Conversation

Suppose I have a prophetic vision that a giant space goat will eat earth unless we do an expensive ritual. I raise a gofundme from friends and do the ritual. No space goat attacks. You’re welcome. Many real problems have this structure. Unfortunately so do many made-up problems
13
230
The space goat problem is a cousin of the black swan problem. A rare problem that we do anticipate and solve leading many to suspect that it was never a problem to begin with. And often they’re right.
2
55
Y2K was probably the clearest example of a space goat I consider real. People anticipated the problem and put in a ton of work and basically nothing happened. But it’s entirely plausible yo me that a ton of really bad shit *could* have happened.
Replying to
If climate action works, it will be another. Covid response is a weaker example. Enough people actually died that you can’t argue it isn’t real, but you can’t make the case (I do believe it) that if it had been left entirely unmanaged it would have been vastly worse.
1
37
Nuclear holocaust is a rare real space goat everybody agrees is real because 2 data points were persuasive, and the extrapolation from megatons to gigatons is relatively easy to independently check.
2
34
On the other side, basically any explicit religious threat model to which prayer is proposed as the answer is a good example of an entirely fake space goat species that nevertheless persuades a large majority of humans. That’s my bullshit threshold.
3
35
Your space goat/response proposal has to be at least more serious than “doomsday is here/atone for your sins and pray” for me to pay attention. Which means I ignore the biggest fears of probably 50% of humanity. Though occasionally I guess the could be mythologized real things.
1
19
One of my unpopular opinions is that Pascal’s wager is bullshit. You treat an unmotivated and unsound ontological expansion as merely a very low probability proposition that you can multiply with very high disutility to arrive at a “real” risk. Something from nothing!
Quote Tweet
Replying to @vgr
a close relative to pascals wager, why take the small chance of eternal oblivion? just do [expensive thing]
3
33
Allow me to demonstrate. I agree a space goat is an extraordinarily unlikely thing to exist. As small as you like. P(space goat) = ε —>0. BUT NOT ZERO! HOW DARE YOU! THE KNOW-IT-ALL-HUBRIS! I assert the disutility of earth getting eaten is -1/ε A real risk. We must act.
1
23
The takeaway from the space goat problem is that the sheer scale of our technological abilities today means if you believe in one, you must act with no expectation of future reward or recognition, or broad validation. You can only expect to be a solipsistic messiah.
1
21
Sometimes we get lucky though. Sometimes the space goat eats the Flying Spaghetti Monster that’s also coming for us, and dies of food poisoning. And we never even notice our near miss.
1
22