Opportunities and challenges for traditional Chinese linguistics research in the era of big data
大数据时代背景下的中国传统语言学研究机遇与挑战

摘要

数字化技术的迅猛推进使得海量语言数据的采集、存储与分析成为可能，这一趋势正在深刻改变语言学研究的方法论格局。中国传统语言学（即“小学”，涵盖文字、音韵、训诂之学）拥有两千余年的学术积淀，但在数据驱动的研究范式面前，其既有方法和话语体系面临着更新压力。本文从大数据的基本特征出发，结合索绪尔关于“语言”与“言语”的经典区分，讨论数据驱动方法为传统语言学带来的三重机遇：语言现象的量化感知与自动化处理、基于语料证据的语法合法性梯度判断、语言共性与多样性的经验性解释。与此同时，文章也分析了数据质量参差、计算资源门槛、田野调查被弱化以及个人隐私与伦理等现实挑战，并就传统语言学如何在保持自身学术品格的前提下吸收计算方法提出若干思考。

Abstract

The rapid advancement of digital technologies has made possible the collection, storage, and analysis of massive linguistic data, a trend that is profoundly reshaping the methodological landscape of linguistic research. Traditional Chinese linguistics—known as xiaoxue (“elementary learning”), encompassing the studies of script/characters (wenzi), phonology (yinyun), and exegesis/semantics (xungu)—has accumulated an intellectual heritage of more than two millennia. Yet in the face of the data-driven research paradigm, its established methods and discursive systems are under pressure to renew themselves. Starting from the basic characteristics of big data, and drawing on Saussure’s classic distinction between langue and parole, this article discusses threefold opportunities that data-driven methods bring to traditional linguistics: the quantitative perception and automated processing of linguistic phenomena, corpus-evidence-based gradient judgments of grammatical well-formedness, and empirical explanations of linguistic universals and diversity. At the same time, the article examines practical challenges, including uneven data quality, the threshold of computational resources, the weakening of fieldwork, and issues of personal privacy and ethics. It concludes with reflections on how traditional linguistics can incorporate computational methods while preserving its own scholarly character.