Additional details weigh down hypotheses. The more steps your argument requires, the less probable it’s correct. None of your 15 things seem obviously objectionable to me, but any argument that requires 15 steps to be correct is unlikely.
I’ve been a bit concerned about Peak Oil for a while now and, as with many politically hot topics, it’s hard to find objective facts or rational analyses in the din of Greens and Blues screaming at each other. I very much appreciate you (and others!) stepping in to point out what a rational analysis requires rather than specifically arguing for one side versus another.
I disagree with the second comment on pragmatic grounds, though:
I doubt the marginal utility of you specifically worrying about it is greater than the marginal utility of worrying about FAI, or any other neglected existential risk.
I’m not aware that the Peak Oil debate has really been resolved for rationalists. That makes it a neglected existential risk (specifically a “crunch”) in my mind.
If Peak Oil really turns out to be a problem, there won’t be an FAI crisis. There won’t be enough power to run any computer that can go FOOM. Even if there’s some possible alternative energy source that might be able to replace oil in theory, part of the concern with Peak Oil is that we’ve already run out of time to replace the energy infrastructure. I honestly don’t know if this is true, but I don’t know that it’s false either and I’m not aware of anyone who has conclusively answered this concern from a rational point of view.
The main concern of Peak Oil doesn’t need so many hypotheses to point out a source of concern (noting that I don’t know how true each of these points are):
The production pattern of individual oil fields, and the rate of oil field discovery, makes it reasonable to conclude that there should be a global peak in oil production (or maybe a bumpy plateau, but there should still be a decline at the other end wherein we have basically exhausted usable oil).
We either have passed that midpoint or will do so very soon (i.e., within the next few years). (Note: Most people who are concerned about this seem to think we already passed it around 2005 to 2007.)
Practically everything in our modern world is driven by oil—sometimes literally (e.g., gas-powered vehicles), but in many ways not (e.g., fertilizer, lubricant for machines). Were oil to vanish fairly abruptly, some Very Bad Things would happen (akin to mass starvation, places like LA and Las Vegas becoming totally unlivable, breakdowns in transportation and communication across long distances, etc.).
Oil is so energy-dense that nearly all alternatives fall short of being realistic replacements. Most of them don’t produce energy quickly enough, and even if there are some that do it’s still debatable whether we have enough energy left to replace our energy infrastructure to maintain our current level of civilization.
Our main source of electricity is coal rather than oil, but coal is going through a similar peaking problem. (Note: Many people screaming about Peak Oil miss this point.)
Again, I don’t know if any of these points are true. But they strike me as individually reasonable, and together they hint at a crunch. I think many of DavidPlumpton’s other points are perhaps unnecessarily specific (e.g., the nature of the resulting food crisis) in the same way that specifying a paperclip-maximizer (as opposed to, say, a GAI Roomba) is unnecessarily detailed to point out the FAI problem. The point is that if Peak Oil is a real concern, then by some mechanism our heavy reliance on vast amounts of energy is going to cause a general collapse in civilization from which we may very well not recover for several million years.
We might not agree with whether DavidPlumpton’s presentation of the problem (or mine, for that matter) is the most rational. However, I do think that it’s important to recognize this as a viable existential risk, and at least from where I stand there does seem to be reason to take it at least as seriously as we do the FAI problem. At the very least, I think it’s worthwhile for us to rationally work out at least the rough likelihood of Peak Oil being a problem since we’re one of the painfully few communities who are likely to do so (as most others are obsessed with just shouting out applause lights on this topic).
I agree that a data-driven discussion on Peak Oil from a rationalist perspective would be a very good thing, although I don’t think the data supports your concerns (see here and here).
If I did a lot more research and spoke to some people I know in the energy forecasting business, would an article about peak oil as existential risk be something people would want on LessWrong?
I agree that a data-driven discussion on Peak Oil from a rationalist perspective would be a very good thing, although I don’t think the data supports your concerns (see here and here).
I actually don’t know what the data say. I’ve encountered gobs of blatantly contradictory information about this topic, much of which claims to be objective data. Most of the debate about whether Peak Oil is really a problem seems to come down to name-calling. I’ve personally found truth to be insanely opaque when a debate reaches that point.
That said, I would love it if the concerns I mentioned turned out to be balderdash! But that makes me inclined to be a little extra-careful of evidence that just happens to make my wishes seem more true.
If I did a lot more research and spoke to some people I know in the energy forecasting business, would an article about peak oil as existential risk be something people would want on LessWrong?
Upvoted for this comment:
I’ve been a bit concerned about Peak Oil for a while now and, as with many politically hot topics, it’s hard to find objective facts or rational analyses in the din of Greens and Blues screaming at each other. I very much appreciate you (and others!) stepping in to point out what a rational analysis requires rather than specifically arguing for one side versus another.
I disagree with the second comment on pragmatic grounds, though:
I’m not aware that the Peak Oil debate has really been resolved for rationalists. That makes it a neglected existential risk (specifically a “crunch”) in my mind.
If Peak Oil really turns out to be a problem, there won’t be an FAI crisis. There won’t be enough power to run any computer that can go FOOM. Even if there’s some possible alternative energy source that might be able to replace oil in theory, part of the concern with Peak Oil is that we’ve already run out of time to replace the energy infrastructure. I honestly don’t know if this is true, but I don’t know that it’s false either and I’m not aware of anyone who has conclusively answered this concern from a rational point of view.
The main concern of Peak Oil doesn’t need so many hypotheses to point out a source of concern (noting that I don’t know how true each of these points are):
The production pattern of individual oil fields, and the rate of oil field discovery, makes it reasonable to conclude that there should be a global peak in oil production (or maybe a bumpy plateau, but there should still be a decline at the other end wherein we have basically exhausted usable oil).
We either have passed that midpoint or will do so very soon (i.e., within the next few years). (Note: Most people who are concerned about this seem to think we already passed it around 2005 to 2007.)
Practically everything in our modern world is driven by oil—sometimes literally (e.g., gas-powered vehicles), but in many ways not (e.g., fertilizer, lubricant for machines). Were oil to vanish fairly abruptly, some Very Bad Things would happen (akin to mass starvation, places like LA and Las Vegas becoming totally unlivable, breakdowns in transportation and communication across long distances, etc.).
Oil is so energy-dense that nearly all alternatives fall short of being realistic replacements. Most of them don’t produce energy quickly enough, and even if there are some that do it’s still debatable whether we have enough energy left to replace our energy infrastructure to maintain our current level of civilization.
Our main source of electricity is coal rather than oil, but coal is going through a similar peaking problem. (Note: Many people screaming about Peak Oil miss this point.)
Again, I don’t know if any of these points are true. But they strike me as individually reasonable, and together they hint at a crunch. I think many of DavidPlumpton’s other points are perhaps unnecessarily specific (e.g., the nature of the resulting food crisis) in the same way that specifying a paperclip-maximizer (as opposed to, say, a GAI Roomba) is unnecessarily detailed to point out the FAI problem. The point is that if Peak Oil is a real concern, then by some mechanism our heavy reliance on vast amounts of energy is going to cause a general collapse in civilization from which we may very well not recover for several million years.
We might not agree with whether DavidPlumpton’s presentation of the problem (or mine, for that matter) is the most rational. However, I do think that it’s important to recognize this as a viable existential risk, and at least from where I stand there does seem to be reason to take it at least as seriously as we do the FAI problem. At the very least, I think it’s worthwhile for us to rationally work out at least the rough likelihood of Peak Oil being a problem since we’re one of the painfully few communities who are likely to do so (as most others are obsessed with just shouting out applause lights on this topic).
I agree that a data-driven discussion on Peak Oil from a rationalist perspective would be a very good thing, although I don’t think the data supports your concerns (see here and here). If I did a lot more research and spoke to some people I know in the energy forecasting business, would an article about peak oil as existential risk be something people would want on LessWrong?
I actually don’t know what the data say. I’ve encountered gobs of blatantly contradictory information about this topic, much of which claims to be objective data. Most of the debate about whether Peak Oil is really a problem seems to come down to name-calling. I’ve personally found truth to be insanely opaque when a debate reaches that point.
That said, I would love it if the concerns I mentioned turned out to be balderdash! But that makes me inclined to be a little extra-careful of evidence that just happens to make my wishes seem more true.
Yes! I would love that. Please!
I think this was downvoted. Would whoever downvoted this please explain why?