Does meat cause cancer? The discursive construction of meat carcinogenicity in a corpus of scientific texts

In 2015, the International Agency for Research on Cancer (IARC) published a report on the carcinogenicity of red and processed meat, incorporating red meat in Group 2A carcinogens (probably carcinogenic to humans) and processed meat in Group 1 (carcinogenic to humans). This announcement attracted immediate interest from other scientists, especially in medical research, where the relation between cancer and food has been investigated extensively for many years. This paper aims to analyze the discursive construction of meat carcinogenicity in a set of scientific papers published in the wake of the IARC communiqué. For this purpose, an electronic corpus was assembled from a range of academic journals featured in the database Elsevier Science Direct, for a total of 384,491 words, which were fully POS-tagged, partially parsed using a systemic functional grammatical formalism, and subsequently analyzed on Antconc. The methodology adopted to analyze these data is a combined corpus assisted discourse analysis approach, focusing mainly on experiential noun group structures, specifically those involved in patterns of nominalization, which typically aim to achieve monorefentiality in scientific discourse. However, in this corpus, the denotational boundaries of meat (what animal-based foods count as meat or meat products; what animals have red rather than dark or white meat; the exact nature of meat processing) are not entirely clear, and this "semantic debate" (Lippi et al. 2016, p. 2) is central to the preoccupations of medical and nutrition experts. Therefore, conclusions show that linguists could make a useful contribution to cancer science by devising a set of universally agreed definitions of meat types, so as to agree on the level of health risk that each may cause.