published in: New Horizons in Prescriptivism Research, ed. Nuria Yáñez‐Bouza, María E. Rodríguez‐Gil, Javier Pérez‐Guerra. Bristol, Jackson: Multilingual Matters, pp. 266-295. https://www.multilingual-matters.com/page/detail/?K=9781800416147 Prescriptivism and Variation: The Greek Word for ‘Coronavirus’ Spiros A. Moschonas, Costas Mourlas and Thodoris Paraskevas 1 Introduction With the advent of the COVID pandemic, newspapers and news websites witnessed a remarkable increase in the use of the Greek translation loan for coronavirus. The newly rediscovered word appeared in four main variants (κορονοϊός, κορωνοϊός, κοροναϊός, κορωναϊός), while several metalinguistic texts prescribing the one or the other variant were also published in various news outlets. In this chapter we report on the effects which these prescriptive texts might have had on the variation of the Greek word for ‘coronavirus’. Our case study is structured as follows. In Section 2, based on available literature reviews, we present the research design that has been utilized in corpus-linguistic approaches that seek to detect the effects of prescriptivism on language change, and we indicate the ways in which our approach differs from and improves on previous work. In Section 3, we define our units of analysis, both at the metalinguistic level (corrective instructions) and at the linguistic level (orthographic variants). In Section 4, we present the metalinguistic corpus consisting of texts that prescribe the ‘correct’ spelling of the Greek loanword. We present the rationale for each proposal and argue that there is an internal logic to prescriptivism, a generator of prescriptions that vary with respect to the values assigned to just two binary metalinguistic features. In Section 5, we present the linguistic corpora that we have searched for variation, which are of three kinds: texts about the pandemic mined from news websites, tweets about the pandemic, and news radio broadcasts. The corpus of news articles is further divided into three phases: an early phase from 2013 and 2014, when the coronavirus was introduced as a technical term for viruses such as MERS-CoV and SARS-CoV; a second phase covering the early stage of the COVID pandemic (December 2019 – April 2020), which is the focus of our study; and a post-phase (a week in May 2020), during which, we hypothesized, the dominant trends, if any, would have crystallized. Radio broadcasts and tweets were only considered during this later phase. In Section 6 we present the quantitative results of our