zompist's Essay on "Advanced Civs"

[Raphael]

A few years ago, zompist posted an essay on how to write fictional "highly advanced civilizations" over on his Patreon. More recently, he made that post public, that is, available to everyone, including people who aren't on Patreon.

The post is at https://www.patreon.com/posts/advanced-64840097

Does anyone who is not on the Patreon have opinions on it?

[linguistcat]

I will take a look at it when I have time, and leave my thought here once I do.

[WeepingElf]

I have taken a brief look at it, and shall read more thoroughly later. It is well-written, and I don't see any problems. A minor addition I'd make concerns the G-type stars: You need a G-type main sequence star; G-type giants are useless (but much rarer anyway).

[bradrn]

An excellent essay. Thanks for mentioning that it was made public!

[sasasha]

It is a great essay. Easily one of the best treatments of this I’ve read, and I like that it deals well with the Fermi paradox without ever mentioning it.

One thought I have on it is that I seem to have missed or forgotten the tidbit about iliu music being a low tech and amateur entertainment... Exciting. Makes me think of this. (Low tech for an iliu, anyway...)

Of course, they probably have on-land and underwater musical genres...

[Glass Half Baked]

The question of stagnation is an interesting one. On the one hand, our society has gotten more interesting and complex, and made more esoteric things to care about, over the years. But on the other hand, every single one of those interesting complexities is something that greatly decreases the probability of survival, and we've already decided we have to get rid of all that stuff if we have any hope of lasting into the far future. We can't say "well, obviously they will have to moved on from the sort of partisan squabbling that leads to preventable catastrophe" and then in the next breath say "they will have higher levels of social complexity than us." Our social complexity has been a death cult for ten thousand years. Any scrutible super-advanced civilization should look like a bunch of idiots one button-press away from annihilation. Getting that far intact should be like winning the galactic lottery, not an inevitable result of progress.

[linguistcat]

I've considered the idea of cultures that intentionally regress, in some ways at least, while keeping high tech for when it's needed.

I forget the name, but such a culture was involved in one of the episodes of the original Stargate series. The main crew arrive at a world that looks neolithic at best with a species of peaceful gatherers (I don't remember them hunting or fishing at all). There is an apparent threat to the species, and the main characters debate if they should get involved. They come to the conclusion that they should. (I'm not remembering clearly what the threat was but I think the human explorers felt they were at fault somehow, and hence why they should try to fix things.) But it turns out this species just chooses to live simply and are actually as advanced as The Ancients who created the Stargate system.

Stargate also shows an advanced species who could solve a lot of the issues humans have, but have issues of their own.

Of course, Stargate is also just on the edge of science fantasy so maybe a bit outside the purview of the essay, but still showing examples of some things Zompist had considered.

The essay itself is very well written and covers a lot of different situations that could occur.

[keenir]

I've considered the idea of cultures that intentionally regress, in some ways at least, while keeping high tech for when it's needed.

I forget the name, but such a culture was involved in one of the episodes of the original Stargate series. The main crew arrive at a world that looks neolithic at best with a species of peaceful gatherers (I don't remember them hunting or fishing at all).

I don't think they even gathered - everything was ornamental.

There is an apparent threat to the species, and the main characters debate if they should get involved. They come to the conclusion that they should. (I'm not remembering clearly what the threat was but I think the human explorers felt they were at fault somehow, and hence why they should try to fix things.)

you're thinking of the Nox.

the threat was a visit from the Goa'uld System Lord Apophis, who was the team's enemy (though Apophis wasn't there in pursuit of them - the planet was a regular place for goa'ulds to explore for advanced tech that was rumored to exist there)

[Glass Half Baked]

Speaking of more fantasy than hard SF, the TOS episode Errand of Mercy makes wonderful use of the trope of the advanced civ that appears to be primative. They are essentially a foil to give the Federation a chance to oppose the Klingon Empire in a way that shows them to be similar in their arrogance and privilege. Obviously a Cold War commentary, but in usual Star Trek fashion it remains relevant in numerous time periods.

[sasasha]

Our social complexity has been a death cult for ten thousand years. Any scrutible super-advanced civilization should look like a bunch of idiots one button-press away from annihilation. Getting that far intact should be like winning the galactic lottery, not an inevitable result of progress.

My jury is still out on this. On the one hand, a super-advanced civilisation will have in technicality some kind of access to super-advanced ways to annihilate itself. On the other, to my mind, there’s no doubt that they will be run by super-intelligent self-aggrandising (if not -replicating) AI that will take potentially unassailable steps to control access to such methods. Only other agents of their own nature and stature would be able to threaten to do so.

More: show

I personally can’t see a far-future in our little bit of space that doesn’t involve a war (however cold or hot) between rival AI governance systems, and my guess is that they will be eliminated or assimilated until one is left (or possibly two, in awkward stasis, though I doubt that would actually last). At that point, when you have one agent powerful enough to run entire solar systems or a local cluster of systems ‒ like Banks’ Culture ‒ assuming zomp’s guess about FTL travel remaining impossible is right, what will truly be able to threaten it, beyond coming into contact with another similar system?

Given, I dream up scenarios set in this kind of universe, and they inevitably revolve around me trying to find opportunities for narrative, and often those are like ‘actor X tries to find a chink in the AI armour so they can influence the governance system in their favour or against a rival’. But often these stories end up feeling contrived; I just think, that at a certain level of runaway development, most problems affecting lowly biological lifeforms will be sorted out, and there will be very few if any chinks in the armour.

Incidentally, the biggest problems in my scenario are totalitarian responses to death and ‘the quota’: as it affects resource management and economic planning so significantly, and can be avoided, nobody’s allowed to die without permission (and accidental death has to be proven as such to the Audit to avoid the total confiscation of the deceased’s assets by the system), and similarly, there’s an enormous waiting list to be allowed to reproduce (in any form). Granted, I don’t find this part of it inevitable, this is me trying to find story in what otherwise looks like a fairly meaningless existence of conscious agents trying to entertain themselves eternally in hyperreal simulations. In other words it’s my version of zomp’s scenario of the system developing madness, which allows for narrative possibility.

Like our own civilisation, resource management is perhaps the biggest existential threat here. This is where the desire for stasis comes in to my governance systems: at a certain point, it stops making sense to desire growth, and you can stop needing growth if you have (a) sorted out responses to infrequent existential threats like star death, and (b) can stop your populace changing (beyond like a standard deviation).

Never really shared these ideas before BTW so they definitely need picking apart! E.g. whether personal property is likely to exist at all in such a scenario.

[sasasha]

Huh, sorry to post again, but, I just thought of another Fermi’s paradox possibility: civilizations deliberately ‘clear an area’ around them, i.e., reach a demonstrably self-sustaining economic model within themselves with an enormous contingency set aside for unexpected challenges, then physically destroy nascent or potential biospheres and strip-mine useful resources in an enormous radius, such that they can get along inside the bubble pretty much unnoticed, and nothing that comes near enough will find anything worth having or that would help them continue their journey.

Hmm... Not sure about it, but could be worth exploring. It again relies on STL travel. Star death would be a concern, as the strategy would eventually preclude expansion. Also to justify the enormous costs of the policy there must be some serious fear of external annihilation. Perhaps it could appear as another form of ‘civilizational madness’ though... And might arise naturally, from the strip-mining that could be a natural economic consequence of such a civilization in its growth phases.

Now, if there was some kind of quasi-magic way to get between disparate locations in the universe, i.e. travel in a neighbouring really small universe and back out the other side, this might work well as you could have as many bubbles / pockets as you liked, which no one else could get to unless they also mastered your transportation system.

But this is getting off-topic and also very science fantasy.

[evmdbm]

I'm now curious as to what Zompist - and others - think of Dune. We must be 30-40,000 years in the future and have an emperor, dukes, barons and counts, some significant advanced technology (some more hand-wavy than others) but also capitalism, slavery (at least on Giedi Prime) despotism and attempted genocide (of the Fremen).

There's a need for a narrative device/convenience to keep us on track with the desert nomads control the oil (I mean the spice) and we Westerners (I mean the Harkonnen/Corrino/Spacing Guild etc) have to have it, but Herbert does seem to have done thinking about what sort of society might develop in the absence of AI. It's not a failure state, but it's not purposeful regression either (at least not totally - maybe denying yourself AI counts, but other technological developments are A Ok)

Then of course it just gets weird with the accession of Leto to the throne

[sasasha]

Isn’t it a failure state? At least, a recovering one. There was a cataclysmic war 10,000 years in our future in which all computers were wiped out, and it was so bad that the injunction against them continues to be upheld (to a zealous degree far past the edge of ‘madness’) for another ten thousand years.

(I think Dune is set 21,000 years from us, and my main issue with it is that culture in general is way too similar to our own)

[linguistcat]

I don't think they even gathered - everything was ornamental.

...

you're thinking of the Nox.

the threat was a visit from the Goa'uld System Lord Apophis, who was the team's enemy (though Apophis wasn't there in pursuit of them - the planet was a regular place for goa'ulds to explore for advanced tech that was rumored to exist there)

I was going off what I remembered from 20 years back or so, so I figured I'd get a few details wrong. Or just wouldn't remember them at all. The point I was trying to make was that advanced races might look primitive but idyllic, while still having access to and knowledge of advanced tech. But it's nice to remember the story more clearly. The idea has followed me since watching, though I haven't had much reason to use it in a story myself.

[zompist]

I'm now curious as to what Zompist - and others - think of Dune. We must be 30-40,000 years in the future and have an emperor, dukes, barons and counts, some significant advanced technology (some more hand-wavy than others) but also capitalism, slavery (at least on Giedi Prime) despotism and attempted genocide (of the Fremen).

There's a need for a narrative device/convenience to keep us on track with the desert nomads control the oil (I mean the spice) and we Westerners (I mean the Harkonnen/Corrino/Spacing Guild etc) have to have it, but Herbert does seem to have done thinking about what sort of society might develop in the absence of AI. It's not a failure state, but it's not purposeful regression either (at least not totally - maybe denying yourself AI counts, but other technological developments are A Ok)

It's been years since I read Dune, but I wouldn't call it an advanced civ at all (in the sense I'm using). As you point out, the basics are a projection of 20th century economics. Is there much in Dune that wouldn't fit in Star Trek (besides the more cynical view of human politics)?

I'd say much the same of Asimov's Foundation, which is supposed to be at least 10,000 years out, but the rad exciting tech is atomic power, and when it's lost people go back to "coal and oil." (Fortunately the planets all kept oil refineries and drilling platforms ready during that 10,000 years of using atomic power...)

On the meta level, I'm not criticizing these guys— they want to write, and we want to read, space opera, in terms comprehensible to 20th/21st century American readers. As for the medieval trimmings, they pretty much come to an author's head naturally, if ironically, when they're trying to think of something "not quite like the present".

[zompist]

My jury is still out on this. On the one hand, a super-advanced civilisation will have in technicality some kind of access to super-advanced ways to annihilate itself. On the other, to my mind, there’s no doubt that they will be run by super-intelligent self-aggrandising (if not -replicating) AI that will take potentially unassailable steps to control access to such methods. Only other agents of their own nature and stature would be able to threaten to do so.

I personally can’t see a far-future in our little bit of space that doesn’t involve a war (however cold or hot) between rival AI governance systems, and my guess is that they will be eliminated or assimilated until one is left (or possibly two, in awkward stasis, though I doubt that would actually last).

If the techbros get their way, sure. But in sf we should take a longer view. No ruling class, or ruling system, lasts forever.

The Achilles heel of your AIs, I think, is open-source. Technological secrets are hard to keep, especially in a world that requires high literacy and scientific knowledge. There's really nothing stopping you from creating your own ChatGPT— the basic methods are the same for hobbyists and the big boys. The latter can just afford much larger training sets.

And sure, ChatGPT isn't a real AI; but an AI will be a technology. Think of nuclear weapons: a lot of work has been done to keep the methods a secret, but they really aren't, if you have a good physics department. The mafia and the terrorists don't, but (say) Pakistan does.

I need to keep from trying to influence writers in my direction, but I can't help pointing out that "AIs vs humans" is a very 20th century concept, ingrained into us by every robot story ever. But an advanced civ has control over its own genome— more broadly, over its own biological character. Whatever the AI's got that you want, just throw it into your body. Cyborg it up!