My Chinese literacy is limited to a few hundred characters; nowhere near fluent. While being able to read it is not necessary to the project I had in mind, it does seem pretentious to be leveling a formal critique of Chinese literacy reform programs without that ability, or the ability to directly access much of the relevant published literature. That, and the fact that many of the experimental designs I propose require access to equipment I don't have at my disposal, are weakening my resolve to follow that particular thesis path.
I should've prefaced that by saying that my undergrad degrees were in psych and biology, and I've been playing catch-up in theoretical linguistics throughout graduate school. Needless to say, it's given my approach to linguistics a very quantitative/psychobiological slant.
"Scientific Babelism" is a term I've run across that describes something parallel to "Scientific Creationism" in biology -- the idea that language diversity is not the result of historical evolution from older to newer languages, but rather of "divine creation" at the time of the Tower of Babel. Here again, I lean on my undergrad background. I'm putting together a webpage about it, and I'd welcome your input:
http://linguisticspage.homestead.com/skeptic.html
I also have an abstract for a conference paper proposal that I'd be glad to share with you if you're interested in this sort of thing. The down side is that I don't know how I could stretch it to thesis length.
Lastly, I learned a little about 'forensic linguistics' this summer, which has to do with using text and style analysis to determine characteristics about the writer, including truthfulness. The name 'forensic' obviously refers to the use of this in a legal setting.
I thought, "Hey, wouldn't it be fun to see if these techniques could tell who's describing themselves honestly in personal ads?" I got this idea after watching the show "Blind Date" continuously for about two months.
I really think forensic linguistics would make a cool career, though academia has always been where I see myself. The problem with this route, thesis-wise, is that I don't have any background in forensic linguistics, and neither does anyone in our department.
* * * * * * * * *
Anyhoo, that's my thesis conundrum in a nutshell. I have ideas, but not the strength in any one of them to decide convincingly which to pursue.
Well that's a weight off 
(at the moment if I were to be involved in a khukuri duel I think I would bet on my being hacked down 
