Cool! I've written the following reply to Dinwar, which also covers a few of your points. I'll address a few more of them next post.
When faced with a very complex system, we can either declare prediction hopeless, or we can look for ways to analyze it that subsume the complexity in some manageable and reliable way.
For example, if I showed you an extremely complex perpetual motion machine with thousands of moving parts operating in chaotic ways, you would likely not be able to model or predict its motion. But, you could invoke the laws of thermodynamics and conclude that my machine will not produce more energy than it uses. I can protest that you're not properly taking into account the exceptional oscillations of part 3326-B, but you can confidently stand by your conclusion.
As another example, the human body contains a literally unimaginable number of "moving parts" interacting in ridiculously complex ways, both within individual cells and throughout the larger tissue and organ systems, including the brain's neural processing which even by itself defies comprehension. Faced with this, how can we make any predictions at all? Yet, I can predict with some confidence that if your breathing is entirely occluded for ten minutes, you'll die.
From the point of view of a single cell (that somehow achieved sentience), this might be much harder to perceive. "Yes, it's become more difficult to acquire oxygen than it was a minute or two ago, but there's still oxygen all around; the water molecules are full of it, and there's even some in the proteins. With so much going on, who's to say a solution won't be found? Maybe there's something even better than oxygen we should be using, and this will give us the kick we need to go find it." (From the same level, death might also be difficult to define. "Sure, the neurons have clearly died off, but the anaerobic bacteria are flourishing more than ever, and there are some exciting new kinds of cells showing up in the mucus membranes that seem to have very promising futures.")
There are two main lines of argument for collapse, roughly corresponding to the two analogies above. Both are stronger than they appear at first glance (simple incredulity is a barrier), but neither is airtight, as far as I've been able to tell.
The first argument focuses on energy. It begins with the historical correlation between energy use and wealth, economic growth, technology, and social complexity. While we might imagine exceptions, such as a comfortable low-effort lifestyle that also consumes little energy due to highly efficient technology, there are no historical examples of this. Instead, even increases in efficiency accompany increases in the overall use of the resource. For example, as cars become more energy efficient, there are also more of them so fuel use only increases. It should also be noted that while technology does increase efficiency, outside of a few exceptions (e.g. computation and communication), it increases it only by limited factors because of physical constraints of the task being performed. While computers were getting a million times faster, automobile gas milage increased about threefold. LEDs are up to twenty times as efficient as incandescent light bulbs, but even another twofold improvement beyond that is unlikely. Many productive uses of energy, including crucial ones like metal smelting, nitrogen fixing for fertilizer, refrigeration, and water and rail transport, are close to as energy efficient as they thermodynamically can be, and have been for some time.
Another reason to focus on energy is that it in theory energy can substitute for nearly all other resources, so focusing on it is the only way to make the question of sustainability a bounded one. If there were no constraints on energy, constraints on land, soil, water, food, space, materials, pollution, and even waste heat would become moot. With enough energy we could desalinate sea water and pump the fresh water to wherever it's needed, raise crops or rain forests in multi-story buildings, synthesize food or soil, extract whatever elements we need in pretty much unlimited amounts, pump heat to alter the climate as desired, and build all the machinery to do all those things.
Substitution the other way, turning other stuff into usable energy, isn't so easy, though. Energy extraction or acquisition has a unique constraint. If we were talking about, for example, gold, then no matter how dilute the remaining gold ores of the world become, there is always some price of gold that would make it worth obtaining. Even if that meant extracting it from sea water or, at the ultimate extreme, synthesizing it with particle accelerators. So, saying that the supply of gold (or tungsten, chromium, etc.) is limited or will run out really is inaccurate.
But extracting or otherwise producing energy does have such a limit. If the energy required to extract or exploit a source of energy exceeds or equals the amount of energy produced, there is no point in producing it. You cannot make a profit or accomplish anything useful by doing so, no matter how high the price or value of energy is.
So, the available supply of usable fossil fuels on earth is limited, and will run out. Before it runs out, the production rate will decrease and the price will rise. Since we're currently dependent on fossil fuel energy for our productivity, the economy will suffer. As economic recessions make it more difficult for both individuals and governments to build alternative sources like nuclear and renewables, growing recession as supplies tighten will continue to run ahead of any such efforts. Without a way to halt that cycle, slow collapse will result.
That is essentially the argument of the Peak Oil doomers, who cried wolf a few years ago. Their doom dates have come and gone, nothing happened, and no one wants to hear it any more.
(Don't forget how that fable ends, though. Eventually there is a wolf, no one heeds the warning, and all the sheep get eaten. Is that the script we're following?)
That's because there are some complications to the energy argument, which makes it less than airtight. The most important of those complications are "nonconventional" fossil fuels, coal, nuclear power, and renewable energy, especially solar.
Nonconventional fossil fuels--oil shale and oil sands--are what's currently making up the difference between still-increasing petroleum production (a little under 2% per year growth) and now-decreasing conventional crude oil production. The latter peaked on schedule, but the overall production has not peaked yet.
There are many indications, though, that nonconventional oil and gas extraction has not been profitable, even though as usual the easiest deposits are being exploited first. The oil price has been roughly flat since 2010 but producer costs have been going up across the board. This means that either the price of energy must rise again, or there will be an industry shake-out that will reduce the unprofitable production, tightening supply and causing prices to rise. (Heads I win, tails you lose.)
Historically, oil price spikes appear to have contributed to 10 out of the last 11 recessions in the U.S. So, recession again: the economy slows down, energy usage goes down, supply and demand go back into balance. The energy price could even drop again (as in 2008), but it won't drop below the marginal production cost. Thus, the fossil fuel supply becomes a regulator on economic growth.
Note, however, that the argument is now based on economics, no longer on physics. It's no longer airtight, but the basic principles (e.g. that producers will not produce at a loss for very long) do seem sound. Also, is perpetual cyclical recession equal to slow collapse, or is there an intermediate state, a stable stagnation? I don't know. (However, I do know that overall economic stagnation plus the rich getting richer is not a stable condition. The people losing out eventually starve and rebel.)
I'm going to skip coal for now. It's already in the mix, already accounted for, not a game-changer. It's one reason any collapse would likely be slow. Does anyone think we're not just going to eventually burn it all (all that can be extracted with sufficient net positive energy, that is) eventually, carbon agreements be damned?
That leaves a build-out of nuclear power and/or solar. Is a sufficiently massive build-out in a sufficiently short time technically possible? Probably. Would it prevent collapse? Definitely. Will it happen? Apparently not. It hasn't happened yet (though a few nations are way ahead of most others), it isn't happening now, and it doesn't appear about to happen. In the U.S., the necessary political resolve might appear around the time rolling blackouts start affecting the neighborhoods congressmen and CEOs live in, but that might be too late.
That's the energy argument, more or less. Note that many otherwise disruptive new technologies, such as the Internet and 3D printing, are irrelevant to the energy argument.
The other type of argument I'll call the organism argument, referring back to the analogy above about the human body. That argument is based mostly on comparing present-day trends with historical observations of the "life cycles" of civilizations or empires, as for example in the writings of Arnold Toynbee or Joseph Tainter. I'm not going to be spending much time discussing that (beyond my personal experiences at the Age of Limits conference, in the other thread), because it comes down to a matter of perception, like trying to judge how similar two pieces of music are. To some people the comparisons seem obvious; others looking at the same thing can't see any connection through the differences. One one side any analogy (e.g. drug lords = warbands on the borders of empire) is considered significant without quantitative analysis; on the other, anything that can be brought up that's "different this time" is considered sufficient refutation, without any requirement to show even qualitatively what difference is actually made or how it would affect the outcome. Not promising for discussion.
