Are we agreed that the rapid rise in CO2 levels, to highs not seen in human history and owing to human intervention, is undisputed fact?
If so, it seems to me that the default extrapolation, from our everyday experience with systems we understand poorly, is that when you turn a dial all the way up without knowing what the heck you’re doing, you won’t like the results. Example include: numerous cases of introducing animal species (bacteria, sheep, wasps) to populations not adapted to them, said populations then suffering upheaval; stock market crashes; losing two space shuttles; and so on.
The burden of proof seems to be on those who insist that yeah, CO2 levels are rising super fast, but don’t worry, it’ll be business as usual (except winters will be nicer and summers will need a little more ice cubes).
Wha...? Is that an argument by surface analogy? Does every increase in every value owing to human intervention lead to a catastrophe? How about internet connectivity? Land committed to agriculture? Air respired by humans? Shoes built? Radio waves transmitted?
How do you even measure the reference classes appropriately?
For some of these examples, yes, there are catastrophic scenarios on record.
Overgrazing in Iceland to name one I’ve seen first-hand. Beaches despoiled by lethal greeen algae in France as a result of intensive pig farming is another. Shoes—that’s perhaps an excessively restricted category, but the Pacific Trash Vortex is one consequence of turning the dial up on manufacturing capacity without adequate control of the consequences. Improved Internet connectivity is having demonstrated, large and undesired effects on industries such as entertainment and newspapers.
Radio waves… no, offhand I can’t think of an issue on record with those, unless EMF sensitivity counts—but I would be hugely surprised if that turned out to be real (i.e. not psychogenic; the discomfort could be real).
You mentioned “failed predictions”, but left those unspecified. OK, here is a list of empirical confirmations of positive feedback loops involving CO2. Arctic ice melt is the one I’d lose sleep over, since the methane sequestered in Arctic ice is a much more powerful greenhouse gas than CO2. Ice melt also has an effect on water salinity which indirectly affects thermohaline circulation.
The causal details of how some of these positive feedbacks could bring about deeply undesirable consequences seem to me to be better established than the details of how runaway AI could lead to the destruction of human values. But I may have more to learn about either.
This isn’t analogy, as in “build something that looks like a bird and it will fly”. More like abstracting away from examples in several categories, to “systems that remain stable tend to be characterized by feedback loops, including both negative feedback (such as the governor) for regulation and positive for growth or excitation”. The latter leads to predictions, e.g. if you observe only one type of feedback in a stable system a search for the other type will generally be fruitful.
For instance, we observe that successful community Web sites tend to become even more successful as enthusiast users take the good news outside. Yet very few sites become very big. We can look for regulatory feedback loops. A good one stems from the joke “Nobody goes to that restaurant anymore, it’s too crowded.” As the audience of a community site increases, its output may become difficult to handle, turning people away feeling overwhelmed. I would predict that LW will run out of new commenters before it runs out of readers, that a lowered influx of new commenters leads to staleness in the contributions of post authors, in turn leading post authors to look elsewhere for stimulation.
Now, perhaps CO2 levels rising through the roof aren’t going to do anything bad. But that’s as much an argument as saying “perhaps I will win the lottery”.
For some of these examples, yes, there are catastrophic scenarios on record.
Overgrazing in Iceland to name one I’ve seen first-hand. Beaches despoiled by lethal greeen algae in France as a result of intensive pig farming is another. Shoes—that’s perhaps an excessively restricted category, but the Pacific Trash Vortex is one consequence of turning the dial up on manufacturing capacity without adequate control of the consequences. Improved Internet connectivity is having demonstrated, large and undesired effects on industries such as entertainment and newspapers.
This raises the issue of what exactly people mean by ‘catastrophic’. None of the examples you give are ‘catastrophic’ on anything like the scale of what some prophesize for global warming. I personally think it is a misuse of the word catastrophe to apply it to the situations you describe. If global warming was only forecast to cause problems on that sort of scale then I don’t think anyone would be seriously contemplating the kinds of measures often advocated to mitigate the risk.
The effects of improved Internet connectivity are having large positive effects on the entertainment industries and newspapers from the perspective of most people who aren’t incumbents in those industries, just as technological progress generally benefits societies as a whole while sometimes reducing the income of groups who made their living from the supplanted technologies that preceded them.
None of the examples you give are ‘catastrophic’ on anything like the scale of what some prophesize for global warming.
That’s because you’re cherry-picking. Having the Gulf Stream stop, one of the possible consequences of Arctic Ice melt, would be very unpleasant.
In other cases the effects we’re seeing are only the start of a chain of effects. The Pacific Trash Vortex is basically us dumping tiny plastic particles into our own food chain, ultimately poisoning ourselves. It’s bad in itself, but the knock-on effects will be worse. Sure, it still pales in comparison to some predicted AGW effects: that’s why the latter has become the more pressing issue.
These examples were direct responses to Silas, who meant to ridicule the initial instances I gave of the class bad things happening as a result of pushing too hard the parameters of systems we understand poorly, on various scales. Many of his own suggestions turn out not to be ridiculous at all, but rather serious matters.
Having the Gulf Stream stop, one of the possible consequences of Arctic Ice melt, would be very unpleasant.
The Gulf Stream makes the difference between Europe and the west coast of North America, not east coasts. Maybe it would be unpleasant, but a catastrophe?
I’ve heard claims that the gulf stream switching off would cause Britain to undergo a climate change that would have consequences I would call ‘catastrophic’, at least in the short term. Some predictions talk about average temperatures dropping by 5-8 C in a matter of months which would have severe consequences for British agriculture and would likely have a noticeable impact on GDP. I’m not sure I put much faith in those predictions however.
This would also be a catastrophe on a different scale from the more alarmist AGW predictions. We’re talking about a major disruption to the British economy but not an existential threat to the human race.
That’s because you’re cherry-picking. Having the Gulf Stream stop, one of the possible consequences of Arctic Ice melt, would be very unpleasant.
I thought you were using that as an example of a potential catastrophic effect of global warming, whereas I was saying none of your examples of things that have actually happened are what I would call catastrophic. I have heard some predictions of what might happen to the climate in Britain if arctic ice melt caused the gulf stream to stop and if those predictions were to pan out then I think ‘catastrophic’ would be an appropriate word to use for the consequences for Britain.
I don’t disagree that some of the predictions for the consequences of AGW are situations for which the word ‘catastrophic’ is appropriate. My point is that some of these predictions are an entirely different scale of disaster from anything you’ve given as an example of actual consequences of human activity to date. The Pacific Trash Vortex cannot reasonably be described as ‘catastrophic’ in my opinion, though dire predictions may exist that if they transpire might justify such language.
Based on the voting patterns, I’m going astray somewhere. We don’t seem to disagree on the facts (high CO2 levels, past environmental damage) and I’m not seeing arguments directed at my reasoning, beyond the criticism of “surface analogy” that I’ve done my best to adress. So I’ll let this be my final comment on the topic, and hope to find insight in others’ discussion.
We quite agree there hasn’t yet been a catastrophe on the scale predicted for AGW: we wouldn’t be having this conversation if there had been. If you read the original post all over again, you’ll find that was its entire point. Don’t demand that particular proof.
The Pacific Trash Vortex cannot reasonably be described as ‘catastrophic’ in my opinion
We don’t want to play dictionary games with the word “catastrophe”. One constructive proposal would be to consider the cost to our economies of cleaning up one or the other of these environmental impacts—including their knock-on effects—versus the costs of prevention. We haven’t incurred the costs of the Trash Vortex yet, it’s not making itself felt to you; but it’s nevertheless a fact not a prediction, and we can base estimates on it.
The typical cost of cleaning up an oil spill seems to be on the order of $10M per ton. The Pacific garbage patch may contain as much as 100 million tons of plastic debris. As an order of magnitude estimate, one Trash Vortex appears to be worth one subprime crisis, albeit spread out over a longer period.
We’re clearly in Black Swan territory, and yet this is just one example picked almost at random (in fact, picked from what Silas took to be counterexamples).
I’m not seeing arguments directed at my reasoning, beyond the criticism of “surface analogy” that I’ve done my best to adress.
Ok, I’ll try and make it more explicit. Your reasoning seems to be that our experience with complex systems that we don’t fully understand is that disrupting them has bad unintended consequences and therefore the burden of proof is on those who suggest that we don’t need to take drastic action to reduce CO2 levels.
I don’t think your conclusion follows from your premise because it seems to me that there are no examples of bad unintended consequences that we haven’t been able to deal with without paying an excessive cost and few examples of bad unintended consequences that even end up with a negative overall economic cost. The only reasonable argument for adopting the kind of drastic and hugely expensive measures necessary to significantly reduce CO2 levels is that the potential effects are so catastrophic that we can’t afford to risk them. There are no examples of similar situations in the past, though as you rightly point out that is not strong evidence that such situations cannot happen since we might not be around to discuss the issue if they had. On the other hand there are lots of examples of dire/catastrophic predictions that have failed to pan out, although in some cases mitigating action has been taken that means we haven’t had the control experiment of doing nothing.
It seems to me that the burden of proof is still very much on those who argue we must take very economically costly actions now because unlike previous problems which have turned out to be relatively cheap to deal with this problem poses a significant risk of genuine catastrophe.
One constructive proposal would be to consider the cost to our economies of cleaning up one or the other of these environmental impacts—including their knock-on effects—versus the costs of prevention. We haven’t incurred the costs of the Trash Vortex yet, it’s not making itself felt to you; but it’s nevertheless a fact not a prediction, and we can base estimates on it.
It’s also important to consider the cost of doing nothing and dealing with the consequences. The trash vortex is a problematic example to use here because there have not been any significant bad consequences yet. It may be a fact that it exists but I haven’t found any estimates of the economic cost it is imposing right now and only vague warnings of possible higher pollutant levels in future.
If the cost of doing nothing about CO2 levels were similar to the cost we appear to be paying for doing nothing about the Pacific Trash Vortex then it would be a no brainer to do nothing about CO2 levels.
Ah that particular idea of all human pleasures being harmful for the environment is pretty much religious. It’s not at all what the impact is like.
Computing is basically blameless in the direct sense for global warming. We should probably enjoy it as much as possible. Electricity is good. Trains are good. Holidaying is good.
Airconditioning is bad. Air travel is bad. Short product lifetime is bad.
The situation is far more positive than some make it out to be. Even the direst climate change predictions necessitates drastic changes in some aspects of life.
AGW can’t take away modern medicine or virtual reality from you.
Why do you think “harmful for the environment” means “leading to global warming”? Lots of things are harmful for the environment. Drying swamps to make railroads harm it. Holidaying leads to decreased “old habitat” biodiversity. Building power plants on small mountain rivers leads to decreased biodiversity, too. Yes, these things are good for us. It just has no bearing on whether they are good for nature.
Are we agreed that the rapid rise in CO2 levels, to highs not seen in human history and owing to human intervention, is undisputed fact?
If so, it seems to me that the default extrapolation, from our everyday experience with systems we understand poorly, is that when you turn a dial all the way up without knowing what the heck you’re doing, you won’t like the results. Example include: numerous cases of introducing animal species (bacteria, sheep, wasps) to populations not adapted to them, said populations then suffering upheaval; stock market crashes; losing two space shuttles; and so on.
The burden of proof seems to be on those who insist that yeah, CO2 levels are rising super fast, but don’t worry, it’ll be business as usual (except winters will be nicer and summers will need a little more ice cubes).
Wha...? Is that an argument by surface analogy? Does every increase in every value owing to human intervention lead to a catastrophe? How about internet connectivity? Land committed to agriculture? Air respired by humans? Shoes built? Radio waves transmitted?
How do you even measure the reference classes appropriately?
For some of these examples, yes, there are catastrophic scenarios on record.
Overgrazing in Iceland to name one I’ve seen first-hand. Beaches despoiled by lethal greeen algae in France as a result of intensive pig farming is another. Shoes—that’s perhaps an excessively restricted category, but the Pacific Trash Vortex is one consequence of turning the dial up on manufacturing capacity without adequate control of the consequences. Improved Internet connectivity is having demonstrated, large and undesired effects on industries such as entertainment and newspapers.
Radio waves… no, offhand I can’t think of an issue on record with those, unless EMF sensitivity counts—but I would be hugely surprised if that turned out to be real (i.e. not psychogenic; the discomfort could be real).
You mentioned “failed predictions”, but left those unspecified. OK, here is a list of empirical confirmations of positive feedback loops involving CO2. Arctic ice melt is the one I’d lose sleep over, since the methane sequestered in Arctic ice is a much more powerful greenhouse gas than CO2. Ice melt also has an effect on water salinity which indirectly affects thermohaline circulation.
The causal details of how some of these positive feedbacks could bring about deeply undesirable consequences seem to me to be better established than the details of how runaway AI could lead to the destruction of human values. But I may have more to learn about either.
This isn’t analogy, as in “build something that looks like a bird and it will fly”. More like abstracting away from examples in several categories, to “systems that remain stable tend to be characterized by feedback loops, including both negative feedback (such as the governor) for regulation and positive for growth or excitation”. The latter leads to predictions, e.g. if you observe only one type of feedback in a stable system a search for the other type will generally be fruitful.
For instance, we observe that successful community Web sites tend to become even more successful as enthusiast users take the good news outside. Yet very few sites become very big. We can look for regulatory feedback loops. A good one stems from the joke “Nobody goes to that restaurant anymore, it’s too crowded.” As the audience of a community site increases, its output may become difficult to handle, turning people away feeling overwhelmed. I would predict that LW will run out of new commenters before it runs out of readers, that a lowered influx of new commenters leads to staleness in the contributions of post authors, in turn leading post authors to look elsewhere for stimulation.
Now, perhaps CO2 levels rising through the roof aren’t going to do anything bad. But that’s as much an argument as saying “perhaps I will win the lottery”.
This raises the issue of what exactly people mean by ‘catastrophic’. None of the examples you give are ‘catastrophic’ on anything like the scale of what some prophesize for global warming. I personally think it is a misuse of the word catastrophe to apply it to the situations you describe. If global warming was only forecast to cause problems on that sort of scale then I don’t think anyone would be seriously contemplating the kinds of measures often advocated to mitigate the risk.
The effects of improved Internet connectivity are having large positive effects on the entertainment industries and newspapers from the perspective of most people who aren’t incumbents in those industries, just as technological progress generally benefits societies as a whole while sometimes reducing the income of groups who made their living from the supplanted technologies that preceded them.
That’s because you’re cherry-picking. Having the Gulf Stream stop, one of the possible consequences of Arctic Ice melt, would be very unpleasant.
In other cases the effects we’re seeing are only the start of a chain of effects. The Pacific Trash Vortex is basically us dumping tiny plastic particles into our own food chain, ultimately poisoning ourselves. It’s bad in itself, but the knock-on effects will be worse. Sure, it still pales in comparison to some predicted AGW effects: that’s why the latter has become the more pressing issue.
These examples were direct responses to Silas, who meant to ridicule the initial instances I gave of the class bad things happening as a result of pushing too hard the parameters of systems we understand poorly, on various scales. Many of his own suggestions turn out not to be ridiculous at all, but rather serious matters.
The Gulf Stream makes the difference between Europe and the west coast of North America, not east coasts. Maybe it would be unpleasant, but a catastrophe?
I’ve heard claims that the gulf stream switching off would cause Britain to undergo a climate change that would have consequences I would call ‘catastrophic’, at least in the short term. Some predictions talk about average temperatures dropping by 5-8 C in a matter of months which would have severe consequences for British agriculture and would likely have a noticeable impact on GDP. I’m not sure I put much faith in those predictions however.
This would also be a catastrophe on a different scale from the more alarmist AGW predictions. We’re talking about a major disruption to the British economy but not an existential threat to the human race.
I thought you were using that as an example of a potential catastrophic effect of global warming, whereas I was saying none of your examples of things that have actually happened are what I would call catastrophic. I have heard some predictions of what might happen to the climate in Britain if arctic ice melt caused the gulf stream to stop and if those predictions were to pan out then I think ‘catastrophic’ would be an appropriate word to use for the consequences for Britain.
I don’t disagree that some of the predictions for the consequences of AGW are situations for which the word ‘catastrophic’ is appropriate. My point is that some of these predictions are an entirely different scale of disaster from anything you’ve given as an example of actual consequences of human activity to date. The Pacific Trash Vortex cannot reasonably be described as ‘catastrophic’ in my opinion, though dire predictions may exist that if they transpire might justify such language.
Based on the voting patterns, I’m going astray somewhere. We don’t seem to disagree on the facts (high CO2 levels, past environmental damage) and I’m not seeing arguments directed at my reasoning, beyond the criticism of “surface analogy” that I’ve done my best to adress. So I’ll let this be my final comment on the topic, and hope to find insight in others’ discussion.
We quite agree there hasn’t yet been a catastrophe on the scale predicted for AGW: we wouldn’t be having this conversation if there had been. If you read the original post all over again, you’ll find that was its entire point. Don’t demand that particular proof.
We don’t want to play dictionary games with the word “catastrophe”. One constructive proposal would be to consider the cost to our economies of cleaning up one or the other of these environmental impacts—including their knock-on effects—versus the costs of prevention. We haven’t incurred the costs of the Trash Vortex yet, it’s not making itself felt to you; but it’s nevertheless a fact not a prediction, and we can base estimates on it.
The typical cost of cleaning up an oil spill seems to be on the order of $10M per ton. The Pacific garbage patch may contain as much as 100 million tons of plastic debris. As an order of magnitude estimate, one Trash Vortex appears to be worth one subprime crisis, albeit spread out over a longer period.
We’re clearly in Black Swan territory, and yet this is just one example picked almost at random (in fact, picked from what Silas took to be counterexamples).
Ok, I’ll try and make it more explicit. Your reasoning seems to be that our experience with complex systems that we don’t fully understand is that disrupting them has bad unintended consequences and therefore the burden of proof is on those who suggest that we don’t need to take drastic action to reduce CO2 levels.
I don’t think your conclusion follows from your premise because it seems to me that there are no examples of bad unintended consequences that we haven’t been able to deal with without paying an excessive cost and few examples of bad unintended consequences that even end up with a negative overall economic cost. The only reasonable argument for adopting the kind of drastic and hugely expensive measures necessary to significantly reduce CO2 levels is that the potential effects are so catastrophic that we can’t afford to risk them. There are no examples of similar situations in the past, though as you rightly point out that is not strong evidence that such situations cannot happen since we might not be around to discuss the issue if they had. On the other hand there are lots of examples of dire/catastrophic predictions that have failed to pan out, although in some cases mitigating action has been taken that means we haven’t had the control experiment of doing nothing.
It seems to me that the burden of proof is still very much on those who argue we must take very economically costly actions now because unlike previous problems which have turned out to be relatively cheap to deal with this problem poses a significant risk of genuine catastrophe.
It’s also important to consider the cost of doing nothing and dealing with the consequences. The trash vortex is a problematic example to use here because there have not been any significant bad consequences yet. It may be a fact that it exists but I haven’t found any estimates of the economic cost it is imposing right now and only vague warnings of possible higher pollutant levels in future.
If the cost of doing nothing about CO2 levels were similar to the cost we appear to be paying for doing nothing about the Pacific Trash Vortex then it would be a no brainer to do nothing about CO2 levels.
Ah that particular idea of all human pleasures being harmful for the environment is pretty much religious. It’s not at all what the impact is like.
Computing is basically blameless in the direct sense for global warming. We should probably enjoy it as much as possible. Electricity is good. Trains are good. Holidaying is good.
Airconditioning is bad. Air travel is bad. Short product lifetime is bad.
The situation is far more positive than some make it out to be. Even the direst climate change predictions necessitates drastic changes in some aspects of life.
AGW can’t take away modern medicine or virtual reality from you.
Why do you think “harmful for the environment” means “leading to global warming”? Lots of things are harmful for the environment. Drying swamps to make railroads harm it. Holidaying leads to decreased “old habitat” biodiversity. Building power plants on small mountain rivers leads to decreased biodiversity, too. Yes, these things are good for us. It just has no bearing on whether they are good for nature.
My favorite one: burning wood for heat. Better than fossil fuels for the GW problem, but really bad for local air quality.
Of course, “leading to global warming” is a subset of “harmful for the environment”. Agreed on all counts.
Computing can’t harm the environment in any way—it’s within a totally artificial human space.
The others (“good”) can harm the environment in general, but are much better for AGW.
Well...