Hacker News Reader

Comment by azernik | Hacker News Reader

This article is a bit unfair to the dark-forest conception of interstellar life. The axiom that Liu put forward was not about some inherently expansionist nature of intelligent life, but about the difficulty of communication and understanding. It takes two round-trips to communicate, but a small multiple of one for a species that can launch an attack that moves at a substantial fraction of the speed of light. And such attacks may be fast enough, or hard enough to detect, that second-strike capability is impossible.

In his books he explicitly compares this to the prisoner's dilemma - maybe you yourself don't want to expand beyond your own star system, but as soon as you know another species exists, you know that if they are hostile and do find you, they will destroy you before you know there's any threat. And so you are incentivized to attack when the other civilization has not yet been alerted to your presence.

View on HN

The article says

> it conceives of every civilization in the universe as a hunter hiding in a moonless woodland, listening for the first rustlings of a rival.

How about

> it conceives of every civilization in the universe as a hunter hiding in a moonless woodland, listening for the first rustlings of another. The hunters are strangers who do not trust each other (or even have a language in common), so can respond with long dangerous negotiations to establish good intent, or with a swift bullet. Everyone knows this, and acts accordingly. The stable state is pre-emptive strikes, zero trust, maximum hostility.

This is a very "me vs them" individual bias vision. It assumes every individual civilization can take another with a "single" obliterating bullet like some street gang conflict.

In practice we are ant colonies in that dark forest. Unless you can understand enough the rumblings of another civilization, you don't know how far they've expanded. They may have fully functional, fast replicating colonies in one of the trillions of Oort cloud objects and you just pissed them off by trashing their historical sites.

We could go insane hypothesizing destruction on light-years bubble scales, but then you can rest assured other civilizations are now aware of your existence and will take note of your undesirable quality as neighbor.

A good answer, though I am merely summarising what I understand of Liu Cixin's argument.

Yes, across interstellar distances, a single strike with a very high velocity object might take out a planetary civilisation before they could react. But as you say, that might not be enough.

It's even worse than "you don't know how far they've expanded", even a fast strike from nearby on the cosmic scale can take hundreds of years, and you can't know how far they will have expanded by then.

Same (I'm GGP)

I think it also makes a lot of assumptions about the nature of interstellar warfare - mainly that against a single-system enemy, attacks are quick and destructive enough - and hard enough to defend against - to destroy all ability to retaliate with high certainty. A sort of MAD gone wrong, with an attacker not having to even worry about the possibility of a second strike by the more resilient elements of enemy forces.

The addendum is that Liu believed it would be very difficult for, if not impossible, for a civilization to hold together across a long enough distance. This was shown when the original interception ships floated far enough and started attacking each other, throwing out pre-emptive strikes when communications broke down and resources were needed.

>The axiom that Liu put forward was not about some inherently expansionist nature of intelligent life

Huh? All the other stuff you said is perfectly true, but in addition to that, there's an axiom in his book explicitly stating that "civilization continuously grows and expands."

The idea the bring not that they will expand to take you over, but that you can't trust a weaker civilization to stay weaker.

Apologies, but this is giving me a brain freeze.

Previously, you talked about chains of suspicion (Liu's term for it). It was couched in a denial of there being any expansionist nature of civilizations. All the other things you said were exactly right, but, at the same time, Liu did indeed suggest expansionist tendencies of civilizations as an axiom for understanding Dark Forest theory.

And now you're saying expansion is relevant to relative strength or weakness of civilizations? You seem to be clarifying what is meant by expansion. But we can define expansion however. My point was more that expansion, however defined was definitely included among Liu's axioms.

Reply on news.ycombinator.com