Detecting Religious Language in Climate Discourse

2025-10-27

AI Summary

This study investigates the distribution of explicit and implicit religious language in climate discourse across secular and religious NGO texts. Methodologically, it employs a dual-path approach: (1) a rule-based model grounded in ecological theology, operationalized via a hierarchical taxonomy of religious terminology, and (2) a zero-shot large language model (LLM). Both methods are rigorously compared on a corpus exceeding 880,000 sentences. Results indicate that the rule-based model identifies significantly more religious utterances, exposing persistent tensions between lexical definitions and contextual semantics—and thereby revealing systematic methodological divergences and limitations in religious language detection. The study advances digital religion research through methodological innovation, particularly in bridging theological frameworks with computational linguistics. Moreover, it provides empirical grounding for understanding how sacred expression persists and adapts within ostensibly secular environmental discourse, contributing to broader debates on religion’s discursive resilience in post-secular public spheres.

Technology Category

• Natural Language Processing: Discourse, Pragmatics & Argument Mining
• Data Mining & Knowledge Management: Semantic Web
• Knowledge Representation and Reasoning: Knowledge Representation Languages

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Sentiment analysis and opinion mining
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Religious language continues to permeate contemporary discourse, even in ostensibly secular domains such as environmental activism and climate change debates. This paper investigates how explicit and implicit forms of religious language appear in climate-related texts produced by secular and religious nongovernmental organizations (NGOs). We introduce a dual methodological approach: a rule-based model using a hierarchical tree of religious terms derived from ecotheology literature, and large language models (LLMs) operating in a zero-shot setting. Using a dataset of more than 880,000 sentences, we compare how these methods detect religious language and analyze points of agreement and divergence. The results show that the rule-based method consistently labels more sentences as religious than LLMs. These findings highlight not only the methodological challenges of computationally detecting religious language but also the broader tension over whether religious language should be defined by vocabulary alone or by contextual meaning. This study contributes to digital methods in religious studies by demonstrating both the potential and the limitations of approaches for analyzing how the sacred persists in climate discourse.

Problem

• Detecting religious language in climate discourse texts
• Comparing rule-based and LLM methods for identification
• Analyzing definition tensions between vocabulary and context

Innovation

• Rule-based model using hierarchical religious terms tree
• Large language models in zero-shot setting
• Comparing methods on 880,000 sentence dataset