Hallucination as Pragmatic Failure: A Theoretical Reframing of Large Language Models
DOI:
https://doi.org/10.64744/tjiss.2025.38Keywords:
Performativity, Human-Computer Interaction, Hallucination, Large Language Models, Felicity Conditions, Speech Act Theory;, PragmaticsAbstract
This paper reframes hallucination in large language models (LLMs) as a pragmatic failure rather than a purely semantic or statistical defect. Building on speech-act distinctions between locution, illocution, and perlocution, we argue that LLM outputs often function as actions without passing through an intentional, context-sensitive layer that would license those actions. In this view, hallucination is an infelicitous performative: an assertion or directive issued without adequate authority, evidence, or situational fit. The paper develops a conceptual mapping from speech-act structure to the LLM interaction pipeline and proposes a pragmatic layer that constrains when generated text may “count” as an assertion, instruction, or commitment. Rather than claiming to eliminate hallucination by changing the probabilistic core of LLMs, the account narrows its performative scope through felicity-aware gating—deferring, hedging, or refusing when contextual conditions are unmet. The contribution is theoretical: a concise framework for understanding intentionlessness in smart systems and for unifying existing mitigations (e.g., retrieval grounding and guardrails) as pragmatic constraints between generation and action.
References
Austin, J. L. (1962). How to do things with words. Oxford University Press.
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’21), 610-623.
Clark, H. H. (1996). Using language. Cambridge University Press.
Fischer, K. (2009). The role of users’ preconceptions in talking to computers and robots. International Journal of Social Robotics, 1(1), 83-93.
Gao, Y., Zeng, A., Zhang, R., Liu, Y., Chen, G., Lin, Z., Wang, X., & Lin, J. (2023). Retrieval-augmented generation for large language models: A survey. arXiv preprint, arXiv:2312.10997.
Gregor, S. (2006). The nature of theory in information systems. MIS Quarterly, 30(3), 611-642.
Grice, H. P. (1975). Logic and conversation. In P. Cole & J. L. Morgan (Eds.), Syntax and semantics, Vol. 3: Speech acts (pp. 41-58). Academic Press.
Hevner, A. R., March, S. T., Park, J., & Ram, S. (2004). Design science in information systems research. MIS Quarterly, 28(1), 75-105.
Huang, L., Yu, W., Ma, W., Zhong, W., Feng, Z., Wang, H., Chen, Q., Peng, W., Feng, X., Qin, B., & Liu, T. (2023). A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions. arXiv preprint, arXiv:2311.05232.
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y., Madotto, A., & Fung, P. (2023). Survey of hallucination in natural language generation. ACM Computing Surveys, 55(12), 248:1-248:38.
Levinson, S. C. (1983). Pragmatics. Cambridge University Press.
Luger, E., & Sellen, A. (2016). “Like having a really bad PA”: The gulf between user expectation and experience of conversational agents. Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, 5286-5297.
Manakul, P., Laban, P., Tsvetkov, Y., & Soricut, R. (2023). SelfCheckGPT: Zero-resource black-box hallucination detection for generative large language models. arXiv preprint, arXiv:2303.08896.
Matthias, A. (2004). The responsibility gap: Ascribing responsibility for the actions of learning automata. Ethics and Information Technology, 6(3), 175-183.
Meta. (2024). Llama Guard 3 model card. Hugging Face.
NVIDIA. (n.d.). NeMo Guardrails documentation. NVIDIA Developer Documentation.
Porcheron, M., Fischer, J. E., Reeves, S., & Sharples, S. (2018). Voice interfaces in everyday life. Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI ’18), 1-12.
Searle, J. R. (1969). Speech acts: An essay in the philosophy of language. Cambridge University Press.
Wang, X., Wei, J., Schuurmans, D., Chi, E. H., & Zhou, D. (2022). Self-consistency improves chain-of-thought reasoning in language models. arXiv preprint, arXiv:2203.11171.
Zhao, P., Li, Y., Yang, Y., Wang, L., & Jiang, H. (2024). Retrieval-augmented generation for AI-generated content: A survey. arXiv preprint, arXiv:2402.19473.
Downloads
Published
Data Availability Statement
No new data were created or analysed in this study.
This article is based entirely on theoretical and conceptual analysis; therefore, no datasets are associated with this submission.
