May. 11th, 2002

maradydd: (Default)
<whine>

My friends-of list has been dwindling for no reason I can identify. Evidently all the people who are jumping ship are people I hadn't reciprocally friended, as it doesn't look like anyone on my friends list has suddenly removed me from theirs. So it's not even remotely important -- evidently I haven't offended anyone I actually know and/or like -- but it's still peculiar. Have I just been too long-winded lately? Yeesh.

</whine>

In other news: it's been an unremarkable couple of days. Gave a talk on Monday, another on Thursday, both went acceptably well. Received 7 pages of data for my syntax final; tried analysing it in class while someone else was giving a talk and didn't make much headway, then sat down with it this evening and basically had the whole thing figured out within the first two pages. ("Ah. It's doing this. I bet I could generalise thus-and-such. Let's see if I can." <flip, flip, flip> "Aha. And here it does that. Yep, that's a generalisation.") It's always nice when the data behaves.

I also made massive headway on the piece which I'm going to turn in for my fiction portfolio, to which I will refer as the Unfortunately Titled Story, as its title character randomly shares his first name with one of the characters (and I use that term loosely) whom my Faithful Readers have met through my journal. The name is a complete meaningless coincidence, as I'd come up with this story idea (and named all the characters) sometime, oh, around last March -- and I'm not really inclined to change them just because. But still. I'd hate to think people were drawing conclusions about people I know just because of an Unfortunate Title of a story.

Anyway, it's 3:30. I'm for bed.
maradydd: (Default)
Those of y'all whom I haven't met in person/stumbled on through someone else's LJ and friended you first: where the hell did you find me, anyway?

Hey, I'm a sociolinguistics teacher. I like to play follow-the-bouncing-meme.
maradydd: (Default)
This morning (ok, afternoon) while in the shower, I started teasing with the natural-language problem of irreciprocal antonyms. I think it started with that old Objectivist saw, "Pleasure is more than merely the absence of pain." One would easily formalise the oppositeness of the two ideas (if one has pleasure, then one is not in pain) as (using > to mean "implies"; damn HTML and its lack of useful high-ASCII symbols) P=pleasure; Q=pain; P > ~Q. But, of course, ~Q does not necessarily imply P -- the invalid converse (which is what Rand was saying). Nor does ~P necessarily imply Q -- the invalid inverse (you might not have pleasure, but you might just be bored rather than in pain). For any other restatement of the principle to come up true, you need the contrapositive: Q > ~P.

This is all a very simple application of modus tollens, and I'm sure any real logician would have seen this far faster than the couple of minutes it took me while I was washing my hair. And, of course, it applies to more complicated constructions: if Meredith and Colin slept together, then Meredith didn't sleep alone and Colin didn't sleep alone; but if Meredith and Colin didn't sleep alone, that doesn't necessarily mean they slept together!

Yet in natural language, the tendency is for people to assume the inverse to be true, especially if it's phrased like that. (This is my intuition, and native-speaker judgments are welcome, but I think "Meredith didn't sleep alone last night and Colin didn't sleep alone last night" has a reading closer to the strict logical interpretation, particularly if your world-knowledge indicates that Meredith and Colin normally sleep together -- perhaps the explicit construction is intended to reject the hearer's implicit knowledge? -- though "Meredith didn't sleep alone last night and neither did Colin" suggests to me that the speaker is being coy in talking about the wild night Meredith and Colin had!) I'm not sure why that is.

NL is so weird and goofy.

I'm also now wondering, at the back of my mind, what opposites we might have (if any) which do have valid inverses or converses. Someone's probably done work on this already. Note to self: Figure out how to look up crap like this. My normally undefeatable research skills already feel taxed by this one.

Profile

maradydd: (Default)
maradydd

September 2010

S M T W T F S
   1234
567891011
12131415 161718
19202122232425
26 27282930  

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags