So the thesis process is progressing, and I have finally chosen a direction in which I am proceeding. I think I'm going to analyze the change of Japanese through the influence of English/foreign language and attempt to track the changes regionally to see if certain areas of Japan were earlier adopters of various foreign words and/or recently adopted non-traditional Japanese approximations of foreign phonemes. I was hoping to research regional newspapers and record the ratio of occurrences between the older approximations like フイルム、チーム、バイオリン vs フィルム、ティーム、ヴァイオリン、etc. I wanted to do a chronological and regional analysis of various newspapers around Japan, ie examine at least 3 newspapers - probably a Kyushu newspaper, a Tokyo newspaper, and maybe a Kansai newspaper - and look at the foreign words in use in each paper at different times, eg Meiji, pre-WWII, post-WWII, and current newspapers.
While post-WWII newspapers are readily available in digitized format, pre-WWII newspapers are considerably, if not entirely not at all, available in searchable, analyzable, digitized format. For example, the Yomiuri Shimbun has fully digitized articles from around 1978, and while it has scanned papers dating back to early Meiji, the full texts do not form an analyzable database and are of little more use than microfilm. The Kumamoto Nichinichi Shimbun similarly has digitized versions available from the 80's, but anything earlier is scanned and, along with being difficult to read, does not constitute a searchable database. I'm expecting the biggest change in Japanese newspapers to be between pre- and post-WWII examples, yet without access to large amounts of prewar data, I can't make any statistically significant arguments showing change over time or distance in Japanese newspapers.
A possible reason for the lack of digitization of old newspapers might be due to the near-complete lack of a standardized writing/printing system in prewar Japanese. Old newspapers are presented in many different orthographical formats, examples being kanji/hiragana hybrids, kanji/katakana, hiragana only, katakana only, and a fairly even spread between with- and without-furigana typesets. The difficulties this would create for successful OCR are understandable but regrettable. I'll have to find another way to chronologically and regionally analyze unique katakana occurrences.
One thing I'd like to look into a bit more is the シェ・クァ pronunciations in Kumamoto and see if they influence the pronunciations of loanwords currently being used by Kumamoto natives that say シェンシェイ （先生）and うんどうくぁい（運動会）. Do they say things like シェーター and アメリクァ？
Currently, I'm reading up on the history of katakana usage and when, how and why it became relegated to 外来語. I'm also reading Loveday's Language Contact in Japan to see what has influenced Japanese in the past and if they're similar to the changes Japanese is currently undergoing. Hopefully, this book will provide something of a bridge between the katakana ideas I'm looking at and some of the evolution stuff by Mufwene that I'd really like to investigate more.
I still really want to look at the phonological changes between dialects of Japan and English, but realize that it's too big and involved of a project for a master's thesis, but would potentially like to research more should I decide to aim for a doctorate in the future. I realize I'm getting ahead of myself a bit, but I have a strange attraction to dialects and don't want to abandon that field of study.
The End for now.
