It's april first so I'm going to talk about the theology I've been doing.

I wanted to play this as straight as I possibly could, but partway through writing the essay I tripped over some difficult questions I don't think I can answer. I can no longer in good conscience project the kind of anguished conviction that I'd need to be able to prank myself and my readers with a bostromian religious conversion. I am sorry. I still think it's an interesting theory, so instead of really trying to sell you on it I'll just summarise it and discuss the flaws. I'll include the rest of the essay, but I recommend it only to those with a special interest in this sort of thing.

Problems to be solved

I'll start by observing two things about the anthropics of our universe that don't make sense under the standard model (insofar as we have a standard model for anthropics)

• Anthropic measure seems to be concentrated in humans/living things, even though most of the things that exist are dead clouds of hydrogen. It is strange that when existence observes itself, despite the relative rarity of living things, observers like you or I find our subjectivity situated in the positions of living things.
• I assume panpsychism as the null hypothesis. It is often thought that any part of the universe that observes must be conscious or sentient to do it, that observing or measuring is intrinsically linked with those patterns of behaviour. It seems clear to me that that's overfitting. The thing you know to possess anthropic measure (you) happens to have conscious behaviour, so you presume there's a link. Your dataset has only one element. That's not enough.
• Think of Alzheimers. If you progressively take away the features of conscious behaviour, memory formation, symbolic thought, would you really doubt that anthropic measure still remains? Subjectivity/experience and conscious behaviour are not the same thing.
• If you don't consider anthropic measure to be a coherent or important concept, I wrote this short for you. The concept of anthropic measure is a mysterious bastard, but a time will come when betting agents wont be able to avoid dealing with it any more.
• The Doomsday Argument. It is strange that we find ourselves in the early stages of a civilization's history, considering that there will be many millions of times more of us later on. In the theory proposed here, I'm exploring a rarely mentioned branch of the doomsday trilemma.

The Teeming Consortium Hypothesis tries to make sense of these anomalies. It claims that

• Living things eventually come to pervade and control almost every region of the universe
• (I still find this kinda plausible. The universe is not designed to be impermeable to superintelligences and things that are not designed to be impermeable to superintelligences (or even regular intelligences) generally wont be?)
• Life will tend towards the goal of making as much life (think; progeny) as possible, and higher efficiencies are possible in simulations.
• Biological evolution, in reality, does not optimise the ratio of living to dead very well. Even if you can build a host planet without any immense core of stone (quite dead, is stone), the process produces a lot of dead bodies, dirt, mounds of skeletons (coral, chalk). Plants? Do we want plants or any other kind of non-sentient life? Better to just spin up a natural-ish physical simulation, while only approximating its non-sentient features.

The measure concentration anomaly is no longer an anomaly; it is explained. We find ourselves in the position of a sentient thing because most of existence has been made into a simulation substrate for sentient things. A sentient thing is the most common sort of thing that a thing can turn out to be. It is now possible to be a panpsychist, and to also expect living things to have disproportionate concentrations of subjectivity, at the same time!

The anomaly of conspicuous youth is no longer an anomaly; it is explained that we are not truly young. We are part of an old thing. At the same time, the doomsday argument may still be sort of true and we might not get the future we expected if we try to lift the stars and aestevate; we will stop resembling Life and we will no longer be so interesting. Perhaps we will be thrown away, or perhaps we will be sent somewhere new

• (A future post might explore the question "What if we make an acausal commitment to providing the living occupants of our simulations with an afterlife beyond the end of the simulation, to boost the probability that we will find we're already covered by such an agreement." I have done a post about theories in that genre before, but I don't recommend that post for various reasons. It was way dry to the point of hostility and some of the arguments it was making fail under some approximate metaphysical relative measure arguments I came across later (in summary, it turned out that we probably couldn't "afford" it, that it isn't a profitable use of resources for most kinds of agents. I may now see a variant that wont not run into this issue.))

I consider The Teeming Consortium Hypothesis less "The universe is weird because god did it" more "The universe is weird because the sorts of gods we're going to build totally would do it".

The problems

I am not certain that these flaws are terminal. All I can say is that my physics/metaphysics is gravely offended by them. If anyone sees a way to overcome these issues, please speak up, it would be very important, if you could.

The problems mainly concern entropy. It seems that, as a rule, there is generally much more entropy in the universe than energy. Before we get uranium, we had stars, burning, exploding continuously for tens of billions of years. If entropy can have anthropic measure, then that's a whole lot of stuff that can't ever be made living. We can get out of it by proposing that entropy doesn't have anthropic measure, does not participate in subjectivity, and while this seems less arbitrary than claiming that dead stuff uniquely has no anthropic measure, ultimately, it still seems like the same crime, though a less severe case.

I don't know that the entropy that exists can ever be contained. See ("The predication of Containment" for more). I was going to leave this as a "maybe we'll figure this out, let's experiment", but now, I'd say it pushes the theory into the red, for me. Note, it is not enough to just slow down entropy for a long time. No matter how long you can keep your machines running, if the entropy gets in, they'll exist far longer in a state of death.

The Sell

Here lies the remains of a failed essay. It's not even proofed. I do not recommend continuing beyond this point. Unless you want to.

I'll define panpsychism as the notion that, essentially, everything that exists possesses some anthropic measure. It is a position from which the universe can be observed.

To me, panpsychism feels like the null hypothesis. This is not an ordinary position, most people, philosophers included, tacitly assume that anthropic measure and the behaviours of sentience are- if not equivalent- self-evidently coincident.

I've found an argument that seems to allow them both to coexist.

I accept panpsychism's assertion that there is no obvious reason that a thing's degree of subjectivity should be entangled with whether or not it implements the behaviour of thought or agency. That seems intuitive, to me. If I were a formless spirit floating in a void before light or time, when I open whatever cognitive medium I have to external inputs, I would not bet that I would find myself in the vessel of a living thing. Living things are very rare, in the grand scheme of things. Dead things are common, so, if all I know is that I exist, I am probably a dead thing.

And yet, despite their rarity, I have opened my eyes and found myself in the position of a living thing. We can play a lot of silly games by selecting arbitrary reference classes, but the reference class of the living does not seem arbitrary. (I think a program for identifying living things would be quite compact.)

Panpsychism, as a conclusion, then, seems unlikely. Despite how physically tiny You are, You is where you have ended up. It stands to reason, then, that there must be something special about You (or if you don't want to seem solopsistic, maybe all humans are special in this way, or all higher mammals, or all living things), some quality that attracts anthropic measure, and since we're good children of Science, we're going to tend to focus on qualities of ours that are salient between taxa: our aliveness, our sapience. Probably, we reason, it has something to do with that stuff.

The Concentrated Reality hypothesis attempts to explain the coincidence between life and measure. It does this from the stance that seems most to me like the null hypothesis: Panpsychism, as I define it: Anthropic measure starts out evenly distributed throughout the universe, and does not inherently favour humans over anything else.

Concentrated Reality then explains how anthropic measure still finds itself concentrated in such small strange things as humans.

The Teeming Consortium premise holds that the majority of species (of those who can agree well enough to be said to collectively value anything) value the abundance of living things.

It's not an odd telos for a consortium of living things to cohere around. I don't think I know many individual humans who have advocated it, but if we ever, somehow, reach far enough out to start interfacing with alien civilizations, and if we had to find some universal morality to build a covenant around, I struggle to think what else could amount from those negotiations. The directive of the teeming consortium handily allows everyone to say, it's okay that you exist. Life is its own purpose. We will continue the work of our collective mother, Evolution. We will tile the universe with the sorts of systems that fascinate us most, and no citizen of our consortium will ever be made sacrifice to a dead, grey monument.

How do they define life? Life is not a negotiated agency rearranging matter. That is what it will build, but that is not its character. Life is shreds of sinews fighting their way up out of the dirt. To invoke life, you must invoke the earlier stages of a bloody, painful, brutally random full life-history. Its parents were not machines. Its parents were chance and soil and physics, and nothing else.

It's often been thought conspicuous that we find ourselves in the early stages of a life-history, rather than in the celestial gardens of some post-organic intergalactic commonwealth. The Doomsday Argument (Or see a video. Interestingly, Nick expresses a reluctance to accept the argument as being instructive) assumes that the conspicuous youth of our upswell implies that there shall be no gardens.

I'll agree that our youth is conspicuous, but Concentrated Reality offers a different explanation: The doomsday didn't come, the teeming consortium is very real, it made its celestial gardens, and we are presently inside them.

Materially, their gardens only loosely approximate our stars and our mountains. Much greater quantities of mass and energy- sometimes to the point of redundancy- are assigned to giving reality to the bodies and minds. Aside from the consortium's imperative to make life the most abundant substance in the universe, there are a few reasons this measure concentration might be a natural requirement of simulation.

• If you want the story to make sense from the perspective of the living, that's the thing you have to get absolutely right. The protagonists wont notice if the trees that fall out in the forest don't make a sound. For all of that, you can use an approximation. An approximation requires much less computing resources. That is to say, less mass and less energy. That is to say, less existence.
• Even when one of these living things turns a microscope upon a nano-scale silicone lattice and demands that it behave less like an approximation, that lattice still requires less resources to simulate than it takes to simulate a mind. The dead are predictable, and the question of which properties of the dead will be perceptible to the living has mostly predictable answers.

You might, now, ask, "What does this hypothesis predict? How can it be tested? What use is it?"

There is nothing vacuous about this hypothesis.

It explains two of our reality's big peculiarities (Life's conspicuous youth, the relationship between life and anthropic measure)

It is predicated upon a number of significant claims about metaphysics, intelligent life, and technology, that we will get into below. If any of them fail to apply, the theory also fails.

The predications might just be predictions. I'm going to call them predications because to me it feels like I'm producing them by rationalising. Sufficiently holistic, sensitive rationalising might not be distinguishable from ingenuous rationality. I don't know. Maybe rationality unmoores us all from common sense and maybe we have no choice but to believe crazy things. All I can say now is that, if the metaphysics to dismiss the predications exists, I haven't stumbled over it yet.

The predication of Permiability

In the territories of our consortium, the measure of the living is boosted, but what about deadzones outside of those territories? What about regions of existence whose physical laws are not so finely tuned as to produce or support life? What about the vast places beyond any consortium's reach where life never stirs?

The premise of Permiability holds that there are very few impermeable deadzones, that the boundaries between different parts of the universe tend to be leaky.

The Concentrated Existence hypothesis seems to depend on the assumption that the universe (that universes, generally) will be found to be permeable. We were wrong when we said no one would ever go beyond the mountains, or the desert, or the oceans. We were wrong when we said no one could go to the moon. Maybe our models are wrong to predict that no one will ever leave the local cluster, and maybe they will be wrong again in so many other ways. If I ask of you; are you confident that no technology will ever let us move faster than c, for now, it wouldn't be wise to give a firm answer. We will struggle to grow as vast, clever and mighty as anything here can, and then we'll meet the walls of the known universe again, and then if we find that the old laws still hold and our expansion has to end there, we will be able to say with some degree of confidence that the multiverse probably isn't Permeable and so the Concentrated Reality hypothesis cannot apply and the dead will always outnumber the living. For now, the question has to remain open.

Let's get closer to formalising this

Define universes as regions of existence that cannot interact with each other.

Permeability makes two subclaims

Innoculation: most universes (a fraction of universes with measure close to one) contain life somewhere. That is to say, in any fairly complex thing, there is a place with the right balance of variation and stability for life to emerge.

Transmissibility: there are very few barriers within universes that an intelligent species cannot pass through.

An issue with entropy

Imagine that a consortium is able to colonise the entire multiverse. For a time, the whole thing teems. Every piece of matter anywhere is ingeniously slotted into a single great resplendent garden.

What happens when the energy runs out.

However long it lasted, there is now just as much Stuff in the universe as their was before, but now all of it is Dead, and Death will have a much longer reign than Life did.

So how can the full coverage occur if entropy exists?

I see two possible answers.

• Entropy, specifically, does not have anthropic measure, and so does not detract from the measure of the living. I don't like this answer. Anthropic measure is weird enough that it might be true, but I don't like it.
• Most of the entropy that exists in a universe can be contained. To break this down into subclaims:
• Reversible computers are possible- that it is possible to instate complex computations that run without consuming energy.
• It is currently inconceivable to us, but it is hardly outlandish to think that we will keep trying to do it forever, and forever is a long time for a technological species. Eventually, we may break through to some level of reality with simple enough physics to create a frictionless wheel.
• Hidden Wells: What entropy naturally exists (exists before any living thing can reach that section of the universe) is relatively small next to stable (largely hidden) usable energy
• A parable to explain why I consider this plausible:
• For a long time, humans derived their energy from forces of nature that they could see with their eyes. Wood sustained their fires. People drove their querns and food drove the people. Water powered their mills. Unearthed fossils powered their engines.
• One day, a great change visited us. Humans discovered that way down in the matter of things hiding behind the shells of atoms was an immense well of energy, far greater than anything they had conceived of before. Even though this new energy had been lying there completely silent and stable in the earth, it was so vast that humans now had the power to destroy themselves, nature, to unleash forces that could turn the whole of the earth's surface into a barren desert.
• I do not think nuclear energy will be the last instance of Life discovering an extremely voluminous, extremely stable source of energy hidden deeper in the matter of existence than we had previously looked. I think this will happen many times. I think this is just how things tend to be. Wells of energy hidden behind the shells of atoms, beyond the reach of nature's roving fires.
• That the entropy can be permanently contained - that the unknowability of some small part of reality will not gradually infect the rest of the universe.
• To illuminate this, it might be fun to ask the following question.
• Say there's a superintelligent line of cells in a Conway's Game of Life system (or, a line of cells whose state we can control).
• A small portion of the grid is configured in an unknown, random state.
• Can a physics gen support a way of containing the entropic part of the system that works most of the time? Can we prove that we can't?

[Epistemic status: A great writer said, last year, "There’s a Jewish tradition that laypeople should only speculate on the nature of God during Passover, because God is closer to us and such speculations might succeed. And there’s an atheist tradition that laypeople should only speculate on the nature of God on April Fools’ Day, because believing in God is dumb, and at least then you can say you’re only kidding.".]