Narrative Media Framing in Political Discourse
Yulia Otmakhova, Lea Frermann · 2025
Read Narrative Media Framing in Political Discourse by Yulia Otmakhova, Lea Frermann (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-00737.txt.
Introduction
Otmakhova and Frermann study how news stories frame politics through narrative, in a paper published in 2025. Frames that cast events as stories, with heroes, villains, and arcs, sway readers strongly. Automated analysis, though, has mostly ignored this device. The authors link elements of storytelling to the mechanics of framing and formalize the connection. They annotate a dataset of climate change news articles by narrative components.
With that data they chart which components dominate across political leanings. They also test how well language models can predict the frames. Finally, they apply the framework without supervision to COVID-19 coverage, where its output matched prior theory. The work gives computational social science a new lens on persuasion. It shows narrative framing can be measured, not just admired.
Catalog record
| CREATOR | Yulia Otmakhova, Lea Frermann |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 77 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-00737
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Narrative Media Framing in Political Discourse Yulia Otmakhova, Lea Frermann Abstract Narrative frames are a powerful way of conceptualizing and communicating complex, controversial ideas, however automated frame analysis to date has mostly overlooked this framing device. In this paper, we connect elements of narrativity with fundamental aspects of framing, and present a framework which formalizes and operationalizes such aspects. We annotate and release a data set of news articles in the climate change domain, analyze the dominance of narrative frame components across political leanings, and test LLMs in their ability to predict narrative frames and their components. Finally, we apply our framework in an unsupervised way to elicit components of narrative framing in a second domain, the COVID-19 crisis, where our predictions are congruent with prior theoretical work showing the generalizability of our approach. 1 1 1 We release our code, data and annotations at https://github.com/julia-nixie/narratives . 1 Introduction Narrative framing is a type of media framing that uses elements of narrativity to highlight some aspects of a complex issue and condense it into a simplified “story” that promotes a particular interpretation Crow and Lawlor 2016 . These elements of storytelling, such as representing an issue through the lens of stakeholders and conflicts rather than direct description of the facts, make narrative frames a highly effective device, particularly in the context of contested issues such as climate change ( Daniels and Endfield 2009 ; Rodrigo-Alsina 2019 ) . Narrative framing can draw the reader’s attention to specific, nuanced aspects of an issue and instill a very precise interpretation that differs from the “default” reading inferrable from its generic or issue-specific frame. To give an example, the text in Figure 1 frames the topic of climate change through a “Polar Bear” issue-specific frame Bushell et al. 2017 which describes the negative effects of climate change on animals ( rising temperatures have certainly put a strain on species ). However, this is not the actual message of the text: it depicts climate scientists as incorrect, while presenting pseudo-scientists from a hero-like angle. It uses devices of narrative framing to replace the default interpretation arising from a topic-like frame (“animals are victims of climate change and humans are villains”) with an opposing idea that animals are doing fine and scientists who claim otherwise mislead the general public. Figure 1: An except from a news article, with hero marked in green and villain in red. Entities that are not main characters are grayed out. The box shows the focal character (here, hero ). The phrases in italic are cues which show that the article has an individualistic cultural story (“the nature can fix itself”) and that it fuels conflict by actively promoting bad science. While the importance of framing narratives for the communication and perception of news has been widely recognized in the social sciences Shanahan et al. 2011 , automatic framing analysis still mostly conceptualizes frames in a topic-like fashion ( Ali and Hassan 2022 ; Otmakhova et al. 2024 ) . Recent work in NLP has studied elements of narratives such as characters or events in news reporting ( Stammbach et al. 2022 ; Frermann et al. 2023 ; Gehring and Grigoletto 2023 ; Zhao et al. 2024 ; Das et al. 2024 ) . However, these studies are either topic-specific, or lose the link to the core mechanism of framing: For framing to occur, an ambivalent issue must be present Sniderman and Theriault 2004 and described in a way that evokes a larger interpretative context (schema) which goes beyond information directly inferrable from the text ( Scheufele and Scheufele 2010 ) . As such, beyond the issue itself (aka Entman 1993 ’s “problem statement”), there are other important mechanisms that turn a message into a frame, such as alluding to the conflict that led to the problem (=its cause), making a moral evaluation , or suggesting a resolution to it. We present a general formalization of narrative framing that comprises all these aspects. We do so by integrating elements of narratology with social and media studies on narrative framing, to establish a framework which allows to identify and structurally represent narrative frames and to distinguish superficially similar frames from one another. First, our framework distinguishes character roles, and we show how issue ambivalence arises when actors in an article are assigned an archetypical role (Hero, Villain, Victim) and one of the roles is drawn into focus. This emphasis in turn evokes a moral evaluation of the personas in the article. As a second component, we position each character in terms of exacerbating or resolving the core conflict of the article. Finally, to link the presentation in the article to the wider set of associations and beliefs already existing in the receiver’s perception ( Nelson et al. 1997 ) , we link our narrative frames to established “cultural stories” that define the attitude towards external control and the sense of unity with the group ( Thompson 2018 ) . Figure 1 illustrates the three components with an example. We apply our framework to analyze media framing of two distinct public issues – climate change and COVID-19. In particular, we make the following contributions: 1. We define elements of narrativity that are essential for narrative framing and are aligned with the definition of a media frame and show that our framework applies across topics (climate change and COVID-19) and domains (news articles and political speeches). 2. We show that framework enables reliable and effective annotation of narrative frames in the news, and improves the automatic detection of narratives. 3. We release a corpus of articles about climate change annotated with narrative frames, and use it to analyze the distribution of different narrative frames across political leanings. 4. We test a range of LLMs on their ability to automatically predict the components of our framework, leaving room for improvements. 2 Background Narratives in political communication Following Fisher 1984 ’s seminal paper coining the term “homo narrans” to illustrate the importance of storytelling for society, narratives in political communication have attracted substantial research attention (see also Bennett and Edelman 1985 ; Patterson and Monroe 1998 ), exposing its effects from a critical vehicle in deliberative democracy Boswell 2013 to its use persuasive device Skrynnikova et al. 2017 . Similar to the concept of framing in general, a principled and empirically testable definition of “narrative framing” has long been lacking. However, recent work has progressed in developing frameworks that are testable and amenable to computational modeling Shenhav 2005 ; Robert and Shenhav 2014 , most prominently the Narrative Policy Framework (NPF; Jones et al. 2023 ), which we build on in this work. The NPF defines a set of generalizable structural elements in political/policy narratives, including characters, settings, plot and moral evaluation which it uses to characterise the operation of narratives on the individual, group and cultural level. The NPF has been particularly instrumental in studying climate change narratives Fløttum and Gjerstad 2017 , and identifying dominant narratives in the discourse Bushell et al. 2017 ; Bevan 2020 . This paper adapts the elements of the NPF into a structured framework designed to support automated prediction. Narrative framing and NLP Narrative framing intersects the concepts of