Well...

It's been quite a while since I've posted anything here, and it might be a good while longer before I post again.

You see, I've graduated. I'm no longer a Linguistics Undergraduate, so I'm not exactly elligible to post here anymore. I could have pulled the rug out from under my collaborators and changed the name, but I don't think that would be fair. I strongly encourage the rest of you to keep posting here, and to get more people on board. This blog gets alot of google search traffic, and was even posted on the LSA 2007 message boards.

As for myself, I've set up a new blog called Val Systems. It doesn't have very much content yet, but that'll change soon, believe me.

Merry, Mary, Marry

So here are my spectrograms. I'm not entirely sure what they show, other than that I do pronounce these words differently, and I already knew that.

One thing of note (to me) is that my /r/ is longer in "Mary" than it is in the other two. That's something I hadn't realised before.

I'm not sure yet how to upload the actual recordings. I need to find somewhere to host them I think, then link to them from here. I'll edit this post to include them once I've figured it out.

Session resumes

Just a quick post to get back into the habit.

Autumn session has started, and having won the first year linguistics prize last year, the pressure is on to keep up my performance. This session sees me doing Phonetics, next session I'll be doing Formal Phonology and Language Description.

I'm finding phonetics more and more absorbing (but isn't that the way of it, the further you look into any branch of linguistics!). We're using Ladefoged, P, "Vowels and Consonants" as our text, which seems so far to have quite an acoustic focus. That suits me down to the ground!

One interesting thing so far is that BBC English (and by extension Australian English) has many more vowels than General American English. GAE: 14-15 vowels, BBCE: 20. And that doesn't include schwa.

I wonder why he doesn't count schwa as a vowel, though? And I wouldn't have thought that BBC English (not quite my dialect, but almost) had so many more vowels than General American English. Imagine not being able to distinguish between "Merry", "Mary" and "Marry"!

So, that's that.

Oh, I have a new job; I'll start as a Natural Language Programmer on 28/3. More updates when I figure out exactly what it is that I'll be doing.

"As elements become more abstract, they become more categorical."

Someone said this to me during a conversation today, and I realized it had to be true. He was specifically talking about language change. In Martha's Vineyard, the older generations centrilized the nuclei of (aw) and (ay) along a continuum of following environments, from voiceless to voiced, and also from anterior to posterior. The later generation developed a categorical rule, and only centrilized before voiceless consonants. I guess this means that the older generation had a close to surface mechanism of centilization, and the younger generation had a more abstract phonological rule.

When he said that abstract elements are more categorical, I knew it had to be true. I wish there was a whole class I could take where people would say things like that and I'd know it'd have to be true.

Back to School

Well, I'm one week back into school, and it seems like I've chosen a rather intimidating course load for my final semester. A professor last semester convinced me to register for the graduate Pragmatics course being taught by a visiting prof. I'm in it with all the same graduate students I've been in classes with before, but I'm still feeling somewhat nervous, and have yet to pipe up in class.

What's really got me sweating bullets is the class with Charles Yang that I've mentioned once or twice. He kicked off class with "This course has never been taught anywhere." There are some mathematical concepts I haven't touched since high school at least (like solving differential equations, sigma notation) and others I never touched ever (vector notation and matrices). I'm particularly excited about the opportunity to be creative in a more universal sense, but this is the closest I've come to being in over my heard in a long time.

That's about all for now. I'm in the midst of writing a naive review of Knowledge and Learning in Natural Language, but I don't think it'll get posted until a bit later.

Christmas Cheer Clarified

I was just so wowed by something so elegant and beautiful that I didn't want to ruin it with explanation in the last post. However, how many people know Philadelphian phonology well enough to make sense of it? The reason I deck the halls with bowels of holly is the result of a few on-going changes in progress interacting with eachother.

The story begins with /æ/. A well known attribute of the Mid-Atlantic and New York dialect regions is their split short-a, [æ], systems. There has been a phonemic split between lax æ and tense æ, or (æh). Lax æ is staying rather stationary, if not lowering and backing ever so slightly. Tense (æh) is raising and diphthongizing: æə->eə.

Enter (aw), which in conventional IPA is usually /aʊ/. In Philadelphia, among other places, the nucleus has fronted to [æʊ]. I don't know the chronology of all of these changes off the top of my head, but the nucleus is definitely associated with the tense short-a, so it is rising in suit. As the nucleus rises, the offglide is lowering, so now most Philadelphians say house something like [heɔs].

Now enter /l/. I think all dialects of English have dark, or velarized /l/'s syllable final, and in alot of them it's vocalizing to something like [ɤ]. Philadelphia is unique in that it's even vocalizing intervocalically and in initial clusters. The example I hear mentioned alot is that if you walk into a Philadelphian shoe store and ask for "New Bounce," you'll be serviced with a pair of New Balance sneakers without hesitation.

If you remember the IPA for non-English vowels, you'll know that [ɤ] is unrounded, and just a bit higher than [ɔ], which, if you've been paying attention, you'll recall is the new offglide for (aw). So what effect does all this vocalization have on preceding (aw)'s? The offglides delete, so for towel you go from [tæɔ.ɤ] to [tæɤ] . This means that there is now homophony between pal, and Powel, and this whole post has been about vals. Apparently this is tough for alot of people to believe, so let me tell you as a native speaker of the dialect, this is absolutely true.

Now enter the Christmas Cheer. While singing the carol "Deck the Halls" to myself the other day, I realized that I have always sung it "Deck the halls with [bæɤz] of holly." I guess it could be a nonsense word bals, but [bæɤz] exists in my vocabulary as bowels. Just to be clear in the progression: boughs [baʊz]->[bæʊz]->[bæɔz]. Now, depending on where in the progression from [ʊ]->[ɔ] someone is, the offglide could very well be very ɤ-like.

I guess that as I was growing up, the data was sufficiently garbled that I determined I was hear hearing a vocalized /l/ rather than a lowered (aw) offglide. It is sung, and it is a fairly low frequency word (I always take a [bæɔ]). Not to get too ahead of myself, but it may be an indicator of things to come. I already have a few tokens of intrusive /l/'s on tape, so we'll have to wait a few age cohorts to see how they sort the mess out.

I think my previous post was more eloquent.

In other Christmas news: I got Charles Yang's book Knowledge and Learning in Natural Language. It's pretty hefty, which makes me nervous about taking his class next semester. I got my parents his lay-people book The Infinite Gift, so maybe now they'll have a better understanding of what it is I do.

And also, go Eagles. I bet Noam Chomsky would be a fan of the birds if he didn't think professional sports wasn't some kind of institutional distraction for the working class.

Time for some Christmas Cheer

I just realized that I deck the halls with bowels of holly....