January 4, 2006

On maintaining a blog

Writing is like taking a shit. It is at its most satisfying when it is something that has to come out, and the greatest pleasure it offers authors is the relief of release. I guess I've been pretty constipated lately.

I'd like to say it has to do with the events of this last summer, but I don't think this is really true. I could be wrong. One's own motivations are never fully accessible. Obviously, it's been hard. We haven't had any success lately in getting pregnant again, but it takes a while for regular ovulation to return after a pregnancy, even a terminated one, and it takes long when you're older than when you're younger.

The grand irony of it all is that I don't actually have any heavy responsibilities for the first time in as long as I can remember, and trying to get pregnant has involved having - well, I suppose this is too much information but what the hell - more and vastly better sex than I've ever had in my life. My Dutch is good enough that I can actually impress people as a rare anglo who speaks the language. I think my Chinese is improving, although it's hard to tell. My wife is actually learning French - she's in Paris right now - and has enough French that I can start to rent films in the language. My presentation in Amsterdam went well, and I think I actually managed to impress the people I wanted to impress. Perversely, things are going really well in every way except one.

Even the political scene is improving. Everyone hates Sarkozy and Villepin and generally thinks Chirac is a joke. Bush appears to be in real trouble on enough fronts that it may yet destroy him. The European economy seems to be on a real rebound.

And yet... And yet I can't write. It's not so much that I'm horrified at the thought, it's that, I just can't seem to do it.

I have enough that I want to say. I've been reading Derrida lately. Initially, it was a matter of sheer curiosity - anybody that widely hated can't just be spouting incomprehensible mush. And I was right. Derrida is not spouting nonesense. I haven't quite decided to what degree I agree with his main thesis, but he has one, and it is quite a doozy, and it most certainly has linguistic significance. I'm really begining to like the guy's work. He actually stood up for dissidents in the old Warsaw Pact and never had any truck with the kind of lightly shrouded ethnocentrism that passes as realism on the left in more recent years.

I think part of the problem is his translations into English. When I started reading On Grammatology, I took a copy of Spivak's translation out of the library with it in case Derrida really proved unreadable. Then I returned it. Derrida is not an easy read by any means. He dances around the subject, conducting detailed arguments with long dead thinkers of various stripes, but he's not that hard to understand when you are used to, first, conventional French writing styles that tend to long sentences of the kind anglophones would consider run-ons, second, the numbered and bullet-pointed style of French intellectual writing that he is clearly rebelling against, and lastly, have some capacity to recognise where he is echoing the authors he discusses. Certainly, his arguments in his first chapter are fairly easy to understand as an extension of Saussure's. If you read him in French, this is all clear enough. In English, I just found him too jarring to follow.

There's other stuff. I'm finding information theory cropping up in such odd corners of linguistics, often disconnected from more advanced ideas in information theory that could in some places really be useful. Like, I just came across Martinet's functional load hypothesis for language change. It seemed to me that his lack of understanding of information theory (which is hardly surprising considering that he proposed his hypothesis in the 1930s) explains many of the short-comings that led to the rejection of his hypothesis. Others have taken up that line of thought, but there's an important component missing from their approaches: the notion that the structural features of a language serve to reduce information loss due to noise. I think a failure to think of language as something that operates in a noisy environment underlies a lot of what's wrong with McWhorter's approach to grammatical complexity in Defining Creoles, for example. In fact, it was in reading his book that it occured to me to look into the idea. For example, grammatical gender appears superfluous to a lot of people. But, studying Dutch, one of the favourite kinds of questions is fill in the blank sentences. For example:

Hij leest de _____.
He reads the _____.

There are a number of coherent possible answers, but the obvious one in English - book - is ruled out by the article de. Boek has the neuter gender, and cannot have de as its article. Krant - newspaper - can go here nicely though.

McWhorter suggests that features like grammatical gender - among other things - are a sort of cruft that a language accumulates over time - that there is a kind of universal basic language structure that doesn't have those things, and that creoles all, to some extent, resemble this basic grammar. It is a kind of resurection of Bickerton's universal language bioprogram. Now, this is simply wrong if one considers languages like Michif creoles - the position that McWhorter takes in his book. Michif has absorbed all the complexities of French nouns and of Cree verbs at the same time, making it by his standards an incredibly complex language. There are ways to get out of this hole, but they all involve seeing Michif as something fundamentally different from a creole. As far as I can tell, Mufwene's approach still seems to explain the data better.

But the thought that occured to me was that complex and seemingly unnecessary features like grammatical gender can serve to reduce noise in oral communications. Languages that mark grammatical gender in articles, like Dutch or French, sharply reduce the number of words that can follow them given their context, while the amount of information that has to be stored in the lexicon to make this gain in noise reduction is very small. In French with just two genders, it's just one bit of information per noun. I suspect that by losing grammatical gender, Haitian Creole has had to create other distinctions to compensate. The thing is, given a reasonable corpus, this hypothesis should be empirically confirmable.

Presumably other grammatical features could be evaluated in the same way. Furthermore, instead of simply offering up a more sophisticated version of Martinet's hypothesis, it seems to me one could recast his core claim: Feature loss in languages should be less likely to occur to the features that most add to noise robustness. This hypothesis can cover a lot of ground - it can explain creolisation not as incomplete acquisition of substrates, but as the result of speakers with different language abilities - some fluent, some marginal - devising mechanisms to increase the noise robustness of their language.

I haven't found anybody going there yet. It offers prospects for empirical linguistics that I think are interesting.

Wow. I think that's the most I've been able to write in months. Unfortunately, it's dull as all hell for most people. Okay, obviously there's been a lot of linguistics on my mind lately.

Still, I wanted to write something. To wish my readers, however many I still have, a happy New Year, and to hope for a grammatological laxitive.

Posted 2006/01/04 22:55 (Wed) | TrackBack

No, no, not dull at all. It's an area which has caught my casual attention in the past, and I'm gratified to hear the thoughts of someone with the appropriate background. I shall have to look into Martinet. Your point about noise-robustness makes a lot of sense - I hadn't thought of it in quite those terms before.

Anyway, it's good to hear something from you again.

Posted by: Tim May at January 5, 2006 1:11

Happy New Year!

By the evidence on this page, you can still write.

Years ago I read somewhere a brief discussion of grammatical features as error-correction mechanisms, pointing out that not all potential errors are equally likely. For example, position-based grammars like English lack the protection against word-reordering errors provided by inflected (is that the right term?) grammars like latin, but word-reordering errors are pretty rare in speech (though maybe a little more common in written language?).

Posted by: Jeremy Leader at January 6, 2006 0:26

Not dull. Very thought-provoking. And I followed the link to the abortion post, which would have been the most menmorable thing on the web last year, had I read it then.

Posted by: Andrew Brown at January 6, 2006 14:29

Not dull to me either (obviously), and it's very good to see you blogging again, even if with difficulty. I'll have to check in more often than every few months now...

(I did read the post then, and it's haunted me since. I'm glad you guys are doing so well. You have my warmest thoughts.)

Posted by: language hat at January 14, 2006 20:25

Which Dialect are you learning in Chinese?

Posted by: jVirus at January 16, 2006 17:28

Thanks everybody - sorry I didn't respond sooner. I couldn't see that there were any comments here, as explained in the post at the top of the main page.

Posted by: Scott Martens at January 17, 2006 9:07

Even after taking a twenty-year leave of absence from writing, you could still be coming out with something that is broadly appreciated - like, say, Tomaso di Lampedusa who only started writing late in life.
I see genuine potential in your hypothesis about feature loss and feature persistence as functions of "noise robustness".

Posted by: Joerg Wenck at January 19, 2006 7:52
Post a comment

Remember personal info?