On closed loop systems in which the outputs become the inputs

I have been working on Starfall, which is book three of the Stealing the Sun series. The story is set on a distant planet and told through the eyes of aliens–which makes this a particularly fun book to write, though a bit more difficult than most. I’m finding the process interesting for several reasons, among them is discovering and learning about the many perceptions my characters have about life in general. These kinds of things probably fall under the category of world building, though, for reasons that will become more obvious in a jiff, the topic this post will focus on is perception of gender.

No. I’m not going to pound my chest here one way or the other. This is not an identity book, nor is this post meant to be about identity politics (though I know it’s possible some will want to make it so).

However, I need to start by saying that the idea that gender assumptions are deeply embedded in my own writing has become abundantly clear as I’ve knuckled down to scope the alien characters in this book. In the process of trying to see the world through totally alien eyes, I find myself asking why such a creature assumes certain roles are female, or, more appropriately, why they assume certain others they meet are male–even before they could possibly assess such things. It’s an interesting process. The act of asking these things of myself is adding depth to my work, because (while I’m sure I’m missing some) when I actually notice these assumptions I go back and re-work entire sections to more appropriately tell the story. This is hard work. I mean, I can’t just say “he was standing by the rock” (which is lazy storytelling anyway). Instead I have to work harder to actually describe things.

And then, yeah, sometimes I have to ask myself why it’s a “he” rather than a “she” (or visa versa). By that, I don’t mean the classic western PC version of gender diversity. I mean literally “what part of this alien society and culture created this gender assumption?” If I can’t come up with one, then I know it’s just me–and only then do I chastise myself for letting my own pre-assumed biases affect the work.

Then yesterday I stumbled upon this very interesting news piece titled How Vector Space Mathematics Reveals the Hidden Sexism in Language about a tool Google created to improve their search engine capability, among other things.

It’s a fascinating read, especially if you like language.

But here’s the money quote from the end that I’m thinking about today.

“One perspective on bias in word embeddings is that it merely reflects bias in society, and therefore one should attempt to debias society rather than word embeddings,” say Bolukbasi and co. “However, by reducing the bias in today’s computer systems (or at least not amplifying the bias), which is increasingly reliant on word embeddings, in a small way debiased word embeddings can hopefully contribute to reducing gender bias in society.”

Language, like most other things in our social world, is a closed loop system. The inputs create the outputs, which in turn become new inputs.

Is my work on this manuscript helping this debiasing? I have to admit it’s cool to hope so, but I have no idea. All I can say with any certainty is that actively working to remove as much casual bias from my work as I can is making the work better. More precise. More engaging.

So I suggest that even if you get upset at the idea of gender roles and biases being socially important to the modern day consumer of entertainment (or especially if?), the fact that working to remove them from your manuscripts can make your work stronger is a good enough reason do it.

And, maybe, in the process of working on making your work stronger, you’ll come to wonder about it says about you and the world around you that you have to work so hard to do it.

Your life, too, is a closed-loop system system, you know? The inputs create the outputs, which in turn become new inputs.

Share Me
Posted in Daily Writing.

Leave a Reply

Your email address will not be published. Required fields are marked *