Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Recontextualized Knowledge and Narrative Coalitions on Telegram

Tom Willaert

TL;DR

This study investigates how conspiracy narratives on Telegram recontextualize prior knowledge by linking to diverse sources through hyperlinks, and how these recontextualizations form narrative coalitions among heterogeneous actors. It adopts bibliometric-inspired methods, specifically a bibliographic coupling framework using the OpenAlex knowledge graph, applied to the Pushshift Telegram Dataset to map channel-level knowledge references and detect community coalitions via Louvain clustering. The findings reveal six major narrative coalitions—spanning science/technology imaginaries, far-right antisemitic conspiracies, Marxist/anarchist discourse, German diagonalism, Spanish-language feminist/leftist networks, and Brazil-focused channels—each shaped by ideological orientation, regional language, and political stance. The work demonstrates a transferable methodological approach for tracing how online conspiracy theories mobilize established knowledge sources, contributing to understandings of the communicability, ontologies, and power dynamics of online conspiracism and highlighting the value of archival social-media data for digital disinformation research.

Abstract

A defining characteristic of conspiracy texts is that they negotiate power and identity by recontextualizing prior knowledge. This dynamic has been shown to intensify on social media, where knowledge sources can readily be integrated into antagonistic narratives through hyperlinks. The objective of the present chapter is to further our understanding of this dynamic by surfacing and examining 1) how online conspiracy narratives recontextualize prior knowledge by coupling it with heterogeneous antagonistic elements, and 2) how such recontextualizing narratives operate as connectors around which diverse actors might form narrative coalitions. To this end, the chapter offers an empirical analysis of links to prior knowledge in public messaging channels from the Pushshift Telegram dataset. Using transferable methods from the field of bibliometrics, we find that politically extreme Telegram channels engage with a variety of established knowledge sources, including scientific journals, scientific repositories and other sources associated with the system of scholarly communication. Channels engaging with shared knowledge sources thereby form narrative coalitions ranging from scientific and technological imaginaries to far-right extremist and antisemitic conspiracy theories. Our analysis of these coalitions reveals (i) linguistic, political, and thematic forces that shape conspiracy narratives, (ii) emerging ideological, epistemological and ontological positions associated with online conspiracism, and (iii) how references to shared knowledge contribute to the communicability of conspiracy narratives.

Recontextualized Knowledge and Narrative Coalitions on Telegram

TL;DR

This study investigates how conspiracy narratives on Telegram recontextualize prior knowledge by linking to diverse sources through hyperlinks, and how these recontextualizations form narrative coalitions among heterogeneous actors. It adopts bibliometric-inspired methods, specifically a bibliographic coupling framework using the OpenAlex knowledge graph, applied to the Pushshift Telegram Dataset to map channel-level knowledge references and detect community coalitions via Louvain clustering. The findings reveal six major narrative coalitions—spanning science/technology imaginaries, far-right antisemitic conspiracies, Marxist/anarchist discourse, German diagonalism, Spanish-language feminist/leftist networks, and Brazil-focused channels—each shaped by ideological orientation, regional language, and political stance. The work demonstrates a transferable methodological approach for tracing how online conspiracy theories mobilize established knowledge sources, contributing to understandings of the communicability, ontologies, and power dynamics of online conspiracism and highlighting the value of archival social-media data for digital disinformation research.

Abstract

A defining characteristic of conspiracy texts is that they negotiate power and identity by recontextualizing prior knowledge. This dynamic has been shown to intensify on social media, where knowledge sources can readily be integrated into antagonistic narratives through hyperlinks. The objective of the present chapter is to further our understanding of this dynamic by surfacing and examining 1) how online conspiracy narratives recontextualize prior knowledge by coupling it with heterogeneous antagonistic elements, and 2) how such recontextualizing narratives operate as connectors around which diverse actors might form narrative coalitions. To this end, the chapter offers an empirical analysis of links to prior knowledge in public messaging channels from the Pushshift Telegram dataset. Using transferable methods from the field of bibliometrics, we find that politically extreme Telegram channels engage with a variety of established knowledge sources, including scientific journals, scientific repositories and other sources associated with the system of scholarly communication. Channels engaging with shared knowledge sources thereby form narrative coalitions ranging from scientific and technological imaginaries to far-right extremist and antisemitic conspiracy theories. Our analysis of these coalitions reveals (i) linguistic, political, and thematic forces that shape conspiracy narratives, (ii) emerging ideological, epistemological and ontological positions associated with online conspiracism, and (iii) how references to shared knowledge contribute to the communicability of conspiracy narratives.
Paper Structure (18 sections, 5 figures)

This paper contains 18 sections, 5 figures.

Table of Contents

  1. Introduction
  2. Methodology
  3. Bibliographic coupling analysis
  4. Data processing pipeline
  5. Findings
  6. Heterogenous couplings
  7. Narrative coalitions
  8. Science and technology
  9. Far-right extremism
  10. Marxism, anarchism, and technology
  11. German-language diagonalism and conspiracism
  12. Spanish-language feminism, anarchism, left-wing politics
  13. Channels pertaining to Brazil
  14. Discussion
  15. Conclusion and avenues for future research
  16. ...and 3 more sections

Figures (5)

  • Figure 1: Illustration of bibliographic coupling of two Telegram Channels. Both Telegram channel 1 and Telegram channel 2 contain at least on reference to url1, url2 and url3. As such it can be argued that both channels are connected with a bibliographic coupling frequency of 3.
  • Figure 2: Overview of the main steps in our processing pipeline.
  • Figure 3: Illustration of a simple citation matrix. Each column represents a Telegram channel, rows represent sources. If a channel refers to a source, we set the value to 1, otherwise we set the value to 0.
  • Figure 4: Overview of 50 most frequent domains in top 10,000 channels by number of links to OpenAlex sources, segmented by: combined source types (a), OpenAlex sources (b), non-OpenAlex sources (c)
  • Figure 5: Visualization of bibliographic coupling network of Telegram channels. Nodes represent channels, weighted edges the bibliographic coupling frequency (BCF) between two channels based on OpenAlex sources cited by both channels. Only edges representing a BCF higher than or equal to 2 are retained. Nodes are coloured according to Louvain communities (narrative coalitions). Node positioning: ForceAtlas2 algorithm, Node size: node degree.