人工智能时代语言研究“第五范式”论略

王春辉

语言战略研究 ›› 2026, Vol. 11 ›› Issue (3) : 14-27.

欢迎访问《语言战略研究》官方网站!今天是
语言战略研究 ›› 2026, Vol. 11 ›› Issue (3) : 14-27. DOI: 10.19689/j.cnki.cn10-1361/h.20260301
专题研究

人工智能时代语言研究“第五范式”论略

  • 王春辉
作者信息 +

A Brief Discussion on the “Fifth Paradigm” of Language Research in the Age of Artificial Intelligence

  • Wang Chunhui
Author information +
文章历史 +

摘要

科学研究历经实验范式、理论范式、计算范式和数据密集型范式,目前正在迈向第五范式,即人工智能驱动科学发现,人工智能不仅是辅助工具,更是知识生产的协作者与启发者。语言研究因其对象的特殊性,在这一转型中首当其冲。语言研究第五范式本质上是“人机协同生成机制”对语言研究的系统性重构。在知识生产上,体现为4个层面的根本性转变:认知主体从人类研究者重构为人‒AI复合认知系统;生成机制从解释性模式转向生成性机制;生产结构从线性流程转向循环系统;呈现形态从规则化走向分布化、参数化的概率表征。在研究实践上,体现为5个转型:研究对象从人类语言扩展到人机混合语言;研究方法上,AI不仅是被分析的对象,更是生成数据、模拟实验、辅助分析的方法论工具;研究目标不仅追求理解人类语言,也追求构建能够生成、理解、交互的语言智能体;研究主体从个体学者向人机协同升级;学科结构由单一学科走向多学科融合。这一转型为语言研究带来研究能力跃迁、理论创新空间拓展等重大机遇,同时也伴生了解释危机、理论弱化、知识权力集中、语言生态风险与伦理问题等深层挑战。面对第五范式的系统性重构,语言研究者不应仅仅是方法论的追随者,更应成为技术治理的批判者与语言伦理的建构者,保持人的主动性,并维护人的最终权利。

Abstract

Scientific research has progressed through experimental, theoretical, computational, and data-intensive paradigms, and is currently advancing toward the fifth paradigm: artificial intelligence-driven scientific discovery. AI is not only an auxiliary tool but also a collaborator and inspirer in knowledge production. Language research, due to the unique nature of its object, is at the forefront of this transformation. The fifth paradigm of language research is essentially a systematic reconstruction of language research “human-machine collaborative generation mechanism.” In knowledge production, this manifests in four fundamental shifts: the cognitive subject is reconstructed from human researchers to a human-AI composite cognitive system; the generation mechanism shifts from an interpretative model to a generative mechanism; the production structure shifts from a linear process to a cyclical system; and the presentation form shifts from rule-based to distributed, parameterized probabilistic representations. In research practice, this transformation manifests in five key areas: the research object has expanded from human language to human-machine hybrid language; in terms of methodology, AI is not only the object of analysis but also a methodological tool for generating data, conducting simulations, and assisting in analysis; the research goal is not only to understand human language but also to construct language intelligent agents capable of generating, understanding, and interacting; the research subject has upgraded from individual scholars to human-machine collaboration; and the disciplinary structure has shifted from a single discipline to a multidisciplinary integration. This transformation brings significant opportunities for language research, such as a leap in research capabilities and room for theoretical innovation. However, it also brings profound challenges, including an interpretive crisis, weakening of theories, concentration of knowledge power, risks to the language ecosystem, and ethical issues. Faced with the systemic reconstruction of the fifth paradigm, language researchers should not merely be followers of methodology, but rather critics of technological governance and builders of language ethics, maintaining human initiative and upholding ultimate human rights.

关键词

第五范式 / 语言研究 / 人机协同 / 人工智能 / 知识生产

Key words

fifth paradigm / language research / human-machine collaboration / artificial intelligence / knowledge production

引用本文

导出引用
王春辉. 人工智能时代语言研究“第五范式”论略[J]. 语言战略研究. 2026, 11(3): 14-27 https://doi.org/10.19689/j.cnki.cn10-1361/h.20260301
Wang Chunhui. A Brief Discussion on the “Fifth Paradigm” of Language Research in the Age of Artificial Intelligence[J]. Chinese Journal of Language Policy and Planning. 2026, 11(3): 14-27 https://doi.org/10.19689/j.cnki.cn10-1361/h.20260301
中图分类号: H002   

基金

国家社科基金重点项目“国际中文教育全球学术组织体系建构理念与实践路径研究”(24SGB015)。本文写作中得到编辑部同人诸多指导,在此深表谢意。文中谬误概由作者负责。

Accesses

Citation

Detail

段落导航
相关文章

/