Literally does not always mean literally: a corpus-based diachronic study on literally as an intensifier

Abstract

This paper explores the adverb literallys linguistic state over the past century. Nine-hundred tokens of literally are collected from the Corpus of Historical American English and a social network service, Twitter. Three decades (the 1900s, 1950s, and 2000s) of data are examined as well as the data of 2016. The empirical data is investigated by literallys seven pragmatic functions and syntactic patterns. The results of data analysis are not consistent, but the overall results support the idea that literally is going through grammaticalization, though it is far from a complete internsifier.