“细读”之后,如何“远观”:数据驱动语言学对文学研究的方法论启示———刘海涛教授访谈

数字人文研究 ›› 2026, Vol. 6 ›› Issue (2): 18-26.

“细读”之后,如何“远观”:数据驱动语言学对文学研究的方法论启示———刘海涛教授访谈

刘永强，浙江大学外国语学院教授;陈子轩，浙江大学外国语学院博士生;刘海涛，复旦大学文科资深教授

出版日期:2026-06-28 发布日期:2026-07-16
基金资助:
中央高校基本科研业务费专项资金资助

After “Close Reading”, How to “Distantly Read”? Methodological Implications of Data-Driven Linguistics for Literary Studies: An Interview with Professor Liu Haitao

Online:2026-06-28 Published:2026-07-16

摘要/Abstract

摘要：

近年来，语言学在计量语言学基础上逐渐发展出数据驱动语言学，致力于从大规模语言数据中发现模式、提炼规律。以细读为主要传统的文学研究，是否也可能走向数据驱动?国内数据驱动语言学的重要代表学者刘海涛教授认为，文学研究并非不能谈论“模式”与“规律”，相反，数据驱动的文学研究已是一种现实趋势。数智时代文学研究的关键，不是用“远观”取代“细读”，而是根据研究问题重新组织二者的关系。在大语言模型迅速发展的背景下，数据驱动文学研究的意义也不只是方法更新，而是促使人文学者重新理解文本、数据、知识、智能与主体性之间的关系。由此，文学研究面临的不是一次简单的技术转向，而是一场从观念、材料到方法的整体性变革。

关键词: 数据驱动语言学 , 模式 , 规律 , 计量语言学 , 文学计算批评

Abstract:

In recent years,linguistics has progressively developed data-driven linguistics on the basis of quantitative linguistics,aiming to discover patterns and extract regularities from large-scale linguistic data.Can literary studies,with its long tradition of close reading,also move toward a data-driven approach? It is precisely with this question that this article interviews Professor Liu Haitao,a leading scholar of data-driven linguistics in China.Liu argues that literary studies need not shy away from discussing “patterns” and “regularities”; on the contrary,data-driven literary research is already an emerging trend.In the digital age,the key for literary studies is not to replace close reading with distant reading,but to reorganize the relationship between the two based on specific research questions.Against the backdrop of rapid advances in large language models,the significance of data-driven literary studies extends beyond methodological renewal to prompting humanists to rethink the relationships among text,data,knowledge,intelligence,and subjectivity.Accordingly,literary studies are facing not merely a simple technological turn,but a holistic transformation spanning concepts,materials,and methodologies.

Key words: data-driven linguistics , pattern , regularity , quantitative linguistics , computational literary criticism

中图分类号:

刘永强, 陈子轩, 刘海涛. “细读”之后,如何“远观”:数据驱动语言学对文学研究的方法论启示———刘海涛教授访谈[J]. 数字人文研究, 2026, 6(2): 18-26.

Liu Yongqiang, Chen Zixuan, Liu Haitao.

After “Close Reading”, How to “Distantly Read”? Methodological Implications of Data-Driven Linguistics for Literary Studies: An Interview with Professor Liu Haitao

[J]. Digital Humanities Research, 2026, 6(2): 18-26.

[1]	保罗·约瑟夫·斯彭斯, 雷纳塔·布兰多, 谢佳／译. 数字人文中的语言敏感性和多样性 [J]. 数字人文研究, 2022, 2(3): 3-20.
[2]	雅各布·格林陆玲怡/译程林/注. 谈非精准科学之价值[J]. 数字人文研究, 2021, 1(4): 12-14.
[3]	特奥多·蒙森石双彬/译. 拉丁铭文全集计划和执行方案[J]. 数字人文研究, 2021, 1(4): 15-.