a wolf angel is not a good angel

The good thing about all this snow (though I think it just stopped snowing) is that my cats stop everything and sleep. More, even, than they usually do.
Since I am at home today, they are happy to sleep on my bed while I “work”.
I’m trying to figure out a formal way of explaining how modals pick out just the possible worlds that are made relevant in intensional verbs, so that they then can give information about instantiations of kinds in those possible worlds. I can sort of see my way through, but I have trouble explaining why the following is infelicitous:

1. There’s a book about giraffes I’m looking for. It must have pictures in it.

Part of the problem is the other (irrelevant) sense of must. The one I want is the one that’s okay in

2. I’m looking for any book about giraffes. It must have pictures in it.

There’s something about (1) that feels uninformative.
In (2), it might or might not be the case that this book you’re looking for exists. A book about giraffes with pictures of course, does, but if the second half said “It must have been illuminated by a monk in the 14th century”, it’s quite likely that it doesn’t. Even if it does exist, you’re probably never going to get it, but that’s not relevant.
In some worlds — not this one — there is indeed a book about giraffes that some monk painstakingly toiled over 600 years ago. In some of those worlds the search is successful. It’s those worlds that the must is referring to — the worlds were I find what I’m looking for. It gives us more information about the result of this search in those worlds.
If we know it’s a book about giraffes, that’s fine. There are lots of those. We then say more and more about this book, which doesn’t, perhaps, exist.
But in (1), we know the book exists. So going into the success worlds to tell us more about it . . . well, why is it wrong? I think it’s because this world might not bea success world (it’s at the Vatican and only the pope may read it; I am not the pope). So you’re giving information about success worlds, but not about the real actual book in this worldyou’re looking for.
I still think I’m missing something.

Today I am working at home. We’ll see how that goes. I seem to spend a lot more time procrastinating, but it’s hard to be sure, since I always spend a fair amount of time procrastinating. In any case, I have 9 pages and need to add, probably, 3, though if I can’t do it I will survive. It feels like I’m cheating, because it’s 1 1/2 spaced and I have lots of room taken up by formulas and updates and everything. However, these are the main point of my proposal, so I guess it’s not really.
I don’t know. I think the problem is that I want it to be really good, and it’s just not, and I don’t know how to make it better.

Yes, another day spent n the coffee shop. But I bought a sandwich, so I don’t feel all that guilty. Of course I’ve still consumed countless cups of coffee, so maybe I should.
It’s 6:30; I want to leave in just over an hour. I’ve written about 5 pages of my semantics paper, draft quality. I somehow suspect that my goal (entire draft, which will need fleshing out) for tonight is unreachable. I write something, I delete it, I rewrite the same thing, I stare at it, I play solitaire, I delete what I wrote. Repeat.
I don’t know *how* to write it, really. I’ve written out most of my major definitions. I say “my”, but of course I barely came up with any of them, and what I did come up with was changed for me. Oh well. First real major research paper in semantics. If I repeat that enough, maybe I’ll feel less guilty. I’ve worked on this one way more, but it’s just hard to bullshit this. Syntax, not so hard. Phonology, almost easier (though minimalism is ensuring it’s a close race). Snark, snark, snark.
I’m pretty much at the end of the “I’m looking for a book about giraffes” sentences. Then I need to get on to the “It must have/has pictures in it” continuations. It should work: the verb sets up possible worlds, the modal should make topical the success worlds, so if the modal’s not there we’re referring to reality, the only world we have available, but if it is there, we instead refer to the worlds in which the search is successful.
I write it here to ensure it’s cleaner in my mind. It’s been explained to me, but I need to be able to explain it or else it’s sure I don’t know it.
It’s sure? Whatever.
A friend and I were discussing genius boy. All three of us want to do semantics. And he mentioned that GB is going to make our degrees look a lot better. I’m glad he’s ahead of me. (a) I need the help now and (b) his successes will make this program look even better for semantics.
I try to recall we’re a quite new graduate program, so it’s all of us now setting the reputation.

Among the students in my department. I have not decided what to make of it.
1.Mushroom
1.Pineapple (tie)
3.Mushroom & pineapple

Finished everything but the references, which need to wait until I am home and can find them. My theory can also explain pair-list answers in English. Yay! And thank you, advisor, for suggesting I look at English, too. Home, home. Finished a paper by 7 the night before? This is a record.

I’ve now finished everything but the conclusion and the trees. (And the references, but I’ll spend half an hour on that tomorrow. I know them all, I just don’t have the details handy. Anyone want to give me EndNote as a gift?)
I need to do some editing of the first four sections, then I’ll write the conclusion. I think I’m on a bit of shaky ground for one part (or wrong, whatever), so I need to figure out how to explain that better. But it’s only 5: this means I won’t be up until 3 am writing the paper. I’ll stay here until 7, when traffic gets lighter. By then I should have finished the writing and some of the trees. I’ll finish the rest of the trees before ER (that show needs to go off the air so I can stop watching it), watch TV and go to sleep.
Then I get to start writing semantics. Sigh.

Sitting in the cafe again (they’re currently out of hazelnut coffee, dammit), finishing syntax. Well, sort of. I’m writing, skipping the examples I don’t have handy and the trees. I will fill those in by the end of tonight, which leaves just formatting and putting in the full references for tomorrow. This is my plan.
I’m about halfway through the final section (except for the intro and conclusion). I’ve explained how I’m modifying other people’s theories. Now I’m showing how it plays out in the simplest possible sentence. Then I’ll move up to a slightly more complicated one, and then to my last, most complicated one and show that I can explain all the weird little details of this language.
I will need to add more guideposts, though: right now I think I have too much of the mystery novel structure, which is never good. I hate reading those: tell me immediately what your solution is going to be! But it’s so satisfying to write. Still, it will need editing. Most of it can come in the introduction, so it’s not so bad. I have a lot of trouble fixing this kind of structure.
I used to write other things — fiction, poetry. I guess it’s turned into just academic writing. I sometimes regret it, but I can never think of anything else to write, anymore.

Yes, my syntax paper still. Blogging while offline: how fun! But no real blogging once I get home — it’s a Law & Order double header tonight!
I know I’m tired, I’m using a zillion exclamation marks (!).
I’ve decided to use phases (possibly unnecessary, but I don’t think it will turn out that way) and a single output syntax (this part might also be unnecessary, now that I think about it. No. If I think a bit more it’s probably crucial. Oh well. Caffeine ho!). At this point, other people who know enough about minimalism to give them nightmares (e.g., anything at all) tell me that those two contradict each other. Well, granted. As they stand. But who says I can’t change them? They’ve probably changed since yesterday, anyways, much less the two or so years ago when the articles I’m primarily using (Chomsky’s Derivation by Phase and Bobaljik’s A-Chains at the Interface) were written. Minimalism sometimes changes by accident when you sneeze.
Chomsky has certainly brought to a high and wonderful level the ability to write totally incomprehensible stuff that you look at and figure, well, hey, it can’t be that stupid, it must be just me.[1] Luckily I’ve heard a lot of talks about or using phases (mostly by the lovely Norvin Richards), so I have a vague idea of how they work, which seems to be something like: send your stuff to LF a bunch of times, and to PF following each LF[2].
I don’t like this. We might interpret a bunch of times (this seems most clear in biclausal sentences, where it *feels* like we might interpret the Matilda thinks part separately from the that Sam is fluffy part). But we only really say things once.
So that’s essentially what I’m saying. I don’t actually know why we need phases (and reading Chomsky’s paper is not helping), but they’re helpful to me. (I’m becoming just the linguist I don’t want to be. I console myself by remembering that I am at least getting real data and drawing what I think are accurate generalisations, and those should stand the test of theory.)

[1] It would make sense if this were what spurred me on to defecting from syntax. (Although now that I’m doing syntax again I remember it’s fun, and will keep the interface in mind.) But it’s not: I generally understand things I read for syntax. I almost never understand whatever I try to read in semantics.
[2] LF: it’s supposed to make us think of but not stand for logical form; it’s more or less what we interpret the meaning of. PF is what we pronounce or hear (think something like phonological form, but not quite, because we can’t have clear terminology: subject and SUBJECT, case and Case, pro and PRO[3]).
[3] When I took my first computer science course in university we used C++ (I think) — something case sensitive, anyhow. We had 6 homework assignments, and the best 5 counted. After the first 5 I had all 9s and 10s, so I didn’t really care very much. But on one of them (in part, make a recursive program to determine the sum of the first n natural numbers), I called a variable n and lost marks for not having the variable name be at least 3 characters. This struck me as just plain stupid — along with the idea of needing to do recursively what was much faster to do with a non-recursive formula — so I decided for my last program to be evil. I named every single variable ‘bagel’, modulo different capitalisation on some letters. I’m not sure the TAs even ever read my source code (the program worked, though).

I’m sitting in a coffee shop (free refills on coffee! Hazelnut, too!) trying to write a paper. Unfortunately, someone else here works here but came to work an hour and a half early and seems to have nothing else to do than talk onto her cell phone very, very loudly. She has a headache; she’s tired; she has three papers due tomorrow and an exam Friday and some other stuff due next week, but I’m trying to ignore that. I also know what she’s eating, what toppings she prefers on Whoppers, and what toppings her *friend* had on a Whopper, once.
I feel sympathy for her having gotten to work early: I’ve done things like that, especially when stressed and tired, and it just makes everything miserable. Hell, I even understand why she’s been on her cell phone (though, dear girl, there’s a big bookstore across the street!). But can’t she just have these conversations — all inane and repetitive — a bit more quietly?
That conversation at Rana’s a while ago comes to mind. I can work with background noise — need to, really. Someone talking to me, however, is too distracting for words.
Also: why has everyone *else* been given a(nother) extension on their syntax paper? Though it’s better if I have mine done by Friday, still.

I am impressed with myself, sort of. I have written four full pages of my syntax paper (including most of my examples, but not all, since they weren’t all convenient). I haven’t written an introduction, I’ve written my “introduction to the syntax of this “weird”[1] language” part, my “this is what someone else argued about quantifiers and reconstruction” part and most of my “how can we explain those crazy questions (independently of quantifiers)” part. Next comes the “hey, the two explanations are totally mutually inconsistent” part. (I will get to at least here by the time I go to bed tonight — actually, by the time I leave the coffee shop, but hopefully further.) Then I get to my “look, my cool arguments can explain why what seem to be contradictory facts are actually totally consistent” part. Then it’s just mopping up: intro, conclusion, why what I’m writing is important and where it fits into the field. Yes, this looks like it will get done in time.
[1] Read, not Indo-European, Chinese, Japanese or, possibly, Hebrew.