ABSTRACT
This chapter demonstrates how corpora can be used to examine the language use of different social groups and how this can change over time. We compare data from the spoken section of British National Corpus 1994 (BNC1994) with the spoken British National Corpus 2014 (BNC2014). The text in both corpora has been tagged for speaker identity (e.g. age, gender and social class), and we examine how age groups use the modal verb may in both corpora. Our research indicates a complex picture of change with some age groups (e.g. 0 to 14) having more use of may in 2014, although others (e.g. 35 to 44) using may less (such changes can be attributed to an age effect). The two corpora can also be used to track birth cohorts e.g. people born in the 1980s would be in the 0 to 14 age group in 1994 and in the 25 to 34 group in 2014. Comparing these two groups helps to discern whether changes are a result of a cohort effect. Our analysis also used concordancing in order to identify how functional use of may changes over time, with the 0- to 14-year-olds using may to request permission more in 1994 than the equivalent 0 to 14 age group in 2014.
