And That’s a Fact: A Rhetorical Perspective on the Role of Fact-Checkers

ABSTRACT

Fact-checking initiatives have rapidly expanded worldwide with the overall aim of assessing the veracity of factual claims made in public. In this paper, we critique how the objectivity norm informs the current role perception of fact-checkers and suggest ways to reinterpret this perception through rhetorical argumentation theory. First, we point to research literature on fact-checking as a site in which role perceptions of fact-checkers and epistemological ideals are represented, (re)created and contested, marking out an objectivist role perception. Second, we highlight recent contributions from scholars who, like us, have noticed tensions between the objectivist role perception and role performances of fact-checkers and constructively recommended different ways of working around these tensions. In dialogue with these suggestions, we propose that rhetorical argumentation theory can be considered a supplementary resource to further advance epistemological discussions on the role perception of fact-checkers. In addition, the vocabulary from rhetorical argumentation theory holds potential for methodological operationalisation to guide the future role performances of fact-checkers. Finally, to develop our suggestion further, we propose a new research agenda embracing action-oriented research as an integrated way of developing fact-checkers’ role and practise together with practitioners.

KEYWORDS:

• Fact-checking
• Epistemic practice
• Epistemological beliefs
• Role performance
• Role perception
• Rhetorical argumentation theory
• Factual claims

Introduction

In the last decade, fact-checking initiatives have rapidly expanded worldwide (Graves 2016; Graves and Cherubini 2016; Lowrey 2017), often along with commendatory remarks from researchers describing the fact-checking practice as “a professionalisation and promotion of journalism” (Eide 2012, 176), “a new democratic institution” (Graves and Cherubini 2016, 1) and “a democracy-building tool” (Amazeen 2020, 99). As of 2023, the Duke Reporters’ Lab counts 400+ active fact-checking organisations globally (Stencel, Ryan, and Luther 2023). In general, fact-checkers are envisioned to assess the veracity of factual claims made in public, correct falsehoods and provide the public with accurate information as an effective counterweight to deception and misinformation (Lowrey 2017; Nieminen and Rapeli 2019; Vinhas and Bastos 2022; Walter et al. 2020). However, in the wake of the laudatory notes, a rising number of critical reflections concerning fact-checkers’ epistemological beliefs and practices have followed (Cloud 2018; Plug and Wagemans 2020; Uscinski 2015; Uscinski and Butler 2013). Moreover, in fact-checking newsrooms, fact-checkers concurrently work to build trust in their role (Graves 2018, 624) and defend their authority (e.g., by offering detailed descriptions of their methodology and obtaining verified membership of the International Fact-Checking Network and the European Fact-Checking Standards Network Project).

While for some, the consequence of their epistemological critique is the complete dismissal of fact-checkers (Uscinski 2015; Uscinski and Butler 2013), this is not the case for us. We believe that the examination of factual matters and how facts are established, made sense of and contested is an essential element when facilitating public debate, and thus, a subject matter that calls for scrutiny from dedicated agents. However, at the same time, we question the epistemological foundations and role perceptions of the current fact-checking practice. What initially caught our attention was the evident discrepancy between the declared mission and actual practice, when a verdict is positioned somewhere in between the definitive “true” and “false”. Obviously, fact-checkers navigate difficult contexts characterised by genuine disagreement over complex issues, lack of knowledge and factual claims of very different kinds with support from data of different types and with varying degrees of certainty (if ever any!). Furthermore, looking towards authorities for definitive answers, fact-checkers experience that just as the facts are not as stable as one could wish for, authority is ascribed and constituted socially and is, therefore, potentially fluctuating (Valaskivi and Robertson 2022). Strictly separating facts and opinions can also be immediately appealing; however, this dichotomy does not account for the interconnectedness between factual claims and perspectives, values, and policy claims (Jasinski 2001a). Lastly, the recent development within the fact-checking practice, where the object of examination has turned from claims by elite actors in public debate (e.g., politicians, pundits, journalists) to citizens on social media (Graves and Mantzarlis 2020), adds a layer of asymmetry between the fact-checker and the fact-checked.

Illustrating some of these complexities, the topical question of whether integration efforts are successful in Denmark has spawned a cascade of fact-checks with verdicts for and against, scrutinising claims on both sides. At one point, the Danish fact-checking initiative Detektor criticised a communal chief constable claiming headway with integration in Vollsmose (Detektor DR 2018a) (a residential area in Denmark that was part of the government’s then-named “ghetto list”) fulfilling certain statistical criteria concerning employment, crime, income, and education rates (Regeringen 2017). The communal police force published the claim of headway in a self-evaluation report and supported it with statistics showcasing that eight out of nine areas in Vollsmose experienced a lower crime rate than the larger neighbouring city of Odense. Fact-checkers criticised this report for distortion and cherry-picking and countered the claim with reference to other experts stating the opposite. Two months later, the same fact-checkers confronted the news director of a Danish broadcasting station following a news segment claiming failed integration on the basis of an interview with 11 Somali women from Vollsmose who apparently did not respond when asked about their Danish language skills, even though they had lived in Denmark for several years (Detektor DR 2018b). Again, the fact-checkers found fault in the argument, questioning the reporter’s interview techniques and generalisation. In the end, as a recipient of these fact-checks, one is still left with the question of whether integration efforts are successful. Which is the better claim?

In an effort to attend to and reflect upon this difficult task taken up by fact-checkers, we, first, point to research literature on fact-checking as a site in which role perceptions of fact-checkers and epistemological ideals are represented, (re)created and contested. Here, the concept of professional roles provides a vocabulary and analytical lens (Hanitzsch 2017; Mellado, Hellmueller, and Donsbach 2016). In a corpus of research literature on fact-checking, including various kinds of fact-checking, not solely journalist fact-checking in public service media as in the example above, we mark out an “objectivist” role perception of fact-checkers. This role perception, we argue, is based on an epistemological understanding of “facts” as objective, stable and always already available and subscribes to ideals of objectivity, empirical justification, and consistency to address and create an informed citizenry. In practice, this idea often leaves fact-checkers struggling in a world more complex, unstable, and fluctuating (Graves 2016). Second, we discuss some of the recent attempts from other scholars who through the lens of the sociology of science and functional hermeneutics have criticised the objectivist role perception and noted a discrepancy between such ideals and actual practices. In continuation of these, we suggest turning to rhetorical argumentation theory as a supplementary resource that compliments the epistemological beliefs in these more recent accounts and offers an additional conceptual apparatus with potential for operationalisation of the methodology. Finally, we outline a research agenda for such further development and qualification of the role performances of fact-checkers. We suggest fieldwork and action-oriented research as means to further understand and develop the fact-checkers’ role (Maruyama and Van Boekel 2014; Wagemans and Witschge 2019). Such reflective development practice will build on studies and discussions of the role performances and hopefully be a way to co-develop an operational understanding guided more firmly by a rhetorical argumentation perspective.

Theory: Professional Roles Between Ideals and Practices

In journalism scholarship, the study of professional roles is a key concept to examine how journalistic ideals, culture, identity, and practices are (re)created, (re)interpreted, appropriated and contested (Hanitzsch 2017; Mellado, Hellmueller, and Donsbach 2016). We posit that the framework, while rooted in journalism research, offers a useful lens to inform our analysis of the ideals and practices examined and created in the research literature on fact-checking, covering practices identifying both as journalistic and non-journalistic. For this purpose, we follow the terminology by Mellado, Hellmueller, and Donsbach of role conception, role perception, role enactment and role performance (Mellado, Hellmueller, and Donsbach 2016, 7). This perspective and the terminology conceptualise roles as situational and negotiated (Mellado, Hellmueller, and Donsbach 2016, 5). At the evaluative level, role conceptions refer to the individual and role perceptions to the societal constructions of the values, attitudes, and beliefs regarding the position of journalism in society. At the performative level, role enactment and role performance denote respectively the individual implementation of a journalistic role and the collective outcome of concrete newsroom decisions (Mellado 2019). This terminology allows for questions about the connection, dynamics, and tensions between an evaluative level of ideals and the performative level and practice of producing news (Mellado 2015, 2019; Mellado, Hellmueller, and Donsbach 2016).

Such tensions are especially visible in the discussion of epistemology and journalism, Örnebring (2016) argues in his review of research on epistemologies and professional roles—covering, on the one hand, epistemological beliefs connected with journalistic ideals and, on the other, the everyday journalistic practice dealing with knowledge claims. It is exactly across these axes that we seek to study fact-checking, as the very core of the fact-checkers’ role and practice is to assess knowledge claims made by others. Thus, fact-checking promises to be a productive site for continued insights into epistemological beliefs and epistemic practices. As part of this conversation, we take inspiration from some of the studies Örnebring summarises. Exemplifying scholarship interested in epistemological beliefs, Hanitzsch (2017, 375 in Örnebring 2016) classifies epistemology along two primary axes in his operationalisation of “journalistic culture”: objectivism-subjectivism (what types of knowledge are believed to be possible) and empirical-analytical (what justification for truth claims are deemed acceptable). From the interest in epistemic practices, Godler and Reich (2013 in Örnebring 2016), through a multimethod study, analyse how journalists determine what could be reported as a “fact”, and Shapiro et al. (2013 in Örnebring 2016) study through reconstruction interviews with journalists how verification is enacted in practice.

Material and Methods: A Reading of the Research Literature on Fact-checking

Our object of study is the body of research literature that examines ex post fact-checking; a particular practice and genre where the fact-checker assesses the veracity of claims made by others, after the claim appears in public. Thus, different from the internal practice of ex ante fact-checking, where journalists review their own work prior to publication (Graves and Amazeen 2019). As of June 2023, the Duke Reporters’ Lab counts 417 active fact-checking organisations globally (Stencel, Ryan, and Luther 2023). Looking across these initiatives, a wide diversity of organisational structures and professional backgrounds is found. Some initiatives are established within or in partnership with legacy media, others follow a NGO model, and its practitioners identify with professional titles such as journalist, activist and domain expert (Graves and Cherubini 2016).

Following the global rise of fact-checking initiatives, research on the practice has similarly grown spanning several disciplines (Nieminen and Rapeli 2019). In this article, we point to this research literature as a site in which role perceptions of fact-checkers and epistemological ideals are represented, (re)created and contested. Across the literature, we find an ongoing scholarly discussion circulating around the ideals and practices of fact-checking. Studies cover topics such as the effects of fact-checking, fact-checking as a practice and profession, and public opinion and perception of fact-checking (Nieminen and Rapeli 2019). Accordingly, some of the studies themselves examine - using the terminology of professional roles - role perceptions through surveys and interviews while other study role performances through content analysis of fact-checks and fieldwork-based analysis of fact-checkers’ ways of working. As we look across this body of work, we observe a site of discursive constructions of the values, attitudes and beliefs regarding the role of the fact-checker in society. Following from the framework of professional roles, we expect that these societal expectations may congrue with and influence practitioners; while at the same time, potentially exist in discrepancy with the fact-checkers’ own ideals and performances as they (re)interpret and enact their practice.

In this paper, our starting point and object of study is the research literature. We borrow an overview of the research literature on fact-checking from three literature review articles concerned with this field of study—namely, Nieminen and Rapeli (2019) reviewing studies on “political fact-checking” (296), Walter et al. (2020) reviewing studies concerned with the “effectiveness of fact-checking in correcting political misinformation” (350), and Vinhas and Bastos (2022) reviewing studies on “fact-checking and the evidence that it can correct prior knowledge and successfully debunk misinformation” (448). Our corpus in the following analysis thus includes studies covered in one or more of these three review articles. As the fact-checking practice is currently evolving in scope (Graves and Mantzarlis 2020; Graves, Bélair-Gagnon, and Larsen 2023), the choice to frame our corpus through these literature reviews excludes the most recent research published after 2022. However, we contend that despite continued development, the research and the practice itself continue to be informed by the foundational perceptions depicted in the chosen corpus.

Analysis: An Objectivist Role Perception Within Research on Fact-checking and Examples of Reinterpretations

While we recognise the existence of various role perceptions in our corpus, we mark out the significant presence of an objectivist role in a substantial portion of the research literature. This perception of fact-checkers’ role is certainly exhibited on a spectrum and to varying degrees, generally, however, we find it to be characterised by first, the overall goal of creating an informed citizenry; second, an understanding of “facts” as verifiable, objective and always already available; and third, reliance upon a consistent methodology as the means to attain accurate and impartial knowledge.¹

An Ideal of the Informed Citizenry

Throughout the research literature, fact-checking is suggested as a remedy for the public debate suffering in “today’s post-truth information setting” (Hameleers 2022, 122) in which “information disorders” (Lazer et al. 2018, 1094) and “misinformation” spread fast (Burel et al. 2020, 28). Fact-checkers are envisioned to be a possible “antidote” (Walter et al. 2020, 351) and “fighters” (Nieminen and Rapeli 2019, 296; Vargo, Guo, and Amazeen 2018, 2043) who can “combat” misinformation by debunking falsehoods (Humprecht 2020, 310) and provide the broad democratic public with “accurate information” (Walter et al. 2020, 352). Starring the concept of misinformation in the role of either disease or enemy places fact-checkers in a committed civic role from which they are proposed to save the public debate. Their audience is described as “citizens” who they “help” by offering accurate information (Amazeen 2013, 10). In some research, fact-checkers are furthermore positioned as watchdogs holding public figures accountable for their deceptive claims (Agadjanian et al. 2019; Amazeen 2015a; Graves and Glaisyer 2012, 3).

Following from this overarching goal, a significant part of the research field is concerned with the effects of fact-checking—in other words, whether fact-checking can correct “false beliefs” and thereby oppose the spread of misinformation. Various empirical studies offer mixed results (Walter et al. 2020)—ranging from evidence of positive and confirmative effects (Fridkin, Kenney, and Wintersieck 2015; Gottfried et al. 2013; Porter and Wood 2021; Weeks 2015) through poor effects and unsupportive evidence (Garrett, Nisbet, and Lynch 2013; Jarman 2016; Lewandowsky et al. 2012; Nyhan and Reifler 2010; Thorson 2016) to examples of “backfire effects” (Nyhan and Reifler 2010), which then again have been doubted and found anomalous (Garrett, Nisbet, and Lynch 2013; Nyhan 2020; Wood and Porter 2019). Despite the ambiguity between the individual studies, the main objective ascribed to the practice is to offset and correct “false beliefs” and thereby create a more informed citizenry, as exemplified by Garrett, Nisbet, and Lynch (2013, 617): “Yet for democracy to succeed, a society must have the means to correct political misperceptions among its citizens”.

In combination, the characterisation of the fact-checking mission firstly mirrors an “information ideal of citizenship” found in models of journalism and deliberative theories of democracy (Ananny 2020, 8; Jarman 2014) that aims to create a citizenry correctly informed of factual matters enabling them to rationally debate and decide on policy alternatives. Secondly, the fact-checkers are positioned in a role suggested to cater to and realise this information ideal.

Epistemological Beliefs About (Verifiable) Facts and the Object of Fact-checking

Central to the role perception of fact-checkers are the epistemological beliefs and assumptions about facts, truth, and knowledge. In the research literature on fact-checking, such assumptions appear in contestations about which claims are and are not possible to fact-check, a discussion that has flowed back and forth among practitioners and scholars (Amazeen 2015b; Graves 2016; Uscinski 2015; Uscinski and Butler 2013).

Political scientists Uscinski and Butler (2013, 162) accuse fact-checkers of subscribing to a “naïve political epistemology” through their use of simplistic true/false judgements (Uscinski 2015, 247), which Uscinski and Butler deem incompatible with the complex and ambiguous subject matter of politics. While they disqualify the practice within the political sphere (Uscinski and Butler 2013, 178), Uscinski offers fact-checkers the category of “verifiable statements” for which the methodology is considered to be appropriate (Uscinski 2015, 249). The distinction between “verifiable facts” and non-verifiable statements—including, but not limited to, those of opinion—also exists in studies less critical of the practice. According to Amazeen (2015b, 4, 2016, 436), the practice is concerned only with statements that can be “verified”, also described in an earlier paper as “knowable facts” (2013, 26), Lim argues that fact-checkers perform better (understood as achieving higher levels of agreement across fact-checking organisations) when evaluating “outright falsehoods and obvious truths” (2018, 1), Hameleers defines misinformation as “objectively incorrect and empirically falsifiable” (2022, 111), and Walter and Salovich describe how fact-checkers exclusively focus on “fact-based statements—something that is capable of being proved or disproved by objective evidence” (2021, 503).

These distinctions and qualifiers suggest a divide between supposedly unproblematic, objective facts and an ambiguous group of political facts and statements. A demarcation similar to the distinction between fact and value judgement that have been called into question by journalism scholars and philosophers of science (Graves 2017; Yarrow 2021). The first category of “verifiable facts” is depicted as a type of knowledge obtained and justified through empirical observations and evidence, and it’s only within this former group, the fact-checker is considered epistemologically equipped to account for and adjudicate statements, thus performing a narrow objectivist role.

Reliance upon Consistent Epistemic Practices

Lastly, we now consider how the research literature represents and evaluates the everyday practices of fact-checkers as they deal with factual claims. Here, we find examples of scholars attending to the production process and routines and others to the fact-checks as outcome. Some have done so through fieldwork (Graves 2016, 2017, 2018; Graves and Konieczna 2015; Haigh, Haigh, and Kozak 2018), which we will revisit in the subsequent section addressing interpretations of the objectivist position. Others have examined fact-checks and policy documents through (qualitative and quantitative) content analysis. While these studies can be described as attending to the role performances of fact-checkers, we also view them as part of the discursive construction of the role perception.

The research literature describes and recommends various epistemic practices as important aspects for fact-checkers to gain credibility, for example “source transparency” (Humprecht 2020) and the consultation of experts (Amazeen 2013, 26). Furthermore, several scholars are concerned with the level of consistency across fact-checking organisations, and therefore, compare published verdicts as well as policy documents (Amazeen 2015b; 2016; Lim 2018; Marietta, Barker, and Bowser 2015). As with the studies examining the effects, the results are mixed. Where Amazeen (2015b, 2016) celebrates cross-organisational consistency, Lim (2018) and Marietta, Barker, and Bowser (2015) disagree and highlight methodological failings within the practice. Nevertheless, all three studies perceive consistent methodology to be a qualification that the fact-checkers should at least strive for in order to realise their overarching goal of opposing misinformation (Amazeen 2016, 433; Lim 2018, 7; Marietta, Barker, and Bowser 2015, 579). Together with the focus on source transparency, the ideal of consistency points towards an expectation of fact-checkers to work in standardised ways and “examine the same facts and […] reach the same conclusions about their truth or falsehood” (Marietta, Barker, and Bowser 2015, 582). Following such epistemic practice, the fact-checkers could—following from the objectivist role perception—better guard themselves against accusations of bias and subjectivity as well as offer more reliable and qualified information to the public.

Reinterpretations of Role Perceptions based on Different Theoretical Resources and Observation of Role Performances

Some scholars have in similar ways critiqued the objectivist role perception of the fact-checker and point to a discrepancy between its inherent ideals and the actual practice, more specifically, Lucas Graves (2016, 2017) and Steen Steensen et al. (2022). Involving theoretical perspectives from social sciences and the humanities, the authors propose alternative epistemological beliefs that might better comply with the role performances and furthermore can serve as a point of departure for future development of the fact-checkers’ role and practice. “The […] goal is to outline a more useful approach to the much-debated epistemology of fact checking by drawing comparisons with studies of scientific work”, Graves states (2017, 519). Similarly, Steensen et al. advance an alternative to the “standard ways of verifying information in journalism, which [they] argue are limited by a binary epistemology in which dichotomies like true/false, reliable/unreliable, and accurate/inaccurate structure the approach” (2022, 2121).

Practical Epistemology, Factual Coherence, and Institutional Facts

To discuss the epistemological beliefs about facts that dominate the role perception of fact-checkers, Graves turns to the sociology of science and Bruno Latour’s conception of the collective nature of fact-making in science: “Scientific facts are deeply collective, argues Latour (2005), made durable by networks of ideas and people and equipment” (Graves 2017, 522). Inspired by Latour’s investigation of laboratory life (Latour and Woolgar 1986), Graves becomes involved as a participant observer in the factual inquiry as it is practised in three US fact-checking newsrooms, understanding fact-checking similarly as a form of knowledge production that exists in material, social, and discursive contexts (Graves 2016, 69–70; 2017, 520, 530). Thus, Graves studies the fact-checkers’ role performances from which he offers “an alternative view highlighting the practical epistemology revealed in the newswork routines and discourse of working fact checkers” (Graves 2017, 518 our italicisation). Fact-checkers navigate a complicated factual terrain where experts disagree and evidence is often incomplete or ambiguous. Here, Graves shows how fact-checkers rather build “factual coherence” in a network of sources (2017, 532), whereas the “correspondence” theories of truth where “a fact is something that is indisputably the case, a true statement one that corresponds to the facts of reality” (Graves 2017, 520) fall short. Inspiration is here drawn from Ettema and Glasser (1998) who use the concept of “factual coherence” describing investigative journalists who often in a similar manner deal with “stories in which official facts and the institutional routines that certify them are in doubt” (Graves 2017, 532).

Supplementary to the sociology of science, Graves situates fact-checkers in an information context characterised by what John Searle has called “institutional facts” (Graves 2016, 70–72, 163–164), defined as “facts produced within the agreed-upon institutional reality human inhabits” (2016, 164). Such definition highlights that facts are more than “brute facts” of the material world, they are also often descriptions of a reality based on collective constructions and categories in a social, political, and economic world; for example, money, borders, and joblessness rates “exist and have meaning only by virtue of some institutional rule-making apparatus” (2016, 71). In moments of institutional unsettlement or instability (Graves 2016, 55, 70–73, 164; 2017, 532), Graves recommends fact-checkers to use their trained capacity with source criticism, especially expert sources, and in the process of establishing factual coherence “tease out useful facts from ideological arguments” (2016, 76) or “distil decisive facts from political rhetoric and ‘spin’” (2016, 116).

Functional Hermeneutics and Source Criticism

Along the same lines, the Source Criticism and Mediated Disinformation project (SCAM), led by Steen Steensen and with the participation of Graves, advances the theory of functional hermeneutics and the concept of source criticism as “an alternative to other practices and understandings of information verification in journalism” (Steensen et al. 2022, 2119). The scope and recommendations are not intended for fact-checkers specifically but for all journalists working with truth-claims and representations of reality. As Graves, Steensen et al. address the dominating objectivist role perception and, more specifically, “verification” as a doubtful governing norm and the “binary epistemology” as an unsuitable structuring principle in journalism (2022, 2122–2123). They argue that the objectivist role perception pushes journalists to interpret information as either true or false and thereby neglect the complexity in the processes of producing and communicating truth-claims, and, moreover, tends to “not adequately acknowledging that degrees of accuracy and reliability depend on socio-cultural context and interpretation” (2022, 2123).

As a replacement for the dominating verification and debunking practices in journalism, the authors urge journalists to employ their trained capacity with source criticism as an alternative epistemology and methodology. According to the authors, “the benefits of source criticism are that it not only is based on a (…) middle-ground epistemology between naive realism and relativism, it is also a distinct methodology of knowledge production” (2022, 2124). As an epistemology, source criticism foregrounds the “tendencies” inherent in all sources and source material from which it follows that all sources and source material must be interpreted and reinterpreted, also in relation to other sources, source material and contexts (2022, 2126). Operationalising source criticism as a methodology, the authors suggest three norms to be applied by journalists in their practice: (1) harness truth-claims with modesty; (2) deploy interpretive transparency; and (3) operationalise self-reflective truth-claims (2022, 2119, 2121, 2128–2133). These norms foreground the carefulness that everyone must approach truth-claims with, others’ and one’s own.

To sum up, common for Graves and Steensen et al. is a radical dissociation from the logical, deductive tradition of reasoning. As Graves states: “In practice, fact-checking rarely proceeds in neat Cartesian steps to a set of hard, elemental facts that either support a statement or don’t” (2016, 69). In a similar vein, Steensen et al. set forth that “the concept of verification […] conforms less well to systems of knowledge production in the sociocultural domain to which journalism belongs” (2022, 2122). Furthermore, they all show a willingness to not only critique but also reflect on the role perceptions and role performances of fact-checkers (and journalists), offering constructive guidance as they transform epistemological reflections into operational methodologies.

Discussion: A Rhetorical Perspective on the Role of Fact-checkers

While Graves compares concepts within fact-checking and the sociology of science, and Steensen et al. connect journalism studies and functional hermeneutics, we propose rhetorical argumentation theory to continue the conversation about the role of fact-checkers. In the section that follows, we argue that rhetorical argumentation theory offers benefits in two ways: First, rhetorical argumentation theory complements the development of an alternative epistemology of fact-checking in continuation of the initiatives by Graves and Steensen et al., and, second, rhetorical argumentation theory offers a nuanced conceptual apparatus, useful for fact-checkers in their everyday work, and thereby assisting the development of the fact-checkers’ methodology.

Noticeably, without adhering explicitly to any theories of argumentation, both Graves (2016, 2017) and Steensen et al. (2022) use a vocabulary of argumentation when describing the fact-checking role and practice. Graves accentuates that facts are constituted through argumentation noting how fact-checkers are “weighing claims” (2016, 114) and assessing the merits of sets of “arguments” (2016, 116; 2017, 531) in matters of “factual reasoning” (2017, 531). Likewise, Steensen et al. refer to “claims” and “truth-claims” as well as the acts of “supporting”, “producing” and “assessing” such claims (2022, 2120, 2124, 2128). To us, this indicates that a theory of practical argumentation could be productive to further negotiate role perceptions, examine role performances and guide the epistemic practice of fact-checkers. The conceptual framework in rhetorical argumentation theory is born out of an interest in practical, societal discourses. Therefore, it is well-suited to both inform the analysis and evaluation of matters of factual reasoning in practice as well as the production of factual arguments by the fact-checkers themselves.

Epistemology: Further Development from Rhetorical Argumentation Theory

The epistemology in rhetorical argumentation theory complements the contributions by Graves and Steensen et al.. In the same way that Graves finds correspondence theories of truth insufficient (2017, 520, 532), and Steensen et al. reject verification because it “carries with it some discursive baggage from its origin in logical positivism and the later development of the hypothetico-deductive method” (2022, 2122), rhetoric has since Aristotle contrasted itself to dialectics and formal logic (analytics) (Aristotle & Kennedy 2007, 1354a; Jørgensen 2013; Perelman 1982, 1–8). A rhetorical field of study would not claim that there are no facts outside rhetorical creation but that knowing and making sense of facts is inevitably a process of rhetorical mediation and argumentation in specific material, social and discursive situations in which the choice of words, the data and the framing influence understandings (for a summary of the rhetorical study of science, see Cloud 2018, 25).

Stephen Toulmin and Chaïm Perelman are among the main figures resuming the confrontation with absolutism in rhetorical theory in the twentieth century and are considered founders of the so-called new rhetoric. Pointing to the limited practical value of formal reasoning, Toulmin (1958) introduced a theory of practical argumentation and key terms such as “argument fields” and “field-dependency”, demonstrating how truths and the establishment of factual claims are highly contextual. Presenting an argument model consisting of six interrelated components (“claim”, “ground”, “warrant” as the three fixed components, and “backing”, “rebuttal”, and “qualifier” as the facultative items), Toulmin highlighted the justification function of practical argument (as opposed to the inferential function of theoretical arguments). According to Toulmin, practical argumentation often begins with a claim followed by a justification, not the other way around as presumed in formal logic where sound premises deductively lead to a sound conclusion. Toulmin, who initially did not receive much recognition for his work in philosophical circles, received a warm welcome in the field of rhetoric where Brockriede and Ehringer introduced his concepts and ensured the widespread of his ideas (Brockriede and Ehninger 1960; Ehninger and Brockriede 1963).

Around the same time as Toulmin and with a similar mission of establishing a contrasting position to formal logic and the understanding of reasoning as “demonstration”, Perelman presented a “regressive philosophy”. With inspiration from Aristotle, he argues that aspects of being are a construction according to probabilities which are in principle always open to change (Frank and Bolduc 2003, 191). In the absence of an authority that can determine absolute truths across time and place, Perelman argued that potential doubt and the perpetual possibility that things can change with new insights and more knowledge is a basic condition: “We see that this status of fact and truth is not guaranteed indefinitely unless we accept the existence of an infallible authority, a deity whose revelations are incontestable and who would guarantee these facts and truths. However, if we lack such an absolute guarantee, such self-evidence, and such necessity as would compel every reasonable being, the facts and truths which are accepted by common opinion or by the opinion of specialists become open to question” (1982, 24). Thus, in principle, all facts can be questioned, and uncertainty and potential disagreement are conditions under which facts are established in the world, even facts about the material world. To emphasise this, a common distinction in rhetorical argumentation theory is between “factual statements” and “factual claims”, where “a factual statement becomes a claim as soon as someone expresses doubt or disagreement” (Jasinski 2001a, 27). Perelman’s reflections about the lack of an absolute guarantee are similar to Graves’ notion about institutional facts and the current instability of authorities and how this affects the knowledge economy. Furthermore, the distinction between “factual statements” and “factual claims” details the observation by Graves of how “of course, every fact is also always an argument, open at least in principle to qualification or refutation. This applies no less in hard science than in political life” (2016, 72).

Rhetoric origins from an interest in practical, public discourse, and rhetorical concepts are therefore in their essence directed at studying and theorising contexts of public discourse and deliberation. While an interest in scientific argumentation is also a part of the contemporary field of rhetoric (“rhetoric of science”) (Ceccarelli 2020; Harris 2017), it is generally not understood as fundamentally different from practical argumentation. Various rhetorical scholars have demonstrated how classical rhetorical concepts are useful when analysing scientific texts and understanding their “effect” - making the point that science is rhetoric (on rhetorical figures as an essential part of the argumentation in science see Fahnestock 1999; on “big rhetoric” see McCloskey 1985). Again, the rhetorical approach seems well aligned with Graves who emphasises that fact-checkers’ practical knowledge production shares basic traits with other epistemic workers situated knowledge production, e.g., investigative journalists or even scientists: “Any useful view of the epistemology of fact checking should understand it as a practical truth-seeking endeavor not fundamentally different from other kinds of factual inquiry” (Graves 2017, 523).

Methodology: On Weighing Practical Arguments

Next to the epistemological reflections, rhetorical argumentation theory holds potential to be operationalised as a methodology for fact-checkers as it includes a fine-grained vocabulary that can not only be used to describe but also evaluate practical, public debate about factual matters. As we now list some of the central concepts, we do not explain them comprehensively but aim to convey the breadth and richness in an introductory manner to hopefully convince researchers and fact-checkers about the usefulness of these concepts. In addition, our future research agenda involves an action-oriented process in which we plan to operationalise the concepts to accommodate the fact-checking practice in collaboration with fact-checkers.

One group of concepts found in rhetorical argumentation theory is the differentiating between four kinds of claims: “designative”, “definitive”, “evaluative”, and “advocative” claims (Brockriede and Ehninger 1960, 52–53; Jasinski 2001a, 27–30). While fact-checkers primarily attend to the first two, an awareness of the distinctions and the interrelatedness that oftentimes make the claims difficult to separate in practice, can help fact-checkers navigate the difficult terrain of practical, public debate (Bengtsson and Villadsen 2024). Another group of concepts are argument typologies, including the widespread and generally accepted typology of warrants containing arguments from “cause”, “sign”, “generalisation”, “example”, “parallel case”, “classification”, “authority”, and “motivation” (Brockriede and Ehninger 1960, 48–51; Jasinski 2001a, 30–41). In factual inquiry, all these argument types can be applied as tools to analyse and evaluate the various attempts to justify designative claims (except for the motivational arguments, which are only relevant in relation to evaluative and advocative claims). Furthermore, the concepts used to describe rhetorical fallacies defined broadly as “defective arguments” (which are not the same as logical fallacies) can be relevant for fact-checkers. This vocabulary has been developed to qualify the discussions of norms and potentially poor or directly unethical argumentative behaviour, distinguishing, e.g., between “fallacies of reasoning” (e.g., “hasty generalisation” or “false cause”), “fallacies of relevance” (e.g., “argument ad hominem” or “argument ad populum”) and “fallacies of clarity” (e.g., “equivocation”) (Jasinski 2001b). In addition, discussions around “authority fallacies” can hopefully further illuminate the tricky issue of reliable sources (see, e.g., Walton 1997). Thus, a rhetorical argumentation framework does not only advance an alternative epistemology, but also a practical, operational methodology and vocabulary that can qualify the examination of factual arguments. So, rather than calling for “verification” of facts, labelling them as either “true” or “false”, concepts from rhetorical argumentation theory can serve as an alternative working vocabulary that more accurately supports fact-checkers in what they are actually doing; analysing and evaluating factual claims and the justification for these claims.

Approaching fact-checking in this way from rhetorical argumentation theory, the fact-checkers’ practice could more accurately be perceived and described as a process of weighing, in other words analysing the factual arguments and evaluating their weight (which arguments are considered most weighty, most credible, most authoritative, or most “coherent” to use Graves’ terminology). The weight metaphor is used in rhetorical argumentation to emphasise the basic idea that people argue “for the purpose of eliciting degrees of adherence or allegiance to ideas” (Perelman 1982, p. xii)—ideas of the world, ideas of how to act in it, and ideas of the values that should guide those actions. Similarly, Christian Kock has applied the weight metaphor to describe the processes of gaining adherence—to a claim and its warrant: “Warrants are not premises about the issue in question but assumptions we rely on about the kind and degree of argumentative weight we may assign to the grounds offered” (Kock 2007, 248). Perceiving the fact-checkers’ practice as a process of weighing highlights how the fact-checkers are arguers themselves. They are actively involved in a constitutive process building adherence to specific understandings of the world when they disseminate their work evaluating the quality of others’ arguments. Here, Toulmin’s concepts of “qualifier”, “rebuttal”, and “backing” (1958, 87–134) can serve as a productive reminder for fact-checkers themselves to construct full, nuanced arguments about factual matters. Using a qualifier, the arguer can indicate to what extent he or she commits to the claim (“presumably”). Using a “rebuttal”, the arguer can consider instances where it might not be the case (“unless”). Using a “backing”, the arguer offers concrete support to establish the more abstract warrant in instances where the warrant is not given in the context (“on account of”). By applying these “facultative” elements, the fact-checkers remain open to the insecurities and contextuality of their own arguments, which is a concrete way in which the fact-checkers discursively can accommodate the norm about “operationalise self-reflective truth-claims” advanced by Steensen et al. (2022, 2131–2133).

Finally, we envision that rhetorical argumentation theory can serve as a foundation for developing an alternative publication style that is more in line with the role performance of fact-checkers. The weight metaphor could, for example, serve to illustrate how arguments with some weight (albeit varying) exist in support of and against the claim in question. Thus, an analysis and evaluation must weigh these arguments against one another and situate the individual factual claim within a wider context and argumentative process. In our introductory example, Danish fact-checkers analysed two contradictory claims about integration efforts in Denmark and ended up criticising them both—and rightly so, as the support for those claims were in both instances problematic. However, the readers could have benefitted from the fact-checkers weighing the support for not only the individual claims of respectively “headway” and “failure”, but also the wider question of integration efforts in Denmark.

As we look towards the current practice of fact-checking, we find only very few fact-checkers who still adhere to a strict true-false dichotomy. More fact-checkers appear to have adjusted their vocabulary using in-between categories such as e.g., “half true”, “half false” etc. (see, e.g., PolitiFact, APP Factcheck, and Correctiv.Faktencheck). We consider this adaptation a sign that fact-checkers are moving towards a more rhetorical mindset, although, without leaving the objectivist vocabulary for good (rather now having a foot in both camps). Moreover, we also find a few examples of fact-checkers, who have abandoned the binary-inspired categories completely and instead disseminate their analysis and evaluation in prose, thereby allowing the audience to more actively take part in deciding which arguments they adhere to (see, e.g., Full Fact,TjekDet and Faktisk). Such practice is more in line with what we envision and could thus serve as examples of best practice for other fact-checkers. However, suggesting a publication style like this, we are aware of studies that conclude that in some cases “blunt wording” (like “false”) is more effective than “pallid” wording (Clayton et al. 2020; Jennings and Stroud 2023), and moreover, that such style implies having confidence in the readers’ capability of digesting the (sometimes) complex fact-check and not least making up their own mind. However, we find this preferable since it is more accurate in terms of how factual statements are constituted in social, discursive contexts as well as in better congruence with what the fact-checkers are actually doing.

As a way of reiterating our argument, we have in the Table 1contrasted our reading of the objectivist role perception as it is represented in a substantial portion of the research literature on fact-checking with our proposed alternative of the constitutive role perception grounded in rhetorical argumentation theory. As we have argued in this article, we consider the constitutive role perception to be a more accurate way of describing the fact-checkers’ practice, while also offering an alternative vocabulary to further qualify the analysis and evaluation of factual claims.

Table 1. Contrasting role perceptions of fact-checkers.

	Objectivist role perception (based on logic, formal reasoning)	Constitutive role perception (based on rhetorical argumentation theory)
Aim: What is the purpose of the fact-checking practice?	Creation of an informed citizenry and accountability within public debate through the correction of falsehoods	Deliberation about factual matters through scrutiny of arguments and evaluation of their weight
Practice: What are fact-checkers doing?	Evidence-based assessment and verification of facts following consistent methodologies	Analysis and evaluation of factual claims (and the support for these claims) in discursive, social, material contexts as part of building adherence to factual statements. In this, fact-checkers play an active part as arguers themselves when disseminating the fact-checks
Publication style: How do fact-checkers disseminate?	Rating claims as true or false	Weighing individual factual arguments situated in wider thematic contexts

Future Research Agenda

An academic discussion of how rhetorical argumentation theory can contribute to the development of the epistemology and methodology of fact-checking is only a first step. As part of a newly funded project taking a rhetorical argumentation perspective on fact-checking, we do action-oriented research (Maruyama and Van Boekel 2014; Wagemans and Witschge 2019), by going into the field—the newsroom of a Danish journalistic fact-checking organisation—to explore and develop the role performance further in collaboration with fact-checkers. Hereby, we follow the call from Örnebring (2016) and hope to add to the conversation about the dynamics between fact-checking ideals and the study of everyday fact-checking practices. Furthermore, the project explores a comparative perspective of fact-checking practices across different media systems enabling a qualitative discussion of similarities, differences, and potential risks and developments in different contexts. In this work, we will also examine how fact-checkers (in different ways) seek to establish credibility and authority as knowledge gatekeepers and position themselves as trusted purveyors of facts. Lastly, the project involves an audience perspective to investigate the experience of being fact-checked and the potential tensions that fact-checkers might create when labelling “false” claims from not only elite actors but also citizens across a range of digital platforms. This part of the project relates to the discussion that Ananny has already raised about the assumptions that platform infrastructures make about public life and its alternative forms (Ananny 2020). With this research agenda, we hope to support the increasing reflexivity of epistemological beliefs and epistemic practices and to discuss with both scholars and fact-checkers how one can apply such reflexivity in the role, practice and communication of fact-checkers.

Conclusion

In this article, we have marked out an objectivist role perception in the academic literature on fact-checking and challenged the epistemological beliefs within this role. As a way forward, we have suggested ways to reinterpret this perception through rhetorical argumentation theory. To develop this argument, we have engaged in dialogue with recent scholarly suggestions that have facilitated reflection about the epistemology and epistemic practice of fact-checking in order to resolve the tensions between the objectivist role perception and role performances of fact-checkers. As a supplement to these ideas, we have argued that rhetorical argumentation theory should be considered a resource to further advance the epistemological discussions on the role perception of the fact-checkers, while the vocabulary from rhetorical argumentation theory holds potential for methodological operationalisation to guide the future role performances of fact-checkers.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Carlsberg Foundation (Carlsbergfondet) (Semper Ardens Accelerate, [grant number CF21-0456], “Tell Me the Truth: Fact-Checkers in an Age of Epistemic Instability”).

Notes

1 Referring to an objectivist perception, we thus point to something else than the critique of the so-called he said, she said reporting style, which—according to some fact-checkers—rely too strictly on an objectivity or neutrality norm, and thus passes along factual claims without assessing and adjudicating their veracity (Amazeen 2013, 19; Graves 2017, 518)