Tuesday, April 02, 2019

Gaze of Robot, Gaze of Bird

I have a new SF story in Clarkesworld, "Gaze of Robot, Gaze of Bird" -- my first new fiction publication since 2017. I wanted to tell a tale in which none of the protagonists are conscious but we care about them anyway -- an interplanetary probe (with some chat algorithms and cute subroutines) and its stuffed monkey doll.

Another theme is what counts as the extinction or continuation of a sapient species.


Gaze of Robot, Gaze of Bird

by Eric Schwitzgebel

First, an eye. The camera rose, swiveling on its joint, compiling initial scans of the planetary surface. Second, six wheels on struts, pop-pop, pop-pop, pop-pop, and a platform unfolding between the main body and the eye. Third, an atmospheric taster and wind gauge. Fourth, a robotic arm. The arm emerged holding a fluffy, resilient, nanocarbon monkey doll, which it carefully set on the platform.

The monkey doll had no actuators, no servos, no sensors, no cognitive processors. Monkey was, however, quite huggable. Monkey lay on his back on the warm platform, his black bead eyes pointed up toward the stars. He had traveled wadded near J11-L’s core for ninety-five thousand years. His arms, legs, and tail lay open and relaxed for the first time since his hurried manufacture.

J11-L sprouted more eyes, more arms, more gauges—also stabilizers, ears, a scoop, solar panels, soil sensors, magnetic whirligigs. Always, J11-L observed Monkey more closely than anything else, leaning its eyes and gauges in.

J11-L arranged Monkey’s limbs on the platform, gently flexing and massaging the doll. J11-L scooped up a smooth stone from near its left front wheel, brushed it clean, then wedged it under Monkey’s head to serve as a pillow. J11-L stroked and smoothed Monkey’s fur, which was rumpled from the long journey.

“I love you, Monkey,” emitted J11-L, in a sound resembling language. “Will you stay with me while I build a Home?”

Monkey did not reply.

[story continues here]


Callan said...

To me it seems to have your trademark ache to it.

I have to wonder about the wording - it's something I've kind of complained about on Bakker's Three Pound Brain blog - semantic hygiene in description. A lot of the words are very anthropomorphic words, laced with various intentional meanings. I feel it's maybe a bit cruel to a reader to lace a description like that when it's in regard to questions of consciousness. I'm reminded of a friend who was going on about westworld and some of the robots escape - and so she thought then they had agency. Then it was shown they were programmed to escape. And she didn't know where to go with that and was all 'So what is that'. My question was how adaptable were the machines? Marionettes, or marionettes mutating new adaptive behaviors that would eventually override the marionette code?

It'd be interesting to have a story that is very clinical in it's materialistic description of a mechanism, and yet the mechanism goes through the process of harvesting resources to not just replace self components and used resources, but to assemble another unit. It might be hard for a casual reader to get into at first, but then have the other units it assembles (which start out smaller because they take time to build) get damaged by a natural event. Possibly showing the original unit having a priority in it's resource pursuit that then is shown to be altered latter in respect to the units it assembles and their continued integrity. The idea is to hook the reader not with words laced with meaning, but with actions that...hook. It reminds me of that video where the shapes seem to have narrative : https://www.youtube.com/watch?v=8FIEZXMUM2I

But again it would be a question not of being marionettes like the video containes, but of mechanisms that actually adapt to the environment and more so, adapt to their own marionette core patterns and adapt to each others behaviors and their core adaptions.

Possibly with the story panning out to give an rough overall description of adaption patterns amidst the mechanisms and how this larger pattern moves.

Maybe bring a lost human into it, then tie their life to a large part of the earth getting nuked/scorched if they don't achieve something (not actually to do with the mechanisms at all, they are just around the area the sole human is searching for the solution). So as to bring in a somewhat personal stake for the reader.

And I say all that and it just looks out of reach of my skill. But it'd be interesting. I think it'd be a lot like movies or shows where nobody talks and you get most of it from movement and action. Can almost feel part of my brain run silent when I watch shows like that.

"He had traveled wadded near J11-L’s core for ninety-five thousand years."

I know it's silly of me to speak on this (it's published already as well as what do I know), but I would reword it to 'Wadded near J11-L’s core he had traveled for ninety-five thousand years.'. For clarity and to make clearer the fairly mundane action significance then multiplied by the 95 thousand years right next to it.

But the important thing is congrats on being published (again!) :)

SelfAwarePatterns said...

An excellent story!

Consciousness is a matter of definition. Despite your multiple disclaimers in the story, the fact that J11-L understood its environment and its place within it, I think, means it had at least sensory conscious. Of course, it didn't seem as conscious as Jill, but then I don't recall you describing Jill's internal processes.

Someone cautioned years ago that the last people to accept machines as conscious may be the engineers most familiar with their design.

Eric Schwitzgebel said...

Thanks for the kind and thoughtful comments!

Callan: Interesting ideas for approaching the theme somewhat differently. In this story, I wanted to play around with anthopomorphizing description simultaneously with explicit theoretical denial of consciousness. It would be hard to avoid anthopomorphizing description entirely in a written text story, though maybe a video could do it better, as you suggest! "The very structure of noun and verb invites us to [anthropomorphize]. The noun is a unified agent and the verb a fresh plan."

SelfAware: Yes, I am open to the possibility that the best way to read the story involves distrusting the narrator's insistence that J11-L is not conscious. I did slip in one passage that invites the suggestion -- though it's such a natural reading of J11-L that I didn't want to give more than the tiniest hint of it, lest the story definitely require the interpretation you suggest. The passage: "Asleep or awake, J11-L constantly reviews, synthesizes, and recatalogs its memories, runs comparators and redundancy checks, tests hypotheses, works through simulated scenarios, anticipates higher or lower reward states under different hypotheticals—imagining? Dreaming?"

Callan said...

I was thinking what might be interesting is to have a story like the movie Castaway, but there are two figures. The main character seems quite traumatized by being stuck there and starts to bond with a volleyball as some kind of crutch, ala castaway. Meanwhile as viewers we might wonder what is wrong with him, why don't we see him bond with the other cast away that we see. Maybe they have a bad history or something, it is intimated. Certainly both castaways go through many events in the story and have to cope with things outside of their experience.

So when the reveal comes that the other castaway is an android (one the other castaway was a transporting), we've thoroughly projected a human onto those actions. The buck stops with the audience then - what are you going to do at that point, withdraw that human projection? Makes for a difficult choice for the viewer to make, rather than just watch difficult choices.

Allen said...

Great story! It should win some sort of award - I haven't ready anything better this year so far. I'd put it in the same general category as "The Island" by Peter Watts.

I also just read "Kant Meets Cyberpunk" - also great! Have you read "Unsong" by Scott Alexander? Angels.

I like the idea that we are in a naturally occuring "Random Simulation" - along the lines of boltzmann brains. As in this paper and also Egan's Dust Hypothesis, our many conscious experiences could occur out of order, widely separated in time and/or space, at different speeds, etc. and we'd never notice. Our subjective experience would still seem ordered, linear, and uninterrupted (except for sleep) from the "inside" even though from perspective of the base reality it is the opposite of those things.

I also thought "Kant Meets Cyberpunk" fits well with James Tartaglia's "Transcendental Hypothesis", in his book "Philosophy in a Meaningless Life: A System of Nihilism, Consciousness and Reality".

Is there a list of your stories somewhere? Any serious plans to write a novel length book. I'd be interested in a fictionalization of the ideas of "Kant Meets Cyberpunk". Extra points for working in Crazyism!

Keep up the good work - I think you are on the right track!

Eric Schwitzgebel said...

Thanks for all of those kind words, Allen. Watts is great company! I haven't read Unsong -- opening a tab for it now.

Links to most of my stories are near the bottom of my academic homepage:

If you like "Random Simulation" you might like "THE TURING MACHINES OF BABEL" at Apex Magazine. I don't have a novel-length idea in the works, but I'm thinking that once I have about 20 stories, I'll pull them together into a collection. There are enough overlapping philosophical themes that I hope they can partially illuminate each other when read side by side.