Abstract

This study examines whether and how Corporate Social Responsibility reports of firms in the automobile industry differ in regards to their linguistic composition and their decoupling tendencies. Based on Basu and Palazzo’s process model, the Volkswagen emission scandal (‘Dieselgate’) is analyzed from a linguistic content perspective by comparing Volkswagen’s CSR reports with those of its biggest competitors. By applying the Linguistic Inquiry and Word Count Software, we identify linguistic characteristics that can help to detect deception markers. More specifically, we show how CSR reports of de-couplers and implementors differ in terms of morphology, syntax, semantics and pragmatics by developing and applying a linguistic analysis’ framework. The data provided mixed support for our hypotheses. The analysis revealed that de-couplers, among other factors, use more articles, more words per sentences, a more negative emotional tone and fewer words connected to risks and anxiety, while verb tense and informal language only had ambiguous or even no influence. The study contributes to the current understanding of CSR reporting, the debate on CSR decoupling, and the linguistic perspective of CSR research. Moreover, it specifies the linguistic domain of Basu and Palazzo’s process model.

摘要

本研究探讨了汽车行业公司的企业社会责任报告在语言组成和脱钩趋势方面是否以及如何不同。基于Basu和Palazzo的过程模型，通过将大众汽车的CSR报告与其最大竞争对手的报告进行比较，从语言内容的角度分析了大众汽车排放丑闻（“ Dieselgate”）。通过应用语言查询和单词计数软件，我们可以识别有助于检测欺骗标记的语言特征。更具体地说，我们通过开发和应用语言分析框架来说明解耦器和实现器的CSR报告在形态，语法，语义和语用上的差异。数据为我们的假设提供了混合支持。分析表明，解耦器 除其他因素外，使用更多的文章，每个句子更多的单词，更消极的情绪语调以及与风险和焦虑相关的单词更少，而动词时态和非正式语言仅具有歧义甚至没有影响。这项研究有助于当前对CSR报告的理解，关于CSR去耦的辩论以及CSR研究的语言学视角。而且，它指定了Basu和Palazzo的过程模型的语言领域。