January 14, 2007

The Eleatic Analysis Debate

Kieran Healy points to, and reproduces in full, a paper by G.E.M. Anscombe that he claims may have been the shortest Analysis paper ever. I'm not sure; I think there's a non-constructive proof of the infinity of Analysis papers, namely that for any paper there's a shorter one.

And I think Anscombe's paper may have been part of the Eleatic Analysis debate, which proceeds as follows: The last 8 pages of a given issue of Analysis are given to a debate on the doctrine of double effect [or something]. The first 4 pages consist of a paper arguing that there is no such thing; the next 2 pages offer a convincing refutation of the first paper; the next 1 pages offer a convincing counterargument; the next 1/2 page offers a convincing counterargument to that; etc. When you finish the issue, what should you believe?

[NB I should probably mention that this is just the lamp puzzle by I think Judith Jarvis Thomson, rephrased as a joke about Analysis.]

Posted by Matt Weiner at January 14, 2007 08:03 AM

After reading the issue, you should decide that the doctrine of double effect is certainly not so important as to merit infinitely many articles about it. As such, you suspend judgment and refuse to think about it ever again.

Posted by: P.D. at January 14, 2007 08:23 AM

That the people who typeset Analysis are very skilled indeed.

Posted by: ben wolfson at January 14, 2007 11:35 AM

Herr Weiner,

We're having a Texas Mini-Meetup this coming weekend (Saturday the 20th) in Austin, just on the off chance that you'll be in Austin for some other reason tjem or feel like making the long drive down (and we can arrange housing/couching if that would help seal the deal).

In attendance will be me, Sir Kraab, heebie-geebie, soubrizquet, Neal the Ethical Werewolf, and maybe even Michael, driving in from Lousiana. On the agenda is dinner then drinks afterwards at the Carousel Lounge.

Just wanted to let you know, it would be cool to finally meet you, but I know it's shortish notice and a longass drive, so we'll understand if you, like, totally snub us.

Posted by: M/tch M/lls at January 14, 2007 05:35 PM

I think it was Douglas Hofstadter (!) who pointed out that the physical form of a book makes it difficult to sustain perfect suspense about the plot; when you get to within a few pages of the end, it's quite obvious the end is near. This could be alleviated by adding some arbitrary amount of paper at the end. But of course if the pages were blank, as you idly flipped through the book you would figure out approximately where the end was, subverting their purpose. So the pages would have to be filled with type. But if it was just random gibberish, or interpolations from Borges or Jane Eyre, you'd realize that too without intending to, as even an unintentional glance at a page reveals enough of the wording to tell two novels apart. So the added pages would have to have no visible difference to the actual novel. But then as you were reading, how would you tell when the actual novel ended? Perhaps this post ended some time ago and you're reading the tacked-on nonsense.

Quite possibly, the editorial board of Analysis has taken this to heart, and the last N and 1/M pages are simply added on to prevent facile convergences.

Posted by: Ben at January 15, 2007 12:46 AM

Douglas Hofstadter and I, in an essay I wrote for my 12th-grade English class on The French Lieutenant's Woman, thankyouverymuch. It was about multiple endings or some such crap.

Another way to do it would be to write your book in continuation passing–style: at the end of each page, a note saying which is the next page, so that the progression of the book would come apart from the progression of physically consecutive pages. If the book were of a decent length, this might prevent one from being able to tell based on purely physical characteristics how close one was to the end (you'd be able to tell that you've seen some facing pages before, or the like, but it would hard to keep everything straight enough to have the sort of immediate indication the diminution of thickness on the right-hand side offers will-you or nill-you). On the other hand, it would be annoying to read.

Posted by: ben wolfson at January 15, 2007 01:16 AM

I have used "one" and impersonal "you" in the same sentence! Horrors!

Posted by: ben wolfson at January 15, 2007 01:21 AM

Obligatory link 1.

Obligatory link 2.

Posted by: Matt Weiner at January 15, 2007 08:43 PM
Another way to do it would be to write your book in continuation passing–style: at the end of each page, a note saying which is the next page, so that the progression of the book would come apart from the progression of physically consecutive pages.

You know what would be a really cool idea? You could make the book title reflect the physical act necessary to read it. It's like a game of Hopscotch , you might say.

I'm not going to ever be able to find where Dougie Hofstadter said that about endings (I don't even have a copy of his doorstop of a book now), so you'll have to either believe that he said it, that I fictionalized it like Marx's fictional borrowing from Hegel (the "first as tragedy, second as farce" quote), or most likely, that I have access to and am reading from your 12th grade English assignments.

Posted by: Ben at January 16, 2007 12:06 AM

I don't have a copy of the doorstop either (or if I do it's in my office, where I am not), but the dialogue that ends with the Achilles being arrested by coppers named Silva and Gould -- having been framed by the Tortoise for stealing a gold box -- is relevant here. On further research, it's called "Aria with Diverse Variations."

Posted by: Matt Weiner at January 16, 2007 07:36 AM

Another way to do it would be to refuse to bind your book, so that it resisted any attempt to place it in order and the only way the reader could divine the approaching of your ending would be to set the signatures to one side as they were completed. If the reader eschewed this cheap shortcut and returned the signatures to the original stack upon reading, she would be sampling-with-replacement from the stack. I have a proof of the number of pages it would take for the reader to be statistically confident she had completed the book, which is unfortunately too small to be written in this margin.

Then much later some indie band from Dorchester could write a painfully catchy song about your disorder and early ending.

Assuming that it was the real ending, that is.

Posted by: Ben at January 16, 2007 07:12 PM