A corpus-assisted study of vague language in corporate responsibility reports of the cosmetics industry

A corporate social responsibility (CSR) report is a form of information disclosure about environmental, health, and human rights issues. It serves as an essential bridge for communication between enterprises and stakeholders. Enterprises can use vague language (VL) to strategically communicate with the public, establish a positive corporate image, and legitimise their activities. In recent years, many scholars have studied the usage and communicative function of VL in general conversational discourse, yet VL in business discourse has not received much attention. In this study, 30 social responsibility reports of 10 world-renowned cosmetic enterprises between 2015-2019 were collected to build a corpus. With the help of the corpus software AntConc, this paper explores VL, specifically in CSR reports, using a corpus-assisted approach. Further, the quantitative findings are complemented by the contextualised interpretation of the communicative functions of VL in CSR reports. The major findings of this research are as follows. There are four types of VL used in CSR texts: quantity, degree, time, and softening stance-taking. In addition, the use of VL can help achieve several communicative purposes of CSR report drafting (i.e., provide an appropriate amount of information, enhance persuasion, and self-protection). Some practical suggestions are discussed for improving the awareness of the strategic use of vagueness in CSR reports.