Podcast Thoughts: Sara Imari Walker

Strong Emergence In Biology

I've been trying to walk 4 or 5 miles every day as of late, to help lose weight, and just stay active during The Neverending COVID. I like a distraction during these walks, rather than just dwelling on my usual thoughts of the day, so I've started listening to Sean Carroll's Mindscape podcast.

I just jump around to the topics I'm interested in, and emergent phenomena is something I've always been interested in, so I listened to a few of those recently.

The Sara Imari Walker episode was definitely interesting - interesting ideas - but highly speculative. The biggest standout for me that surprised me was her belief that somehow there's strong emergence in nature, with regard to life in particular. She believes that you couldn't, even in principle, simulate the high-level complexity from the low level rules. Unfortunately, Carroll didn't well on this long when it came up. I would have loved if we did. Belief in weak emergence and some elegant rule at the bottom its pretty common. People interested in complexity and emergent phenomena commonly think there's, perhaps, something simple analgous to Conway's Game of Life or Rule 110, some automata-like rules that cause all macrostructures to evolve later, directly as a consequence of the rules applied to the initial conditions, reductive yet very hard to predict. Maybe in their mind it takes until generation 10^9999... before they pop out, but that's the typical mental model.

She was very explicit that she doesn't think the link from chemistry up to biology could in theory be simulated on a computer, because of this belief that new physics happen at a certain scale, and we haven't captured them.

A serious belief in strong emergence is going against the grain, and I love that kind of thing. Would love to have heard more from her. Especially since this is something I know Carroll is so strongly opposed to. I would love to learn more about it by way of seeing it defended under a skeptic's intense scrutiny.

Without pressing her, it wasn't 100% certain that she didn't simply mean that it's computationally intractable, or that QM non-determinism throws a wrench in simulation potential, but it did seem pretty clear what she was claiming.

Also interesting was the idea of extending the definition of life to many different levels of complexity, down below the cell and up not just to multicellular organisms like us but to the artifacts we create. This centered around the information-theoretical aspect of her theories, and capturing the information inherent in those objects (the entire baggage of creating a living thing that can finally create that artifact being part of the information that is inherent in the object)

Getting Rid of Consciousness

Interestingly she said consciousness is among the words she wants to get rid of. This was said flippantly when talking about other language like complexity and information being muddy, so I don't know how serious she is.

But if she is serious about seeing consciousness as nothing but a convenient label for a reductively-explainable physical phenomenon that would be...interesting, considering the above wild conjectures.

Still, I'm interested in seeking out her talks, books, papers, and following the links on these ideas. Even if it turns out to be nonsense it takes your mind in different directions that can prove to be useful later in unexpected ways. I would especially like to know if this strong emergence is just an intuition she has about the nature of reality, or if there's definitely grounding to it in some esoteric papers in hers and related fields.

Steven Strogatz

On a prior day I listened to 41 | Steven Strogatz on Synchronization, Networks, and the Emergence of Complex Behavior. I was actually a bit disappointed he wasn't as bold with the conjectures and intuitive leaps. It was a more grounded talk on his seminal paper on small world networks.

At one point he talked about sleep research, and when asked about the function of it, we offered some rather circular reasoning about chemicals building up that trigger you to sleep. Yes, but why! He didn't talk about the essence of why the brain needs to do that: why it evolved a system to make it happen, what function it serves.

I thought, certainly, he would link it to network theories. Perhaps the brain needs to let the network "settle" after a day of learning? Perhaps network reorganizing of some kind needs to happen, the complex of connections that, through learning, tried to find solutions among a massive possibility space, needs to "settle" into a, maybe, "low energy" state, a valley among the solution landscape.

I'm sure I've heard some kind of claim like that somewhere, sometime, but I couldn't possibly place it. I've read so many popsci articles. It could have been the most random Wired or Scientific American article 20 years ago for all I remember.

More thoughts on this episode in another post.