Text Patterns - by Alan Jacobs

Monday, August 24, 2009

building the perfect language. or not.

I enjoy most of the books I read — I admire many of them — I adore some of them — but it is not often that I think “I have no words to express how desperately I wish I had written that.” But just such a longing consumed me as I read Arika Okrent’s scholarly yet delightful In the Land of Invented Languages: Esperanto Rock Stars, Klingon Poets, Loglan Lovers, and the Mad Dreamers Who Tried to Build A Perfect Language. You must buy it, or at least borrow it from your library.

However, you should not get the Kindle version: the book has a number of fascinating charts that are reproduced on the Kindle screen at a size too small and resolution to low to be read. Publishers need to be more responsible about this kind of thing, and Amazon too: if images can't be seen clearly, they should be left out of the Kindle editions, and perhaps made available on a website. Very annoying.

Speaking of websites, the one for the book features a shockingly long list of invented languages, from Lingua Ignota to Proto-Central Mountain, most of which are accompanied by brief translation samples and random tidbits of endlessly fascinating trivia. Consider yourself warned.


  • Ari Schulman said...

    I have a friend who is convinced that "natural language in inherently ambiguous" or something, and that therefore we should switch to an unambiguous invented language. Does this book contain anything in the way of refuting such a notion?

  • Oh yes. Okrent is great on why such ideas are superficially attractive and why they're ultimately impossible to carry out. Consider this:

    "To all the language curmudgeons out there who insist that people ought to speak more logically, I say, be careful what you wish for. You go on about the 'logical' mismatch between 'everyone' and 'their' in perfectly normal sounding sentences like 'Everyone clapped their hands.' You argue that phrases like 'very unique' and 'sufficiently enough' don't make logical sense. You harp on 'hopefully' and 'literally' and 'the reason is because,' all the while calling logic to your side to defend your righteous anger. Before you judge me as some kind of 'anything goes' language heathen, let me just say that I'm not against usage standards. I don't violate them when I want to sound like an educated person, for the same reason I don't wear a bikini to a funeral when I want to look like a respectful person. There are social conventions for the way we do lots of things, and it is to everyone's benefit to be familiar with them. But logic ain't got nothin' to do with it.

    "And oh, how grateful I am. Do you know how good we have it, how much easier our speaking lives are made by the fact that language and logic part ways? Consider the word 'and.' Why, you barely have to know what you mean when you say it! When you say you 'like ham and eggs' do you have to specify whether you like each of those things as evaluated on its own merits separately or whether you like them served together as an entrée? No. You just lazily throw out your 'and' and let context do the rest of the work for you. When you say you 'woke up and ate breakfast' do you mean that you woke up first and then ate breakfast? Or did you do the two things simultaneously? Or, maybe your breakfast was asleep, so you woke it up and then ate it. Pshaw, you say. You know what I mean. Perhaps I do, says the Lojbanist. Perhaps I don't."

    And then she goes on to show how a completely unambiguous, logically consistent language — which Lobjan strives to be — has to have so many words for "and" that it becomes absolutely unusable. We've been going around this block for several hundred years now, but no one learns the lessons of the past failures.

  • Ari Schulman said...

    Excellent, thanks. I expect my friend will spin some yarn about how he doesn't deny it would be practically difficult, but in theory it would still be possible and preferable if people just put in the effort. The point I want to get across to him -- and I'm not sure exactly how to argue this -- is that the task is impossible in theory. Maybe he is just talking about syntactic ambiguity, but I don't think you can fully disambiguate syntax without disambiguating semantics. And meanings change as culture and understanding changes.

    In order to disambiguate language, I think, you would have to disambiguate all knowledge. The task is impossible because the project of grounding philosophy in pure reason has proven impossible. I am not certain of this and think I might be quickly argued into a corner without having better thought this out. But anyway, this book looks like it will be quite helpful in making some headway. Thanks again for the tip.

  • Okrent's book shows that absolutely unambiguous language is easy — all you have to do first is to have absolutely perfect conceptual clarity and universal agreement that you have done so. Piece of cake.

  • Yes, thank you. That's the point I was trying to make. I must get this book.

Post a Comment

[Basic HTML tags can be used in this comment field.]