13 August 2004. Hmmm, Friday, the thirteenth, with a hurricane coming my way. But I feel lucky. Schools are closed so I don't have to sub today, and yesterday I was granted a one-week extension on the deadline for a review I'm writing for American Book Review. I might yet meet the original deadline, which is the 15th, but it's a relief to know I don't have to.

I was almost going to post a new version of the mathemaku below as my entry for today. So this entry is really about what it isn't about. This I think interesting as an example of the kind of dumb/serious subtle problems sharply self-analytical poets like me get tangled up by. What happened is that I suddenly worried about the reflections of pale foliage in the poem. Why didn't I have a mirror image of the text for it? These are reflections in puddles so should be mirrored. After thinking about it, I went to my image of the poem and flipped the text in question.

At that point, I was ready to put it here to show how the poem, already several times revised although not one of my most complicated works, was getting ever-better. Later, though, I realized that I was wrong. Here's the thing: the reflections would not be reversed, only what they were reflections of. That is, once I referred to the reflections as reflections, I was no longer speaking of "pale foliage" but of "reflections." The images in reflections are reversed, not the reflections. If that makes sense. It still doesn't, entirely, to me, I have to admit. That's because I don't feel I'm yet able to say exactly what is going on, and I'm tempermentally not capable of accepting anything one can't be clear about. I'm leaving my poem as it was, however. I want it to remain one of my most accessible mathemaku, so am happy to be able to persuade myself not to have any mirrored text in it.