Wittgenstein and AI Prompt Engineering Techniques: How to Improve Your Prompts for Academic Research Writing with Wittgenstein
Dr. Alvin Chan, DBA
Global Professor of the Practice
Golden Gate University, United States
Author Note
Dr. Alvin Chan, DBA
Correspondence concerning this article should be addressed to Dr. Alvin Chan, Global Professor of the Practice, Golden Gate University, United States.
Email: alchan@ggu.edu and alvinchan88@gmail.com
Acknowledgement
The author gratefully acknowledges Steven Wu Bing Shuo, a 16-year-old student at Hailiang High School in Zhejiang, China, whose curiosity, questions, and interest in the philosophy of Ludwig Wittgenstein and artificial intelligence inspired the development of this article.
Disclosure of Generative Artificial Intelligence Use
The author used a generative artificial intelligence tool to assist with drafting, organising, editing, and formatting parts of this manuscript. The author formulated the research references and the philosophical framework, reviewed and revised all content, verified the appropriateness of quotations and references, and takes full responsibility for the final article.
Abstract
This article applies Ludwig Wittgenstein’s philosophy of language to prompt engineering for academic research writing. It argues that a prompt is not merely an instruction to a machine but a language game: a language exchange whose meaning depends on purpose, audience, context, evidence, and the practical activity it supports. Drawing principally on Wittgenstein’s later philosophy, especially the ideas of meaning as use, language games, family resemblance, rule following, and certainty, the article offers eight practical techniques for improving prompts used in academic writing. These techniques are: clarify the research task; define audience, genre, and purpose; supply relevant context and evidence; specify disciplinary conventions and structure; use iterative dialogue; guard against linguistic bewitchment; verify claims and sources; and practise ethical disclosure. The article also provides prompt templates and examples for literature reviews, research questions, argument development, methodology, discussion, and revision. It concludes that effective prompting requires epistemic humility: artificial intelligence can support academic writing, but it cannot replace the researcher’s responsibility for judgement, evidence, interpretation, and accountability.
Keywords: artificial intelligence, academic writing, Wittgenstein, language games, prompt engineering, research writing, generative artificial intelligence, scholarly communication