David R. MacIver's Blog

Using your voice

Over the last year it feels like my brain has thoroughly been captured by work, and as a result my writing has starved a bit, especially about nontechnical subjects. I’m going to try to return to writing with the best method I know for ending a dry spell: Daily notebook posts!¹

Ironically, or possibly appropriately, I’m going to start with one of the things that I think has kept me away from writing a bit recently: I seem to be much more into the spoken word at the moment. I haven’t been reading much recently either. Or rather, I’ve been “reading” lots of books, in the sense that I’ve been listening to a lot of audiobooks.

I don’t think there’s anything wrong with audiobooks, but they’re much better for fiction than for nonfiction for me. I can listen to podcasts where people are having a conversation for interesting non-fiction thoughts (though it’s a very different experience than reading a book), but the nature of listening to an audiobook is very much that you are carried along at the narrator’s pace. If you want to reread something you didn’t understand, or pause for reflection or to have another unrelated interesting thought. The conversational mode on the other hand is much easier to dip in and out of, because it doesn’t tend to be a tightly layered argument.

This is, I think, one of the big differentiators between listening and reading: With the written word, the default is not reading. With the spoken word, the default is to listen. It requires a conscious decision to stop.²

This can also be a feature! One of my big use cases for audiobooks is something to listen to as I fall asleep. There, the difficulty of getting caught up in my own thoughts is an advantage. It makes it much easier to avoid getting caught up in thinking spirals.³

As I’ve pointed out in the past, these strengths of audiobooks are also not entirely a good thing, which is part of why I’m trying to get back more into the written word.

There’s another interesting combination though that I’ve often felt is neglected (certainly I neglect it, and my impression is that I do it a little more than most people do): Reading out loud. This combines the spoken and written word in an interesting way. It lacks the “continuing is the default” aspect of the spoken world, but it’s a little harder to unintentionally stop than it is with reading silently. Using your voice is an anchor to the task.

It also sometimes is just the right format for a medium. I’ve been reading Arthur Machen’s “The White People”⁴ today, and the middle of it is this long continuous stream of words. I originally encountered it via the podcast “Weird Studies”, who also did a reading of it, and I couldn’t make it through the stream of words. It almost forced me to zone out. So I thought I’d try it as written word. Turns out, it’s also hard to read.⁵

But… there’s something to the experience of reading it aloud. It almost forces you into a kind of singsong, with its continuing paragraph-free flow of prose that jumps from one scene and thought to the next without pause. It’s an aesthetic experience that I think requires you to immerse yourself in it, and by reading it aloud, you make it more “yours” in a way that allows you to inhabit an otherwise difficult to reach state of mind.

Another place I’ve been experimenting with the spoken voice is at work. I’ve been chatting to a codex on fast mode / light reasoning in a particular codebase⁶ about how I think things should work. It’s about 80% my braindumping, and 20% codex fact-checking what I said against the actual code, telling me where it’s different, and asking me questions about how I think specific things should work, with the goal of producing notes.⁷ It’s been very useful, because I can just ramble my thoughts out without having to structure them in the way I would for even fairly rough writing like this. There’s no real attempt at linearity.

It also requires me to block out a meeting room in order to do it, because I can’t do that when around colleagues really. Not because I’m embarrassed to do so, but because it would actively impair their ability to do work.

This is a somewhat frustrating feature of the spoken word: It imposes on those around you in a way that the written word doesn’t. Another conversation I’ve got going on with ChatGPT⁸ is a discussion about the nature of sin (more on that in another post), indeed inspired by Machen and some of the discussion around it, and in order to be able to carry that conversation on on the train, I’ve been typing messages to it and letting it reply in voice. It’s not a bad compromise, and gives me plenty of pause to think, but it does mean that a lot of what I’m thinking isn’t actually captured, which makes it a much worse experience than the conversation about code.

One solution to this of course is to do more spoken word things that engage with those around you. Conversations, of course, but also spoken word activities. I did a play reading recently⁹, and that was a fun activity.

This feels like a cop out though. Many of the things I want to do with the spoken word are not treating the spoken word as a goal in and of itself, but an instrumental tool for achieving some end, and using that tool for different things doesn’t get me any closer to that end.

This touches on a recurring feeling I have: The desire to speak and not be heard. Sometimes the word I want to speak is actually a scream, and there is nowhere to let that out that doesn’t impose on other people.