The Universe leaves out the Details
🌌

The Universe leaves out the Details

‣
Menu

How to solve problems 🦢Solving Aging

Fiction đź““The Hike

Other Projects Arcadia DEEPWAVE

You can comment on everything. Or give me (harsh) personal feedback.

Made with đź’™ by me (how to).

Dedicated to my family.

The Universe leaves out the Details

Status: Draft Epistemic status: Hypothetical Early thoughts on my emerging metaphysics

If you splash around in a bathtub, what do you observe? Well, some waves. But they dissipate quite quickly. But if you get lucky, you get these whirling spirals, which can stay around for a while.

This points to a fundamental insight: We have a stability bias in our observations. Whatever you observe was not only caused at one point, it also did not (yet) dissipate. If you want to answer: why this exists all you have to figure out is why it still does - as in a sufficiently energetic system, everything gets created.

There are two types of solutions: robust and resilient systems. Robust systems are things that require extreme activation energy to be changed and we find them for every force: planets, atoms, particles. They arise by simple positive loops: mass attracts mass, so all mass in the influence of Earth got sucked into Earth. The fate of robust systems is simple: they either get sucked up into a larger system (like the moon with Earth) or get dissipated, if more than their activation energy is added (the sun explodes).

But at the human scale, we observe very few things of this shape: many natural things break if you add energy. Burn a log of firewood and it is completely gone ten minutes later. Poke an animal with a stick and it dissolves. So how can they still exist, given that it has been a long time since God splashed around in the primordial soup?

The answer is what I call “replicative watermills”: patterns that tap into the river of energy flow and produce … themselves. This is what early life was: just the reverse Krebs cycle - a simple set of autocatalytic reactions that at the end of one “rotation” produced the two units of the input.

These watermills have a problem though: they are only stable under very narrow conditions. If the river runs dry, overflows, or is filled with dirt they break. So why do we still observe them, and that everywhere?

Evolution has direction

The classical answer - mutations increase fitness - explains some of this: if you slightly change the watermills you end up with energy buffers, overflow protections, and entropy export. But without something else, we would still be some chemicals around a black smoker. Classical Darwinism is a “satisficer”, there is no telos or direction, no objective better, just an eternal cycle until the world runs dry. But the true story of life is one of unlocking new rivers. Of environment builders. If life is successful (and right now it looks like it) we will observe more and more watermills, even at places entirely unimaginable for simple chemistry (like turning all of the sun’s energy into space stations). Life is an optimization process for capability (the capacity to turn more and more of the universe’s lakes into reservoirs).

This increase in capability comes add a cost: complexity. Being able to get energy from the Arctic requires different solutions (fat, warm blood) than from the air (flight, high metabolism), some incompatible. I write about this fundamental capability-complexity tradeoff in 🧵Evolution favors Simplicity, and about one solution in 🏡Emergence as Self-Domestication - basically that if you have a sharing of labor plus simple communication you can build “ecosystems”, like organisms, colonies or actual ecosystems like coral reefs. This “multi-scale competency hierarchy” is a neat solution to the problem, but it also has some deep metaphysical implications.

Attention is all that matters …

You do not know how many blood cells you have. Your blood cells do not know the laws of quantum gravity. On first glimpse, this just sounds like a limitation. Maybe, it’s just too hard for biology to give that information to you. But I’d argue that even if you knew, you couldn’t do anything with it. You actually have access to very fine bodily signals in deep meditation states - but your attention is limited. So the system constraints heavily what you can be aware of, such that the choice of what to focus on becomes easier.

This is true for all systems - on all scales. The core question of life (which is where we differ from robust systems) is: what matters, rather than what exists. If you observe your environment closely, you notice that nearly everything does not. 99.99999% of your experience is filled with fungible things: the slight color patterns on the wall behind you, temperature fluctuations on your skin, unread messages in your spam folder, etc. Through bad memes around “everything is connected” and the “butterfly effect”, we have a picture of life in which even the tiniest things matter, but it’s the exact opposite: life only happens when there arises a thing that decides itself to matter infinitely more than any fluctuations of the universe.

… because Abstractions are Fundamental

My current (probably highly controversial?) metaphysics is that this is actually deeper than we think. Rather than there being an infinitely detailed generator function that produces reality (called physics) - which we only observe through many levels of compression and filters, I believe that the universe itself gets locally compressed once a system has perfectly captured the underlying generator function. When you look at a wooden table you see what is there: a pattern of arbitrary and fungible wood lines - there is nothing more. You can dismantle it and figure out what it’s made of, but at that moment you care about it, so the universe generates more details. Think of how we write computer games these days: compress away all details that don’t matter and generate them on the fly. If the universe where a simulation, this is how it would be computationally most efficient, but it is also a simpler explanation. Any observer will notice over time that most parts of the things they observe don’t matter, so his internal model approaches the maximally reachable compression rate, which is exactly when you have perfectly seperated experiental noise from the generator function. But at that point there is only an observer who is losslesly observing the generator function - so speculating that there is anything else is weird. Now just introduce entropy exporters (life) whose primary activity is to remove noise from their environment (the generator function of one level below them) and produce a compressed representation to the scale above - and it really becomes meaningless to speculate that there are actual cells in your brain or quantum physics in cells.

Again, you can observe cells in your brain if you look hard enough, which looks like trying to see the world from the perspective of a neuron. Because of course cells exist in your brain - from the perspective of cells in your brain. The universe does not look different from different angles - it is different. There are enough observers at enough layers that it even is useful to speculate the existence of a “ground-level generator function”. But this does not imply that the universe is not highly compressed nearly everywhere.

But what about the actual Butterfly Effect?

If the universe does not contain details, how do you explain the instances in which it does? When a single mutation due to a quantum interaction causes a shift in your experience? This critique presupposes causality - but that only really exists for robust systems (you can throw a rock at another, which might or not might break it). Watermills behave differently: they change through phase transitions. Think bank runs: if everyone tries to withdraw cash at the same time, the system collapses. So if people suspect that one might happen soon, they need to withdraw money immediately, before it is too late. It is actually impossible to determine who “caused” the event, a few possible tiny events caused a very large amount of fungible low-level events, and the dynamics of the macroscale system changed. Biological systems are extremely resilient to this kind of phase change (because most phase changes are towards death) - and therefore different than the weather - but they can still have phase changes (and even optimize for them - see the critical brain hypothesis). This does not imply that there was a first mover at a lower level - just a discontinuity in the underlying generator function which caused a phase transition in the abstraction.

Wait, so if I look twice, I might see different things?

Yes. Living systems did evolve memory, without which you cannot detect change, crucial for survival. But there are large parts of the world which are not memorized in detail by anyone. So how can you proof that those are actually different? If someone blindly puts a deck of cards in a box and we open them 100 years later - how can we tell that they are different? We can’t - and the universe can’t either (given that the simplest possible function is f(x) = 0, I expect “is consistent over time” to be a core property of all generator functions - but this is an assumption, and I am unsure if it is actually simpler than “doesn’t matter”).