Which One of Should and Had Better: A Corpora-based Analysis

Abstract

The ultimate goal of this paper is to provide a frequency analysis of should and had better within the Corpus of Contemporary American English, the British National Corpus, the Time Magazine Corpus, the Hansard Corpus, and the Corpus of Historical American English. The COCA clearly indicates that should may be the preferable type for Americans in eight genres, whereas had better may not. From the overall frequency of should and had better, it is clear that Americans tend to avoid had better since there is always a danger or problem if the hearer does not follow the advice. The BNC clearly shows that as in the case of the COCA, should is the preferable type for British people in seven genres. This may imply that the type should may be preferred over the type had better by British people. When it comes to the TMC, it is interesting to note that the type should was the preferable type for educated Americans. It is significant to note that the type should may have been the most preferred by educated Americans in the 2000s, but it may have been the most undesired type in the 1920s. With respect to the HC, it is worth noting that the type should was preferred by British politicians for 200 years from 1800 to 2000. From the overall frequency of should and had better in the HC, it is clear that the type should was preferred over the type had better by British politicians for 200 years. Finally, as for the COHA, it is noteworthy that the type should was preferred over the type had better by Americans from 1810 to 2000. The frequency of should reached a peak in 2000, but should had the lowest frequency in 1810. This in turn suggests that the type should was the most preferred by Americans in 2000, but it was the least preferred among Americans in 1810.