Censorship of Online Research Journals in China

Lead Research Organisation: University College London
Department Name: Information Studies

Abstract

This research is the first to investigate the involvement of academic publishers in Chinese censorship practices. Since 2017, academic publishers have restricted access to research outputs in China at the request of state regulators. Cambridge University Press (CUP) and Springer Nature have removed over 1,315 journal articles featuring keywords such as 'Tiananmen', 'Tibet', 'Xinjiang' and 'Hong Kong' from their online platforms. LexisNexis (owned by Elsevier's parent company, RELX) and Taylor & Francis have removed a legal information database and 83 academic journals from sales packages sold to Chinese universities.

Due to constraints on library budgets and Open Access mandates, journal publishers are increasingly reliant on revenue growth in emerging markets outside the US and Europe. At the same time, the parameters of academic discourse in countries such as China have narrowed, particularly in relation to civil unrest in Hong Kong, the persecution of the Uighur ethnic group in Xinjiang, and the origins of the novel coronavirus, COVID-19, in Wuhan. These events raise urgent questions about the conditions of access to the Chinese market for scholarly research. To what extent have these censorship efforts affected the availability of resources in China? Are they encouraging scholars to self-censor? How should academic publishers respond to future demands to restrict access to politically sensitive content? Are they capable of resisting these demands? And is it in their interests to do so?

This is the first research to draw on work in political theory to establish a theoretical model for the censorship of academic journals. Some commentators have questioned the applicability of 'censorship' to a commercial transaction between an academic content distributor (publisher) and purchasing agent (importer), neither of whom have altered the content of academic output - i.e. the written text of academic journals and books sold to Chinese universities. Nor do either agent have any influence over the legal framework in China, which prohibits the distribution of politically sensitive material. However, insofar as the scholarly record is an integrated body of intellectual work-itself a text, as well as a corpus-the actions of publishers to limit access to individual articles within collections of content is editorial in nature. Such interventions may be justified in terms of inaccuracy or research ethics concerns, or due to external pressures to obscure research outcomes that undermine powerful individuals and institutions. Any effort to distinguish one form of intervention from the other must account for the motivations of agents to edit the scholarly record. CUP and Springer Nature's actions plausibly constitute self-censorship due to their complicity with the CCP's content restrictions, which are motivated by a desire to avoid political accountability.

My primary research will model empirically the consequences of this form of censorship. I will conduct in-depth, semi-structured interviews with 20-30 academics that have had their research censored in China, and 20-30 publishing professionals with expertise in the Chinese market for research periodicals. I will then use text mining and sentiment analysis to explore quantitatively the view that article-level censorship encourages self-censorship. This will be the first quantitative study of academic self-censorship involving the corpus of affected Asia Studies literature. The primary output of this PhD will be a risk analysis report, modelling the impact of censorship complicity in terms of reputational harm, impediments to research communication, and the financial consequences of forms of resistance. The aim of this report will be to inform industry guidelines via the Committee on Publication Ethics and the International Publisher's Association, drawing on my professional contacts in academic publishing and further connections made during my research.

Publications

10 25 50