 17, 2023). As conservatives criticize ‘woke AI,’ here are ChatGPT’s rules for answering culture war queries. The Verge. Retrieved from: https://www.theverge.com/2023/2/17/23603906/openai-chatgpt-woke-criticism-culture-war-rules
Waisbord, S. (2024). Antidemocratic Publics: The January 6th Mob and Digital Organizing. In Media and January 6th. Eds: Costley White, K., Kreiss, D., McGregor, S.C., & Tromble, R. Oxford University Press.
Heesoo Jang is a PhD candidate in the Hussman School of Journalism and Media, University of North Carolina at Chapel Hill, where she is also a graduate affiliate with the Center for information, Technology, and Public Life. Starting September 2024, she will be an assistant professor in the Department of Journalism at the University of Massachusetts, Amherst.
Shannon C. McGregor (PhD, University of Texas – Austin) is an associate professor in the Hussman School of Journalism and Media and a principal investigator at the Center for Information, Technology, and Public Life – both at the University of North Carolina, Chapel Hill.
[1] Copyright © 2024 (Heesoo Jang & Shannon C. McGregor). Licensed under the Creative Commons Attribution Non-commercial No Derivatives (by-nc-nd). Available at https://politicalcommunication.org.