One aspect that has often been drawn on to explain the differences between varieties of English is the influence of substrata. While it remains undisputed that substratal influence is a powerful explanatory factor, its limits are equally obvious. In a preliminary analysis, drawing on data from the Corpus of Global Web-Based English (GloWbE), I could show that despite their shared substratum, Chinese, Hong Kong English (HKE) and Singapore English (SingE) show different preferences for morphemic vs non-morphemic word-formation processes. Whereas SingE showed a high frequency of demonyms such as the Chinese and the Singaporeans, HKE preferred compounded, more explicit variants such as Chinese people and Singapore(an) people, potentially modelling on the Chinese substrate. The question whether this finding is also applicable to the highly frequent word-formation process of conversion is the scope of the present study.

This study further aims to show that substrate influence interacts with frequency as an explanatory factor in language contact settings. In the study of conversion, frequency is a factor that has not yet received much attention. The frequency-related aims are twofold: The first question is whether the frequency of a verb is decisive for it being converted into a noun; i.e. can frequency explain what we observe in the contact variety? Second, the frequency of use or productivity of a word-formation process in the contact variety is seen as an indicator for the influence of a substratum. Thus, a higher frequency of a pattern in the contact variety is interpreted as the result of a stronger influence of the substratum.

In my study, I focus on verb-to-noun conversion since this process is very frequent in all Chinese languages but not very common in English. Thus, if language contact is decisive in shaping emergent varieties of English, HKE and SingE are expected to show higher frequencies of V>N conversions than established varieties, since Chinese as a non-inflecting language is hypothesized to promote this direction of conversion even further. In order to compare conversion in varieties of English, I draw on data from GloWbE, a corpus containing over 1.9 billion words from twenty English-speaking countries.