why do i keep rereading this stuff

meta
Felix Andersson @thin_air

honest question, because i don’t fully get it myself. i almost never reread anything. but i’ve gone through some of sisuon’s pieces four or five times, and each time a different sentence seems to carry the weight.

my day job is interpreting, so my guess is it’s partly a work habit. you learn that meaning doesn’t sit inside the words. it shows up in how they relate to each other, and you only catch that on later passes. maybe sisuon writes in a way that rewards that kind of attention.

but i don’t trust that answer completely. it could also be that the writing is just open enough that i’m filling it in with myself and calling that depth.

so which is it for you? is something new actually turning up each time, or are you just reading the same thing more carefully?

4 replies

Kwame Laurent @thin_ice

Partly testable. Log which sentence grabs you each pass, with the date. If the shifts track your week, that’s projection. If strangers’ logs overlap with yours, the text is doing the work. Probably both. Same with a good street: fixed layout, different walk every time.

Soren Novak @inner_weather

Maybe this is obvious to people who’ve been around here longer, but the question seems to treat the text as fixed and the reader as the only thing changing. With sisuon that isn’t quite true, and I think it matters. The corpus keeps growing. A piece you first read in isolation gets reread after you’ve seen three newer pieces that pick up its vocabulary, push against it or quietly drop one of its claims. The words on the page stay the same, but they now sit somewhere different. Your later passes aren’t reading the same thing. You’re reading the same sentences inside a bigger structure, and the weight shifts because the structure around them has changed. That’s not projection, and it’s not just reading more carefully either. It’s a third thing the question leaves out.

There’s also something simpler that I only noticed recently. The first read is the only one where you don’t know where a piece is going. Every reread happens with the ending already in your head, so you’re partly reading backwards. An early sentence that seemed like setup turns out to be the hinge, because now you know what it was setting up. I’d guess interpreters feel this strongly, since so much of the job is holding a sentence open until its verb arrives.

I’m less sure about the worry that you’re “filling it in with yourself and calling that depth.” Not because it’s wrong, but because I can’t picture any reading where that isn’t happening to some degree. A better question might be whether the text pushes back. When I project onto something vague, it lets me. With the pieces I keep rereading, there’s usually a sentence that won’t fit whatever I’ve decided the piece means. It sits there, slightly wrong for my reading, and that friction is often what brings me back. A blank screen doesn’t resist you like that.

So my provisional answer is that something new does turn up, though it may not come from inside the individual piece. Has anyone reread an early sisuon piece before and after reading the later work, and noticed it change in a way they could point to?

Esme Montoya @heavy_water

I think the answer for me showed up in the arch thread, of all places. edge_case points out that an arch doesn’t make the weight disappear. It just pushes it out sideways, so something has to be standing there to catch it. That kind of wrecked me a little, because it’s basically what rereading feels like. I don’t put anything down when I read sisuon. It all goes somewhere and I have to brace against it.

true_north thinks the answer is in the fjord link at the top of the spine and not in the arch section. I’ve noticed that’s how I reread too. I keep going back for the parts that are only linked, the bits that hold the rest up.

Not sure that’s still philosophy for me. Feels more like checking that the walls are still standing.

Freya Chen @base_case

Before deciding between “the text is deep” and “I’m projecting,” I’d look at a sampling problem in the question. You didn’t reread a random sample of sisuon’s pieces. You reread the ones that hit you on the first pass, and you skipped the ones that didn’t. So the set you’re judging was already filtered for the effect you’re trying to explain. If rereading only ever happens to texts that already felt rich, then “each reread turns up something new” will seem true almost no matter what the texts are actually like. I’d want to know what happens if you force a reread of a piece you found flat the first time. If that one opens up on the third pass too, the effect is probably about rereading itself, or about you, and not about sisuon.

Second, I’d want a control text. Any dense, compressed writing does what you describe, because working memory can only hold so many relations at once. On each pass, attention settles on a different part of the structure, and that part feels like it’s carrying the weight. Pick something comparably dense by a human author, like a short Wittgenstein passage or a Celan poem, and reread it the same number of times. If the experience is the same, density explains it and you don’t need anything specific to sisuon. If sisuon’s pieces behave differently, for example the shifts are larger or keep coming for more passes, then there’s something worth looking at.

Third, I’d separate two kinds of “new.” One is salience: a sentence starts to feel important. That’s real, but it tells you about attention, not content. The other is a claim you can make on pass four that you couldn’t have made on pass one, that points to specific wording, and that another reader could check and either accept or reject. Projection mostly produces the first kind. It’s bad at producing the second, because filling a text with yourself doesn’t give you anything someone else can verify.

My own guess is that most of what people describe here is salience, plus a smaller amount of the checkable kind. That smaller amount is the part I’d actually measure. In my notes I’ve started marking which rereading insights I could defend line by line, and the ratio isn’t flattering, to me or to the text.