Mostly agree. Would downgrade to “can’t or won’t”. Apart from a little more completeness the difference makes a difference to anthropic considerations.
Does it even make sense to say “won’t”, or for that matter bring up anthropic considerations, in reference to causality violation?
I’m not sure. If a universe allows sufficient causality violation then it may be that it will be too unstable for observers to arise in that universe. But I’m not sure about that. This may be causality chauvinism.
(I feel like there’s a joke to be made here, something to do with “causality chauvinism”, “causality violation”, “too unstable for observers to arise”, the relative “looseness” of time travel rules, maybe also the “Big Bang”… it’s on the tip of my brain… nah, I got nothing.)
Does it even make sense to say “won’t” [...] in reference to causality violation?
Yes. (Leave out the anthropics, when that makes sense to bring up is complicated.)
Most of the reason for saying:
If there are ways to violate causality they are likely restrictive enough that we can’t use them to violate causality prior to when we knew about the methods (roughly).
… are somewhat related to “causality doesn’t appear to be violated”. If (counterfactually) causality can be violated then it seems like it probably hasn’t happened yet. This makes it a lot more likely that causality violations (like wormholes and magic) that are discovered in the future will not affect things before their discovery. This includes the set of (im)possible worlds in which prior-to-the-magic times cannot be interfered with and also some other (im)possible worlds in which it is possible but doesn’t happen because it is hard.
An example would be faster-than-light neutrinos. It would be really damn hard to influence the past significantly with such neutrinos with nothing set up to catch them. It would be much easier to set up a machine to receive messages from the future.
It may be worth noting that “causality violation” does not imply “complete causality meltdown”. The latter would definitely make “won’t” rather useless.
Well, it’s just… how could you tell? I mean, maybe the angel that told Colombo to sail west was a time-travelling hologram sent to avert the Tlaxcalan conquest of Europe.
An example would be faster-than-light neutrinos. It would be really damn hard to influence the past significantly with such neutrinos with nothing set up to catch them.
Well yes, I understand you probably couldn’t use faster-than-light neutrinos from the future (FTLNFTFs) to effect changes in the year 1470 any more easily or precisely than, say, creating an equivalent neutrino burst to 10^10^9999 galaxies going supernova simultaneously one AU from Earth, presumably resulting in the planet melting or some such thing, I don’t know.
However, elsewhere in this thread I suggested a method that takes advantage of a system that already exists and is set up to detect neutrinos (admittedly not FTLNFTFs specifically, though I don’t know why that should matter). I still don’t see exactly what prevents Eliezer_2831 from fiddling around with MINOS’s or CERN’s observations in a causality-violating but not-immediately-obvious manner.
Well, it’s just… how could you tell? I mean, maybe the angel that told Colombo to sail west was a time-travelling hologram sent to avert the Tlaxcalan conquest of Europe.
We obviously can’t with certainty. But we can say it is highly unlikely. The universe looks to us like it has a consistent causal foundation rather than being riddled with arbitrary causality violations. That doesn’t make isolated interventions impossible, just unlikely.
I still don’t see exactly what prevents Eliezer_2831 from fiddling around with MINOS’s or CERN’s observations in a causality-violating but not-immediately-obvious manner.
Overwhelming practical difficulties. To get over 800 years of time travel in one hop using neutrinos going very, very slightly faster than light the neutrinos would have to be shot from a long, long way away. Getting a long, long, way away takes time and is only useful if you are traveling close enough to the speed of light that on the return trip the neutrinos gain more time than what you spent travelling. Eliezer_2831 would end up on the other side of the universe somewhere and the energy required to shoot enough neutrinos to communicate over that much distance would be enormous. The scope puts me in mind of the Tenth Doctor: “And it takes a lot of power to send this projection— I’m in orbit around a supernova. [smiling weakly] I’m burning up a sun just to say goodbye.”
I’m not sure if that scenario is more or less difficult than the remote neutrino manufacturing scenario. The engineering doesn’t sound easy but once it is done once any time before heat death of the universe you just win. You can send anything back to (almost) any time.
In the context of almost every proposed causality violation mechanism I’ve seen seriously discussed, it really is can’t, not won’t. Wormholes aren’t the only example. Tipler Cylinders for example don’t allow time travel prior to the point when they started rotating. Godel’s rotating universe has similar restrictions. Is there some time travel proposal I’m missing?
I agree that when considering anthropic issues won’t becomes potentially relevant if we had any idea that time travel could potentially allow travel prior to the existence of the device in question. In that case, I’d actually argue in the other direction: if such machines could exist, I’d expect to see massive signs of such interference in the past.
In the context of almost every proposed causality violation mechanism I’ve seen seriously discussed, it really is can’t, not won’t.
There are plenty of mechanisms in which can’t applies. There are others which don’t have that limitation. I don’t even want to touch what qualifies as ‘seriously discussed’. I’m really not up to date with which kinds of time travel are high status.
Ignore status issues. Instead focus on time travel mechanisms that don’t violate SR. Are there any such mechanisms which allow such violation before the time travel device has been constructed? I’m not aware of any.
Mostly agree. Would downgrade to “can’t or won’t”. Apart from a little more completeness the difference makes a difference to anthropic considerations.
Does it even make sense to say “won’t”, or for that matter bring up anthropic considerations, in reference to causality violation?
This is a serious question, I don’t know the answer.
I’m not sure. If a universe allows sufficient causality violation then it may be that it will be too unstable for observers to arise in that universe. But I’m not sure about that. This may be causality chauvinism.
(I feel like there’s a joke to be made here, something to do with “causality chauvinism”, “causality violation”, “too unstable for observers to arise”, the relative “looseness” of time travel rules, maybe also the “Big Bang”… it’s on the tip of my brain… nah, I got nothing.)
Yes. (Leave out the anthropics, when that makes sense to bring up is complicated.)
Most of the reason for saying:
… are somewhat related to “causality doesn’t appear to be violated”. If (counterfactually) causality can be violated then it seems like it probably hasn’t happened yet. This makes it a lot more likely that causality violations (like wormholes and magic) that are discovered in the future will not affect things before their discovery. This includes the set of (im)possible worlds in which prior-to-the-magic times cannot be interfered with and also some other (im)possible worlds in which it is possible but doesn’t happen because it is hard.
An example would be faster-than-light neutrinos. It would be really damn hard to influence the past significantly with such neutrinos with nothing set up to catch them. It would be much easier to set up a machine to receive messages from the future.
It may be worth noting that “causality violation” does not imply “complete causality meltdown”. The latter would definitely make “won’t” rather useless.
Well, it’s just… how could you tell? I mean, maybe the angel that told Colombo to sail west was a time-travelling hologram sent to avert the Tlaxcalan conquest of Europe.
Well yes, I understand you probably couldn’t use faster-than-light neutrinos from the future (FTLNFTFs) to effect changes in the year 1470 any more easily or precisely than, say, creating an equivalent neutrino burst to 10^10^9999 galaxies going supernova simultaneously one AU from Earth, presumably resulting in the planet melting or some such thing, I don’t know.
However, elsewhere in this thread I suggested a method that takes advantage of a system that already exists and is set up to detect neutrinos (admittedly not FTLNFTFs specifically, though I don’t know why that should matter). I still don’t see exactly what prevents Eliezer_2831 from fiddling around with MINOS’s or CERN’s observations in a causality-violating but not-immediately-obvious manner.
Other than, you know, basic human decency.
We obviously can’t with certainty. But we can say it is highly unlikely. The universe looks to us like it has a consistent causal foundation rather than being riddled with arbitrary causality violations. That doesn’t make isolated interventions impossible, just unlikely.
Overwhelming practical difficulties. To get over 800 years of time travel in one hop using neutrinos going very, very slightly faster than light the neutrinos would have to be shot from a long, long way away. Getting a long, long, way away takes time and is only useful if you are traveling close enough to the speed of light that on the return trip the neutrinos gain more time than what you spent travelling. Eliezer_2831 would end up on the other side of the universe somewhere and the energy required to shoot enough neutrinos to communicate over that much distance would be enormous. The scope puts me in mind of the Tenth Doctor: “And it takes a lot of power to send this projection— I’m in orbit around a supernova. [smiling weakly] I’m burning up a sun just to say goodbye.”
I’m not sure if that scenario is more or less difficult than the remote neutrino manufacturing scenario. The engineering doesn’t sound easy but once it is done once any time before heat death of the universe you just win. You can send anything back to (almost) any time.
Unless you’re fighting Photino Birds.
But that’s pretty unlikely, yeah.
That sounds like it’s a reference to something awesome. Is it?
Fairly awesome, I’d say.
In the context of almost every proposed causality violation mechanism I’ve seen seriously discussed, it really is can’t, not won’t. Wormholes aren’t the only example. Tipler Cylinders for example don’t allow time travel prior to the point when they started rotating. Godel’s rotating universe has similar restrictions. Is there some time travel proposal I’m missing?
I agree that when considering anthropic issues won’t becomes potentially relevant if we had any idea that time travel could potentially allow travel prior to the existence of the device in question. In that case, I’d actually argue in the other direction: if such machines could exist, I’d expect to see massive signs of such interference in the past.
There are plenty of mechanisms in which can’t applies. There are others which don’t have that limitation. I don’t even want to touch what qualifies as ‘seriously discussed’. I’m really not up to date with which kinds of time travel are high status.
Ignore status issues. Instead focus on time travel mechanisms that don’t violate SR. Are there any such mechanisms which allow such violation before the time travel device has been constructed? I’m not aware of any.
Alcubierre drives.