Have Large Language Models Enhanced the Way Civil & Environmental Engineers Write? A Quantitative Analysis of Scholarly Communication over 25 Years

TL;DR

This study analyzes 25 years of civil and environmental engineering abstracts to detect linguistic shifts indicative of large language model assistance, revealing recent changes in writing style, complexity, and tone since 2023.

Contribution

It introduces a vocabulary-based methodology using NLP to quantify LLM influence on scholarly writing in civil and environmental engineering.

Findings

Post-2023, increased lexical diversity and complexity in abstracts.

Decreased passive voice and hedging language in recent abstracts.

Stylistic shifts suggest widespread LLM adoption in scholarly communication.

Abstract

Large language models (LLMs) have rapidly emerged in civil and environmental engineering (CEE) research, education, and practice as tools for project ideation, execution, and communication. However, it is unknown how prevalent LLM adoption is across CEE scholarship and whether it measurably alters research prose. Inspired by recent analyses of biomedical research, this study uses a vocabulary-based frequency-shift methodology to detect linguistic signals of LLM-assisted writing in a large corpus of CEE literature. A total of 149,452 abstracts published by the American Society of Civil Engineers from 2000 through 2025 are analyzed to quantify deviations from long-term vocabulary trends. Prior to the introduction of LLMs in 2022, CEE publications exhibit long-term trends toward longer abstracts and sentences, greater use of segmenting punctuation, higher required reading levels, and a shift toward active, first-person verb constructions. Beginning around 2023, however, the frequencies of many stylistic marker words (e.g., enhance) sharply depart from historical trajectories, accompanied by deviations in multiple semantic properties. Abstracts classified as likely LLM-assisted exhibit increased lexical diversity, comma use, and complexity, with reduced passive voice and hedging language, producing prose that is more segmented, complex, and confident. The AI contribution of this study lies in the use of natural language processing to identify population-level linguistic signals of LLM-assisted text, applied to quantify the prevalence of LLM use and its influence on the vocabulary, structure, and tone of engineering scholarly writing. Together, these findings provide the first large-scale, data-driven assessment of how LLMs are beginning to reshape scholarly communication in CEE.