The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics

Peng Xu

doi:10.47297/wspiedWSP2516-250026.20261004

PDF
导出
分享
收藏
专辑

The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics

10卷第4期

浏览量: 0 Downloads: 0

The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics
The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics
School of Languages and Cultures (School of International Communication and Exchange), Shanghai University of Political Science and Law, Shanghai, 201701, China
2026年10卷第4期页码：122-126
纸质出版：2026-04-30
DOI：10.47297/wspiedWSP2516-250026.20261004
Accepted：

Scan QR Code

Peng Xu. (2026). The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics. 国际教育与发展（英文版）, 10(4), 122-126.

Peng Xu.The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics[J].Journal of International Education and Development,2026,10(04):122-126.
Peng Xu. (2026). The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics. 国际教育与发展（英文版）, 10(4), 122-126. DOI： 10.47297/wspiedWSP2516-250026.20261004.

Peng Xu.The Contributions and Historical Limitations of Chomsky's Generative Linguistics to Contemporary Computational Linguistics[J].Journal of International Education and Development,2026,10(04):122-126. DOI： 10.47297/wspiedWSP2516-250026.20261004.

摘要

Abstract

This paper systematically evaluates Noam Chomsky's generative linguistics in terms of its contributions and historical limitations for computational linguistics. Chomsky's formalization of syntax—particularly context-free grammars (CFGs)—provided the mathematical foundation for early natural language processing systems and syntactic parsing algorithms. However

his theoretical idealization of linguistic competence over performance

the computational intractability of transformational grammars

the marginalization of semantics

and his rejection of probabilistic methods created persistent obstacles for engineering real-world language applications. By tracing the historical shift from rule-based symbolic approaches to statistical learning and then to deep learning

this paper argues that Chomsky's symbolism laid the foundational groundwork for computational linguistics

and that the future of the field lies in integrating symbolic and connectionist approaches—"walking on two legs"—to achieve both interpretability and robustness.

关键词

Keywords

references

Chomsky N. Aspects of the Theory of Syntax [M ] . Cambridge, MA: MIT Press , 1965 . p. 8 .

Chomsky N. Syntactic Structures [M ] . The Hague: Mouton , 1957 . pp. 13 - 24 .

Earley J. An efficient context-free parsing algorithm [J ] . Communications of the ACM , 1970 , 13 ( 2 ): 94 - 102 .

Marcus M. P. , Santorini B. , & Marcinkiewicz M. A. Building a large annotated corpus of English: the Penn Treebank [J ] . Computational Linguistics , 1993 , 19 ( 2 ): 313 - 330 .

Yang , C. Universal Grammar , statistics , or both? [J ] . Trends in Cognitive Sciences , 2004 , 8 ( 10 ): 451 - 456 .

Jurafsky D. , & Martin J. H. Speech and Language Processing: An Introduction to Natural Language Processing , Computational Linguistics, and Speech Recognition[M ] . 1st ed . Upper Saddle River, NJ: Pearson, 2019 : p. 14 .

Peters P. S. , & Ritchie R. W. On the generative power of transformational grammars [J ] . Information Sciences , 1973 , 6 : 49 - 83 .

Pereira F. Formal grammar and information theory: together again? [J ] . Philosophical Transactions of the Royal Society A , 2000 , 358 ( 1769 ): 1239 - 1253 .

Charniak E. Statistical parsing with a context-free grammar and word statistics [C ] // Proceedings of the Fourteenth National Conference on Artificial Intelligence (AAAI-97) . Menlo Park, CA : AAAI Press , 1997 : 598 - 603 .

Żelasko P. , Pappagari R. , & Dehak N. What helps transformers recognize conversational structure? Importance of context , punctuation, and labels in dialog act recognition[J ] . Transactions of the Association for Computational Linguistics, 2021 , 9 : 1225 - 1239 .

Deletang G. , Ruoss A. , Grau-Moya J. , et al . Neural networks and the Chomsky hierarchy [J ] . arXiv preprint , 2022 . arXiv: 2207.02098 .

Piantadosi S. Modern language models refute Chomsky's approach to language [C ] // Gibson, E., & Poliak, M. (Eds.), From Fieldwork to Linguistic Theory: A Tribute to Dan Everett . Berlin : Language Science Press , 2024 : 353 - 414 .

Portelance E. , & Jasbi M. On the compatibility of generative AI and generative linguistics [J ] . Nature Computational Science , 2025 , 5 ( 9 ): 1 - 9 .

Smith N. A. , & Johnson M. Weighted and probabilistic context-free grammars are equally expressive [J ] . Computational Linguistics , 2007 , 33 ( 4 ): 477 - 491 .

Marcus G. Deep learning: a critical appraisal [EB/OL ] . arXiv preprint , 2018 . arXiv: 1801.00631 .

Tal ,. L, & Baroni, Marco, B. Syntactic structure from deep learning [J ] . Annual Review of Linguistics , 2021 , 7 : 195 - 212 .

Chrupała G. . Symbolic inductive bias for visually grounded learning [J ] . Computational Linguistics , 2019 , 7 : 6452 – 6462 .

文章被引用时，请邮件提醒。

Submit

Research on the Tourism Development Strategy of Red Tourism Scenic Spots

Analysis on Short Video Supporting Traditional Cultural Industry Innovation and Development

Teacher Emotional Support and Second Language Acquisition: Cross-Cultural Differences in Psychological Needs Based on SDT and Pedagogical Responses

A Comparative Study on the Objectives and Implementation Paths of Basic Education Supervision between New Zealand and China

AI Large Model Coupled Competency Modeling: Research on the Growth Path and Career Development Empowerment of College Student Assistants Transforming into Collaborative Managers

Related Author

No data

Related Institution

No data

⁰