Abstract
The present paper focuses on non-standard functions of like in spoken discourse. Like has received wide attention from scholars over the past 30 years and with the recent publication of Alexandra D'Arcy's exhaustive work "Discourse-Pragmatic Variation in Context: Eight hundred years of LIKE", the subject appeared to be depleted.
Nevertheless, non-standard like is a phenomenon that is in motion. The launch of the new Spoken British National Corpus 2014 presents a chance to compare the current use of non-standard like with data from the original Spoken BNC1994.
Their comparable architecture renders them suitable for diachronic research.1 The aim of the paper is to compare the frequency of use of the following non-standard functions in the two corpora: hedge, focus marker, discourse marker, and a quotative marker. Comparing the relative frequency of both standard and non-standard like in BNC2014 (13,781.87 instances per million words) and the BNC1994 (3,668.45 instances per million words) the results clearly imply a steep rise in use, most likely caused by the spread of non-standard like.
The paper examines two hundred examples of non-standard like, which had to be extracted manually from randomly ordered results. Only 167 hits were needed in BNC2014 to obtain 100 relevant tokens in comparison to 426 hits in BNC1994.
The results suggest that more than half of the tokens of like in the BNC2014 corpus are non-standard uses, while in the BNC1994 it is approximately every fourth token. Preliminary results suggest that like used as a hedge, a focus marker, and a discourse marker are devices that are consistently employed by speakers of British English in both corpora.
However, the relative frequency of non-standard like rapidly increases in the BNC2014 data. The quotative marker is of particular interest, since it can be shown to have quadrupled its frequency in the new BNC2014 compared with the older BNC.