....the 100m deaths are additional to the normal death rate? Therefore, an increase of ~10%?

Last edited Thu May 22, 2014, 07:26 AM - Edit history (1)

They are not falsifiable and therefore not science.

Consider, for example, the scientific consensus that air pollution is one cause of the increase in childhood asthma. By your logic, it's "not science" for anyone to attempt to estimate the number of increased asthma cases that would result from a particular action, because it could never be conclusively falsified.

It's one thing to bear in mind the limits of studies about complex matters. It's quite another to say that, because no such study can be conclusively established to be true, we might as well just throw up our hands and build the smelting plant upwind of the elementary school. After all, we can't determine the number of asthma cases we'll cause, so we're entitled to take the action that would make sense if the number were zero. That conclusion simply doesn't follow.

You can prove that an increase in air pollution increases the rates of childhood asthma by comparing childhood asthma rates in areas with differing amounts of pollution. Global warming impacts cannot be separated out like that because, well, the effects are 'global'.

You write, "You can prove that an increase in air pollution increases the rates of childhood asthma by comparing childhood asthma rates in areas with differing amounts of pollution." No, people who want to disagree can argue that it's coincidence or there are confounding factors. Correlation by itself doesn't prove causation. And, of course, there usually are confounding factors, so any figure that someone comes up with for the increased number of asthma cases caused by a particular source of pollution will be just an estimate.

You write, "Global warming impacts cannot be separated out like that because, well, the effects are 'global'." The linked article notes that effects will be greater in the world's poorest nations. One obvious point is that sea level rise will affect different areas differently. Bangladesh is more vulnerable than Switzerland. But Bangladesh has other problems, too, so if the death rate goes up more there than in Switzerland, climate change deniers will still refuse to believe that anthropogenic global warming had anything to do with it.

The fact is that scientists in many fields routinely make estimates based on data and analysis but not subject to conclusive proof or disproof. Climate scientists create models of the Earth's climate and try to improve them, without ever imagining that a model of so complex a system will be perfect.

You can prove that the impacts of rising seas will hit Bangladesh harder than Switzerland. This proof of course, assumes that sea levels will rise by some catastrophic amount. What is funny is that the complete lack of empirical evidence that climate change will be catastrophic doesn't seem to phase anyone around here. Yes, climate change will have some serious effects, but around here you can read posts about how it will result in the extinction of the human race--a position that cannot be found in peer review literature. In EE there is nothing worse than even slightly underestimating the effects of climate change, but no limit to the most absurd exaggerations imaginable.

Especially in light of estimates of the fraction of world GDP that some have estimated would be required to mitigate the effects of adding "fossil carbon" to the atmosphere - much smaller than 3.2%.

Of course, part of the problem is using stupid measures like GDP to assess how well economies function in the first place.

Mainly, these estimates fail to take into account the growing fragility of the world's financial system.

The problem with all of these prognostications (not just the economic ones) is that all the interlocked systems they try to assess are fraught with non-linear behavior, sudden excursions and tipping points. Add to that the fact that the world system (the whole thing, from natural processes and biota up to human societies and their myriad systems) is a panarchy. As a result there's really no way to apply linear models to one small portion of it and say anything conclusive about larger portions of it, let alone the system as a whole.

My general belief is that the world system has entered a chaotic region on a number of fronts simultaneously (e.g. climate, economy, food supplies and energy supplies) at the same time, and the change to a new region of stability is likely to be sudden and irreversible. The new region is guaranteed by thermodynamics to be lower in overall potential than the current one, so the shift is likely to be very damaging to some system elements (like people and their systems).

The progressively worsening news reports on the world's climatic, social, political and economic developments seems to support this admittedly pessimistic view.

How soon will the shit hit the fan? Or to put it another way, how many years have to pass without incident for you to admit you were wrong?

I have no way of knowing when or how such a fragile, complex, thermodynamically expanding system will rupture. I just know it will, because such systems always do, sooner or later.

I'm not trying to make you a believer, and I'm not attached to being "right". I just have my own views. If you don't share them, that's OK with me.

When you set a date, you are making a firm prediction that offers up a scenario where you have to admit you were wrong. Since you claim you are not attached to being "right", you should not have a problem with that.

Steve Hallett calls it for sometime in the 2030s. Guy McPherson thinks he's an optimist and we'll be extinct before that decade ends; presumably that means the whole economic house of cards falls much sooner than that!

Not that it would be fair to hold GG to these timetables. And in any event, it's a bit of a pointless exercise, for suppose he says something like "everything collapses in 2025." If the actual data of a collapse is 2030 or 2035 I'd say the prediction was awfully good (what difference does a decade or two make if something that cataclysmic happens)? Yet in 2026 you'd be saying the prediction was flat-out wrong (and you would; the whole idea is that this is something like a phase change, where there's a sudden discontinuous change, not some gradual slide).

I think there's plenty of room to challenge the rigor of his analyses, but this particular flavor of "put up or shut up" challenge misses the point entirely. If he's ever proven right in the way you demand, we won't be in a position to have any further discussion at all!

I am of the understanding that the severe rise in temperatures being predicted will happen over the course of decades, not weeks. As a result, I think you can look at where things are going and make reasonable judgements. For example, Guy McPherson claims that temperatures will rise 2-4 degrees C over the next two decades. Now, if ten years go by and temperatures have risen by 0.3 degrees or less, I think it would be fair to say at that point that he was wrong. In other words, for the "doomers" to be taken seriously, things have to be moving in the direction of doom.

I don't think physics is on McPherson's side on some of what he predicts. Biologists I know are also pretty skeptical of his arguments predicting extinction.

But more generally, when you have a lot of critical indicators looking bad at the same time, it's hard to tell which one is most likely to go catastrophically wrong first. To focus entirely on things like surface temperature and sea level rise is to miss most of the picture. There's a whole witch's brew of population growth, industrial growth, degradation of water resources, ocean acidification, etc. and only a small subset of those could suffice to bring down civilization. It would be easier to pooh-pooh doomsaying if we had more of a track record of acting decisively against these threats as they are recognized, but we don't. So while one way of being wrong is to say civilization collapses in 2035 and have that not happen, a far worse way of being wrong is to conclude in 2036 that there was never anything to worry about in the first place and have everything fall apart a few decades later out of complacency.

I claim (as do many systems scientists) that it is impossible to predict the place or timing of significant ruptures in a system this complex. Because of that, it would be hubristic for me to predict a date and time, and humble for me to decline.

Here is a four-page paper that outlines very clearly (in 18 very short paragraphs) why I take the position I do on the world system, and why I decline to make predictions of failure.

How Complex Systems Fail (PDF)

You are attempting to play a game of gotcha, which I also decline. It would only be ego and hubris that would make me rise to such bait.

Since you refuse to make predictions, I have to ask if there is any empirical event that would make you take a step back and admit that you were wrong. Just to be fair, I will give you my breaking point: if at any point in time after 2025 temperatures rise to where the computer models say they will be I will admit that I was wrong.

As I've pointed out, the complex system of global industrial civilization is showing a lot of potential failure points. Just as its complexity prevents me from determining a particular time of failure, it also precludes me from predicting where or how the failure will happen. Civ probably will not collapse if just one of its component systems breaks down, and predicting what convergence of failures would cause a general collapse is even more impossible.

In the past I've admitted that I was wrong about nuclear power being an acceptable source of low-carbon energy, as well as admitting that I under-estimated the potential of wind power, and about Fukushima snapping the global supply chains that run through Japan. But my concern isn't about whether or not this or that aspect of the system breaks down. As the paper I linked above made clear, complex systems always suffer failures. The question is whether those failures might converge to cause an irremediable system-wide cascading failure leading to collapse.

Which leaves me with the question of why you care if I admit to being "wrong"? Why is my behavior that important to you?

Why should I care if you admit to being wrong? The simple answer is that I don't really. What I do care about is the fact that you post quite a bit around here, and between DU and your blog a fair number of people read what you write. As a result, I think it is important for those people to have a metric by which they can judge if what you say is true. By failing to offer any objective metrics or empirical scenarios which would prove you wrong, you are declaring yourself as an authority. In my opinion, that is a giant 500 year step back to a time when humanity believed truth was determined by the credentials of the speaker rather than the degree to which what they were saying agreed with reality. The idea of falsifiability is crucial to science. If you do not present your ideas as being falsifiable, you are not talking science, but religion. That's fine, I just want everyone to know it.

I would put myself in the category of "natural philosopher". If I called myself a scientist, I'd be running experiments of one sort or another. Instead, I witness events, try to determine what the non-obvious drivers might be, what connections might exist between apparently disconnected events, and suggest potential courses of development that might result from the trends I see.

I don't think that "truth" is a valid concept, at least in any absolute sense. I think that people who base their worldview on some cnotion of absolute truth - scientific, religious or otherwise - are probably deluded.