Topic

So I was thinking today, reading Mufwene's Ecology of Language Evolution, and I came up with a possible idea for my thesis. I've been wanting to do something regarding dialects and divergence and evolution in the creation of dialects. I'm also interested in why phonological differences appear between various English dialects (eg. pronunciation of "I" in American and Australian Englishes), while they don't really appear in Japanese dialects (eg. something written わたし is pronounced "Wah-tah-shee", with little/no change in vowel or consonant pronunciation in any dialect of Japanese). In Mufwene's book, he explores the influence of the lexifier (base language of a creole) on the structure and vocabulary of a developing creole. However, he doesn't attempt to explain the effect of the lexifier's phonology on the resulting creole's phonology. With English having been influenced by so many languages throughout its long history, its possible that phonological differences in English dialects are the result of extended contact with foreign languages and foreign speakers of English. Japan, on the other hand, has had a famously isolationist history and foreign language influence had been largely non-existant save early Chinese influence and recent English influence.

An interesting thing to note is the current influence of English on Japanese phonology (eg ヴェ and ティ sounds etc. becoming more and more common). These kinds of additions to Japanese phonology seem to be nationwide rather than isolated to certain dialects, maybe due to widespread dissemination of English through TV and music rather than concentrated English influence in distinct regions. Another point to look into would be if Japanese people tend to use the "non-Japanese" constructions like ティ and ヴェ when using 標準語, but tend to use more classically Japanese approximations like チ and バ when using loanwords in dialects. Use corpuses! That Japanese corpus that's in the process of being completed now could definitely be useful. Maybe a corpus of 熊本弁 would also not be out of the question. I'd need to have a sample size large enough for a study to be statistically relevant, and, from casual toying around with English corpuses, words occur at such low frequencies that an immense corpus would be necessary to make any sort of statistically significant conclusion about differences in pronunciation between dialects and 標準語.

Lots of ideas. Doesn't relate directly with evolution, but I really shouldn't attempt to push evolution as a reason for studying. It will be useful regardless of what I decide to research. While I pulled out the 進化学 a lot when trying to get accepted to school, it's really only peripherally related to what I intend to study. We'll see what happens.