Here I will give a reply in two parts, the second much longer than the first, so those who just want the short of it will know where to stop, while those who want to fully hear me out on this can go on to the rest.
Even theism is slowly evolving into atheism, if you look at the broader trends, as pockets of fundamentalism continue to shrink. Even the larger networks of para-fundamentalism are draining, as members increasingly flee to more liberal religious notions, which essentially trend in the direction of humanistic deism, which is just a stop on the road to secular humanism. It isn't happening as fast as many claim, but it is happening. But regardless of where that trend will end, as I argue in Sense and Goodness without God (pp. 303-07), the overall trend throughout society is toward a safer, healthier, more moral world.
But our discussion focused mainly on the merits of technology in this grand picture. Neopagans seem distrustful of technology. And some accuse me of having too much faith in it. Once at an event in Ventura an atheist came up to me and claimed environmental over-exploitation led to the fall of Roman Empire and will soon lead to the fall of the American Empire (and oddly someone in San Francisco asked me about a similar analogy just recently). Being an expert in Roman history and technology, I knew the first premise wasn't true. The failure of the Roman Empire was political, not economic. Economic causes contributed, but were only able to affect the situation because of political failures. And those economic disasters were not caused by dwindling resources anyway (much less "environ-mental over-exploitation"), but by loss of faith in the fiduciary value of the currency produced by an empire that had been locked in a destructive civil war for fifty years. All of this would have been averted by successfully enacting an effective constitutional government ensuring the peaceful succession of power. But that didn't happen (and was unlikely to), thus dooming the empire, as all empires that fail at this are doomed. History proves that out.
Though "environmental over-exploitation" was certainly a reality in the Roman Empire, and caused a great many problems that better resource management could have averted, it didn't cause the empire's demise, and probably never would have. For the reality is, in any society that solves problems with technological innovation, technology can always compensate for the effects of environmental damage and declining resources. That had already been proven by the Romans themselves. Their solution to damaged and declining fresh water resources was one of the wonders of the ancient world. Less well known are all the many ways they actually continued to increase agricultural output even as they were reducing the fertility of the land. They would have met the growing challenges of dwindling fuel supplies the same way (coal exploitation was only just beginning in Roman England, and we all know now how much more there was yet to dig out of the ground, and what the Romans could have done with it). And so on for everything else.
Which incidentally is exactly what America has also done. We are using the same solution to the same water problem: technological innovation. We are using the same solution to the same agricultural problem. And now we are using the same solution to the same fuel problem. Which, as with the Romans, is only now starting to hit us, long after we faced the earlier hurdles of water and agriculture, though the issue of water is now returning and we are back to tackling it again. Same problem, same solution. And history shows it pretty much always works. We're just a million times better at it now, thanks to the Scientific and Industrial Revolutions and the three centuries of intellectual capital and infrastructure development that ensued. Which is why if resource depletion and "environmental over-exploitation" didn't topple Rome (and it didn't), it certainly won't topple America. It was only when technology and innovation declined, after Rome fell, that the effects of environmental exploitation caught up with the world and did it in. Gradually, technology and innovation returned, accelerated, and eventually brought us the Renaissance and the Industrial Revolution.
After explaining all this to the gentleman in Ventura who made that claim, he responded with a creepy smile, "Oh, so you're a technophile!" evidently confusing values with facts. Just because it is a fact that technology can overcome any environmental loss does not mean I think over-exploiting the environment is okay or that we shouldn't pursue renewable technologies as much as possible or that the best solution is always another technology (since oftentimes less is more). To the contrary, it obviously makes more sense to do things well in the first place, and not have to invent ways to solve the problems that lousy management and bad design create. But nevertheless, we can always invent ways to solve the problems that lousy management and bad design create. I know this, because I have two thousand years of history proving the point, more times than I can count.
Of course, technology only works when you use it. Population growth is a serious environmental problem, for example, but we solved that one decades ago--people just refuse to use all the technologies we invented for the purpose, many because of superstitious voodoo beliefs promulgated by an uppity potentate who thinks he was chosen by God, others because of primitive stupidities opposing the liberation of women, still others because First World countries who ought to know better, can't figure out that funding the sound use of technologies in Third World countries will actually benefit the First World countries more than amply to justify the cost. And so on. But these are all political failures, ideological failures. That's where our focus should be, because it's those things that are getting in our way, not technology. Similarly, the problems technology creates are largely the result of similar stupidities on the political and ideological stage. Technology must be used wisely to work. Thus increasing our wisdom is what we should focus on, not abandoning technology (which, apparently, is the argument of Earth 2100, though that series outrageously exaggerates what will otherwise happen--indeed, exaggerates beyond all rational plausibility--but that's typical of the media).
I encountered a similar pessimism about technological progress in my discussion with Haukur the Neopagan. And that's where my longer digression begins...
Part II: The Long of It
In what follows, I respond to quotations from comments made in that Tao Te Ching thread. Leading up to this, I had defended the point against Haukur that nature is indifferent to us and the source of considerable threat and harm to our interests (as much as also the source of considerable benefit and good), that we have reduced or defeated a great many of her threats (and amplified her benefits) with social and technological advances, and that most problems even of our own creation will eventually be solved by future advances in technology and society--as long as things continue as they so far have been going, and right now I have no reason to believe they won't (though I remain vigilant). Only an extreme (and extremely improbable) global disaster could throw us back into another Dark Age. And even that might only delay or slow things down. After all, we recovered from the last Dark Age, and continued advancing on, well beyond the Romans. It just took a bit of time.
But...
Haukur said... When I try to predict the future I find it difficult to believe that the present period of peace and prosperity (in the West) will continue indefinitely.
That's beside the point. WWI and WWII, even the Civil War, did nothing to slow social and technological progress. To the contrary, in many ways they accelerated them--apart from advances they spurred in technology and industry, women's suffrage in the U.S. was largely a consequence of the lesson of WWI that women were vital to modern warfare, a fact proven almost immediately in WWII; Japan became a free democratic nation, and in a near reversal of its previous course, is now a master of technology and science surpassing in some cases even the U.S.; the Civil War ended slavery, industrialized the South, and expanded all our civil rights with the 14th Amendment; etc.
War still sucks, and I'd rather go slower if that's the cost of peace, but the inevitability of future wars is not going to change our present course any more than past wars did. What would have to happen is something far more catastrophic. Like a severe asteroid impact or a nuclear attack on numerous major American cities. Indeed, the destruction of a few cities wouldn't even be enough, it would have to be a lot more than that. As we can see from the history of Germany and Japan, mind-bogglingly vast erasure of economic infrastructure and human capital from endless firebombing, even the nuking of two major cities, had practically no effect on the advancement of those nations. It's only been fifty years and look at them today. Compare them now with where they were in 1938 or even 1910.
There are often brief cyclic reversals (McCarthyism, Nazism, the Great Depression, even the present economic downturn) but history so far shows these are quickly overwhelmed by the stronger forces of social progress. That's why it would take a hell of a lot more than the worst things that have ever happened since the fall of Rome, to finally throw civilization off track again.
Haukur said... Our current civilization has enormous potential for destroying itself.
So does nature. Just one Shoemaker-Levy and we're done for. It's just not likely. Neither is our own self-destruction. Killing the human race is a lot harder than you think.
Haukur said... Weirdly, this is something most people understood 20 years ago but don't anymore.
Actually, there was a lot of irrationality in that. Almost every film and novel depicting total nuclear warfare greatly exaggerated the actual effects of such a thing, and exaggerated even more the ease with which it could occur in the first place. What scared people 20 years ago was mainly rhetoric--the bluster of two nations in a Cold War, whose only viable weapons were threats they were never going to carry out. Now such a war is even less likely than it was then. And it was pretty unlikely then.
Yet if such a war occurs, and it is on a truly global scale (and not a localized skirmish, like Iran and Israel nuking each other, which wouldn't much affect the long term course of global civilization), that could derail the current pace of social and technological development. In the worst case scenario, it could give us a few hundred years of Road Warrior. Or, to go back to a real example, another early Middle Ages, since the outcome of such a war would be in most respects identical to the fall of the Roman Empire. But it wouldn't wipe us out, it wouldn't produce armies of mutant zombies, and wouldn't prevent another recovery eventually--another Renaissance, Enlightenment, and Industrial, Scientific, and Democratic Revolutions--when we'd get to try again.
Even biowarfare can't do it. As with Small Pox and the Plague, a percentage of the population is always immune, and virulence always declines as a germ spreads. And chemwarfare certainly can't cut it, since it's beyond anyone's economic or industrial ability to make enough chemicals to wipe everyone out. By the nature of the case, all capacity to make chemicals would be lost long before completing the task (since killing the people kills the labor and infrastructure necessary to continue the manufacturing). Nor will global warming do it. We know for a fact the earth has been far hotter than even the worst global warming projections predict, with far higher concentrations of CO2 than we will ever produce, and yet life still thrived in those periods. And humans already have the technological means to ensure we're among the survivors should this happen again.
Not that any of these things wouldn't suck, or couldn't throw us back into a new Dark Age if they occurred with enough severity. But they won't end the human race. So far only nature has the power to do that. She has bombs so big they can melt entire planets. We couldn't even do that if we tried.
Haukur said... Can we really keep the peace for the decades to come?
We have for centuries so far, despite terrible wars, plagues, and economic collapses. Not decades. Centuries. Why would you not expect centuries more?
Haukur said... And can we really transition to a much lower use of fossil fuels?
We will have done so within two hundred years. Guaranteed. Unless, as I've said, the catastrophic happens and the reset button is pushed and another Dark Age intervenes. But even that will only delay the inevitable. Just as it did before. As a historian I see and think in the long haul: I see the world in centuries and millennia, not decades. Most people have such a narrow view of history they fail to see the big picture.
All the things that frustrate you now are trivial in the long run. Oil companies and their graft and greed and lobbying don't matter for sh*t. A hundred years and they're done for. They can't stop it. They are doomed. Within fifty to a hundred years alt power (like solar, wind, seismic, and thermal) and alt fuels (like algal biofuels, hydrogens, and other artificial fuels), will be so cheap, oil won't be able to compete, and materials science will have advanced so far we will be able to cheaply manufacture plastics without petroleum. We won't run out of oil by then (despite repeatedly failed predictions we will), we just won't have any reason to buy that expensive crap anymore. Apart from those world-shaking catastrophes I skimmed earlier, there is nothing that can stop this from happening, any more than anything could have stopped the electric light, the television, the internet, or the iPod. Cities were lit with gas lamps a hundred years ago. They will be lit with a combination of local solar, thermal and seismic power a hundred years from now (just as emergency phones on freeways already are). The gas lamp companies couldn't stop the electric light. The oil companies won't fare any better. It won't happen in "decades," but it will happen. History proves it, logic confirms it. And science explains why.
Haukur said... How can we do that? Either we need some really accelerated technological innovation in this field (something which tends to be blithely assumed with, in my opinion, far too little evidence) or we are going to have to adapt to a drastically lowered level of energy usage.
First: Why does it have to be accelerated in order to happen? From the first electric generator to the television took over a hundred years. Why are you so insistent such developments must happen in ten years or it won't happen at all? That's not how progress works. Even the abolition of slavery took nearly a hundred years after the Bill of Rights; women's suffrage, over a hundred years. Yes, faster would be better. But realistically, history shows that's just not how it works.
Second: We already have "drastically lowered level of energy usage." Almost every single technology today uses far less energy to do the same work than its counterpart eighty or a hundred years ago. Right now we are already transitioning from incandescent to diode lighting, a tenfold increase in energy efficiency, and advancing (LED efficiency has doubled every three years since 1960 and shows no signs of stopping yet). That took fifty years to begin, and will probably take fifty years to effectively complete. And that will happen. It's already happening, and there is nothing likely that will stop it. Yes, increases in population and per capita production have increased overall energy consumption, but this has driven and motivated the very advances that will solve all energy problems in the long run. It will suck for a while, and occasionally more than usual. But not forever, and not catastrophically.
Third: We have always had the strongest incentive to make these advances. Power costs money, so the less power you need to do the same stuff, the more money you make. This basic principle has driven technological advances in energy efficiency for the last fifty years and will continue to do so for the next fifty, and beyond. What is most frequently not understood today is that oil survives as our principle fuel not because of stupidity or politics or corruption, but because it is simply cheaper than every other alternative--far cheaper, in fact.
Large-scale electricity supply is an obvious exception due to efficiencies of scale, but notably we had that solved decades ago--nuclear power. People are just too stupid and scared to use it, another example of refusing to use the technology we already have. Although now the innovation of solar-thermal power is actually a better alternative to nuclear (and won't be blocked by superstitious fear), and such plants are now starting to appear in the U.S. and will inevitably replace all coal and natural gas plants within fifty to a hundred years.
Otherwise, electric cars, trucks and trains will replace gas once we develop batteries that are light and cheap enough--and technologies like supercapacitors are already in the works, which will solve that problem soundly in less than fifty years. Local power generation (e.g. solar panels on homes, seismic generators on bridges and street lamps, small wind turbines on the roofs of hospitals, etc.) will rise as the cost of manufacturing and maintaining the necessary equipment declines. And alt fuels will ultimately become cheaper than mined-and-processed oil. It's just a matter of time. As history teaches us, you can only quibble about when, not if.
Haukur said... The second possibility will be hard to do peacefully and however it happens it is likely to lead to less affluent lives for our descendants.
That depends on how you define affluence. It is often remarked that Swedes have much lower incomes than Americans, yet Swedes have so much affluence available to them from public services (e.g. health care, mass transit, vacation laws, etc.) that their real net wealth is by some accounts actually greater than the average American's.
Similarly, to replicate fifty years ago the wealth we have available to us today due to increases in productive efficiency and advances in technology would entail the same conclusion: that we are vastly wealthier now than our parallel selves would have been then.
For example, imagine what it would have cost in 1960 to have all the then-equivalent capabilities of an iPod, cellphone, computer, and cable and internet service. And that's just some of the things you have at low cost. That's how wealthy you are now. You're fat rich. We just obsess so much over dollarsigns and the remaining failures of our social system (almost entirely represented in the cost of housing and health care), that we fail to see how affluent we actually are.
Now project fifty years into the future. What will your parallel self have then? What would it comparatively cost today to have the equivalent capabilities?
Pay attention to history. It teaches us the future. I do think eventually we will start getting by with less, but by then, "less" will actually be considerably more than we have now.
Haukur said... ...the more actively [atheist bloggers] want to embrace atheism and rationalism, the more they tend to embrace super-optimistic ideas about the future.
I'm a realist. Realists base their predictions of the future on the examples of the past. If the result looks optimistic to you, you are the one being unrealistic, not the optimists.
But we must take care not to confuse hope with prediction. I can talk about what I want the future to be, and what I think it could be if certain things happened that are unlikely to happen (like the widespread embrace of secular humanism and naturalism) unless I promote this causal connection and make it happen (which is presumably what most "atheist bloggers" are doing). But that's not the same thing as making predictions about what's likely to actually happen.
That things will go better if we embrace secular humanism is a reason for promoting secular humanism, just as much as it was a reason to promote democracy and human rights (and those have turned out pretty well where they were achieved). Such promotion thus requires describing the possible future that will result. That shouldn't be mistaken for optimism. It's instead an argument for change. Hence I distinguish that project from making realistic predictions about what is most likely going to happen given the way things actually are so far.
If in doing the latter anyone sounds excessively optimistic, it may simply be because they are downplaying the bad things that will still plague us or pop up in new forms, and choosing instead to emphasize the markers of progress that will be realized. Or it may be because their time-scales are unrealistic. Typically great advances take centuries, not decades. But only rarely do I see people predicting things that will never happen. Take space colonization for example: I doubt we will see any steps toward that even in the next hundred years (much less the next twenty), but five hundred years from now it will be a practical certainty.
Haukur said... Getting yourself cryogenically preserved becomes "massively overdetermined", even at 100K$ a pop.
I don't understand what the phrase "massively overdetermined" means in this context. Cryogenics advocates are advocating present technologies, not future ones, except insofar as they assume future technologies will be able to remake them from the material that present technologies are supposed to preserve, and on that point they are absolutely right. The technologies they expect will exist some day. That's inevitable.
The only questions are (a) do any present technologies really have the ability to preserve the necessary information (personally, I doubt it, but that's not a prediction of the future, it's a judgment of the existing technologies) and (b) can they preserve it long enough (since body and brain reconstruction from any stored material will not be available for at least a hundred years, possibly two hundred). As to what it costs, that's again just a present reality, and not relevant to future predictions. In two hundred years I predict resurrection will not only be readily available, it will be as common and affordable as a cable TV subscription is today. In fact, it will probably be a standard component of life and health insurance policies.
Haukur said... [Or claims that] the "Singularity" is just around the corner.
I agree the Singularity stuff is often muddled nonsense. I just don't know many advocates of it. Those who do advocate it are often unrealistic about the physical limits of technology, and particularly the nature of IQ. They base their "predictions" on two implausible assumptions: that advancement of IQ is potentially unlimited (I am fairly certain it will be bounded by complexity theory: at a certain point it just won't be possible to think any faster or sounder or more creatively) and that high IQ is predictive of accelerating technological advancement. History proves otherwise: even people ten times smarter than people like me produce no more extensive or revolutionary technological or scientific output, much less invent more technologies or make more discoveries--in fact, by some accounts they often produce less in those regards than people of more modest (though still high) intelligence.
However, Singularity fans are right about two things: machines will outthink humans (and be designing better versions of themselves than we ever could) within fifty to a hundred years (if advocates predict this will happen sooner, then they are being unrealistic), and the pace of technological advancement will accelerate. However, this is already accounted for by existing models of technological advancement, e.g. Moore's Law holds that computers double in processing power every three years, Haik's Law holds that LED's double in efficiency every three years, and so on (similar laws probably hold for other technologies, these are just two that have been proven so far). Thus, that technological progress accelerates is already predicted. The Singularity simply describes one way this pace will be maintained: by the recruitment of AI.
It therefore doesn't predict anything remarkable, and certainly doesn't deserve such a pretentious name. Because there will be a limit, an end point, and it won't resemble a Singularity: there is a physical limit on how fast thoughts can be thunk and how fast manufacturing can occur, quantum mechanical limits that can never be overcome, by any technology. Once we reach that point, the pace of technological advancement will cease to be geometric and will become linear, or in some cases stop altogether. For instance, once we reach the quantum mechanical limit of computational speed and component size, no further advances will be possible in terms of Moore's Law (even Kurzweil's theory that it will continue in the form of expansion in size ignores the fact that we can already do this now, yet we don't see moon-sized computers anywhere--a fact that reveals an importantly overlooked reality: what things cost).
Ironically, the same has been discovered about actual singularities: they, too, don't really exist, and for the same quantum mechanical reasons (see my discussion here).
Haukur said... I found an atheist blogger (and a good one too) who's still holding out for neuroscience to tell him whether beliefs and desires are real or not.
He is only being unrealistic if his time-scale is too small. Otherwise, he is quite right to expect this. We are already building unbeatable lie detectors with existing brain scan equipment. With a hundred more years of progress in brain scanning resolution we will inevitably be able to map every single brain process down to the individual synapse and firing event. Once we can do that, we will be able to locate exactly where beliefs and desires exist in the brain, exactly what they consist of (both physically and computationally), and be able to fully explain why they exist and how they work. You can bank on it. You just won't likely live long enough to collect. :-)
50 comments:
Atheism is evolving to theism? After listening to your debate with William Lane Craig, I doubt that's true.
Sam MacKinnon: Craig represents a dying breed. His brand of conservative Christianity is like the proverbial frog slowly boiling to death. Just look at the statistical trends, on all relevant markers, over the course of the last eighty years or so, then project another eighty years out, and you'll see what I mean.
This was a post of rational hope. I have no rebuttal or counterpoint or addendum ... but I feel optimistic for the future of my human family!
"I'm a realist. Realists base their predictions of the future on the examples of the past. If the result looks optimistic to you, you are the one being unrealistic, not the optimists."
Love that - I get so tired of neocons, Christian apologists and other ethical mutants acting as though their dickish, cynical, demonstrably wrong approach to life and politics was just 'common sense."
I've never been one to believe this inevitable, "manifest destiny", of human technology.
"technology can overcome any environmental loss"
I think this is, at best, too strongly worded. It is a mistake to simply conflate technology and resources. Technology is just knowledge, techniques, or tools that you ultimately have to apply to resources. It's trivial to create a thought experiment where you give a population of people some massive amount of technology and some horribly degraded environmental condition (such as the desertification of the entire planet or the heat death of the universe) so that they all die. Through this resource-technology space there are various areas of survival or destruction including our current state. Technology is highly dependent on previous advances so you are essentially worming your way through this space and I just don't see how we are guaranteed to reach any arbitrary height of technology ("space colonization"). It's all well and good to just plot our past trajectory, but technology, by its very nature, is impossible to predict and resource depletion is similarly difficult. Simply extrapolating forward may be realistic, but it's by no means a guarantee of anything.
So, I don't see how it is inevitable that we are going to reach any pinnacle of technology. We may get to a technological dead-end where we used up all the technologically accessible resources necessary to reach some further step.
"But it wouldn't wipe us out, it wouldn't produce armies of mutant zombies, and wouldn't prevent another recovery eventually--another Renaissance, Enlightenment, and Industrial, Scientific, and Democratic Revolutions--when we'd get to try again"
Except we've already used up all the easily accessible resources when WE did all those things. These guys will be working with sticks and stones (and whatever resources we may have left over from when we stepped up to something new). They may very well be trapped unless you want to consider the geological time required to regenerate whatever resources we may have used but then all bets are off (if some reasonably intelligent species continues to exist or evolve they may get another chance--but they won't be us and still no technology is ever inevitable).
Ultimately, my case is fairly easy to make because you made a very strong claim: technology will inevitably continue to improve--no exceptions. You can soften it to some lesser degree and it becomes a probability calculation of how likely some given technology is and none of us really have the required information to answer that with any accuracy for any non-trivial technology (beyond predicting hard drive capacity, etc.)
Ultimately, I'm a pragmatist. I put most of my eggs in the basket assuming things will all go to plan (at least during my lifetime), and generally ignore doomsday scenarios. If the world climate undergoes some unstoppable positive feedback loop and re-stabilizes in some non-life-supporting state, I'm just dead anyway. I do reserve a few eggs, though, for the possibility of social disruption, food shortages, etc. that may accompany something like a difficult transition to non-fossil fuels. It's entirely possible, or at least, in my opinion, possible enough to take into considerations.
I do think there is a kind of irritating skeptical blind-spot that many (typically technologically inclined) atheists have towards inexorable scientific progress (and yearning to transfer their consciousness into some type of cyborg). We're all human with our various biases working under relatively sever conditions of incomplete information.
Now that I've attempted to poo-poo your hope, inspiration, optimism you may continue with your regularly scheduled day. :)
Shane, how dare you rain on our Futurism parade?!
Wow, thank you. That's quite an extensive reply. I'm unable to offer anything substantial in return now or in the near future since I'm expecting to become a father sometime in the next 16 hours or so. Let me just say that while I hope that the simple homebirth my wife has planned will succeed, both she and I are grateful that there is a 21st century hospital five minutes away, in case it does not.
Great post!
Richard said...
>>I'm a realist
>>it is a fact that technology can overcome any environmental loss
That's not realism, that's exactly what Haukur described. Irrational, hyper-optimistic faith in reason.
Our technological approach to life - domination of life - has resulted, amongst many benefits, in an environment that is increasingly toxic to our being and fundamentally endangering our survival.
Remember Socrates, the original man of science and hyper-optimism, and his final pessimistic word on the scientific approach to life: "I owe a rooster to Alcepius." (Life is a disease.)
All of this irrational blather about science and morals, science and perfection, is on the same self-deluded course.
Any opinion of John Michael Greer's theory of catabolic collapse? He proposes that the Roman Empire was top-heavy with its army and its administration, and that it was short of the revenue source that the Republic had: conquest and plunder. The Germans and the like were too poor, and the Parthians were good at fighting them off.
I think that Shane has a legitimate concern about keeping our technology going; we will need it for building a renewable-energy infrastructure. And the best way to do that is not to dawdle on renewable-energy development but to try to get it as ready-to-go as possible.
Wind energy is already starting to become economically competitive, though solar still does not seem quit there.
Rick,
Thanks for the food for thought.
Ben
Richard,
On the whole I agree though where you seem very confident that technology will provide remedies, I think we must remember that the transition periods between energy sources or technologies can be rough. Had the transition from whale oil to petroleum happened a decade later, we may have exterminated the whales and suffered rough times with very little oil. Now too, while oil is not going to run out, the cost could make it unavailable to many people. Fortunately I agree that we already have the technologies to ease that transition, provided we act quickly.
Bug generally, just because it has always happened in the past doesn't mean it will always happen in the future. It's certainly possible that the environmental damage could become severe enough that the world human population gets cut back to a small fraction. Bounce back? Eventually, probably, but what a rough time it'll be. It's still just hope that we can change in time, not a certainty.
-nuclear power. People are just too stupid and scared to use it, another example of refusing to use the technology we already have.
Nuclear power has had a lot of hype but despite advances still continues to be one of the most expensive forms of power available. Nuclear power plants have always been heavily subsidized to bring the costs down to that of other forms and the government is rightly reluctant to continue this. Smart and rational people avoid nuclear power because it's expensive to build and produces expensive power, time-consuming to build, relies on hard to obtain materials, consumes large amounts of water (which is becoming increasingly rare and costly) and still has unresolved issues of waste disposal.
http://climateprogress.org/2008/06/13/nuclear-power-part-2-the-price-is-not-right/
http://www.americanprogressaction.org/issues/2008/nuclear_power_report.html
Excellent post, but I must make two physicist comments:
1) While it is true that quantum gravity should nullify singularities, we don't have a proven or even demonstrably plausible theory of quantum gravity yet, so that holding that there are singularities within black holes is still reasonable. There are certain directions that would justify this, for example ones considering GR to be a classical reality that is behind the emergence of the classical from the quantum.
2) Is achieving enough resolution and amplification to see every synaptic firing using current imaging principles reasonable? Without crunching the numbers, my gut feeling is that the systems are too noisy and the signals too weak for this to be feasible.
Richard: Can you poke any holes in the following semi-doomsday scenario? Some people are religiously opposed to birth control. Since it seems that about half of the differences in personality are genetic, it's unlikely that there's any genes out there that would guarantee that you'll want to have 8 kids, but there probably are genes out there that substantially increase the chance that you will have quite a few kids for religious reasons. In an environment like modern America, this means a substantial increase in evolutionary fitness--we're talking 100 percent or more fitness increase. If that's right, these genes should increase exponentially in frequency, dominating the population in a few hundred years. When that happens, the population will be kept in check by religious wars rather than birth control. It won't quite be doomsday, but it life on earth won't be very much fun.
I'd like there to be something wrong with this scenario, but what?
UH - I think this is only a doomsday scenario if religiosity and birthrate are highly heritable and confer a strong advantage. I don't think either are the case, in fact I don't think they're heritable at all.
Tyro--Why do you doubt they're heritable? That they are is for me just an educated guess, but one based on the fact that so many things are, in part, heritable.
As for conveying an advantage, by definition higher birthrate is an advantage, as long as it doesn't, say, inhibit the ability of your kids to have grandkids because you spread your resources too thin. And in a rich country, spreading resources too thin is not a issue when we're talking in terms of 1 kid vs. 8 kids. You may be being misled here by the "survive-and-repoduce yes/no" measure of fitness realize *quantity* of reproduction matters.
Side note: the situation is slightly less bad if there is no deep link between religion and birthrate. Still means lots of early death down the road, but it avoids the global theocratic hellhole problem.
None of these factors seem enough to diffuse the problem. My main hope is that somehow the technological and cultural winds will shift, making currently-adaptive genes non-adaptive, but I don't see how that would happen.
UH - because they vary strongly with culture and environmental conditions which themselves change quickly. If there's a genetic component, these features should be long-lasting and be preserved through the generations but this is demonstrably not the case. Take a look at the number of children of children immigrating from Asia or Africa. Their parents and grandparents often have many siblings yet their birthrate is predicted by their countries, not by their family history. If there is an effect, it appears so weak as to be drowned out by other factors.
I think you're falling for the adaptationist fallacy, that constructing a story whereby a behaviour may result in an advantage means this behaviour is a heritable trait. It needs to have a genetic component and the behaviour has to be strongly favoured. Neither appear to be the case.
That they are is for me just an educated guess, but one based on the fact that so many things are, in part, heritable.
Most behaviours are not heritable. Imagine a Gilmore Girls gene or a Drives-Pickup-Trucks gene, even though these behaviours may be expressed by parents and their children. The evidence is not sufficient to rule out the null hypothesis. Until that happens, I wouldn't worry.
No plausible quantum gravity theory that eliminates the singularity? Not according to New Scientist cover story 13 Dec 2008 "Recycled Universe: from big bang to big bounce". The story seems to suggest that "Loop Quantum Gravity" is at least plausible, and suggests a universe collapsing then bouncing to form the universe we know and love. Is the story misleading?
?
Important clarification: What I'm worried about here is not birthrate in general--which obviously seems to be determined largely by environment. The trait of interest here is "birthrate in environments which cause low birth rates in most currently-existing humans." There are middle class Yankees of the current era chosing to have huge families, what I'm suggesting is that what makes them different from most Americans is likely a combination of culture and genes.
Also, your Gilmore Girls and Pickup Truck examples involve a fallcy of your own: you assume that because there is no gene whose sole function is to regulate those things, then there must be no genetic component there. But obviously it could be true that there are behavioral genes which have, among their seval effects, a tendency to like things with certain features found in pickup trucks or the Gilmore Girls. Consider porn: probably no genes specific to it, but no one would watch it if we humans didn't have the genetic programming we do.
UH - Actually no, I didn't ignore that point. That was my point.
You're pointing to a very specific behaviour in a culturally-connected group of people and assuming there's a strong genetic component which would be heritable and would outweigh cultural changes. I don't know what the stats are right now, but a few years ago when I checked, most western nations were averaging less than replacement rate and it was only because of immigration that their numbers were growing. This is true even in countries which used to have huge birth rates just a short time ago. What's notable is that the pop notions of evolutionary psychology would predict that couples would be driven to have at least two kids but this isn't the case.
So if there is any heritable component, I'm saying that observations indicate it's so weak that it's easily drowned out by cultural changes. If it's this weak, it will almost certainly not spread (there are equations for this, I can dig them up if you like). Populations that are having lots of children are doing so for ideological or religious reasons and looking at their family history, I doubt that you'd find any significant differences. What distinguishes them isn't their genes but their beliefs and these aren't heritable. Can be indoctrinated but not genetically inherited.
I could be wrong and there may be some sub-populations somewhere that are different but I think the data is good enough to justify a negative stance until something changes. Are you aware of any studies which found a strong genetic component to birth rates?
BTW: there are several famous families (e.g.: royalty) who have had dozens or even hundreds of children (from the man, obviously). It shouldn't be hard to see if this was inherited.
Once again, a thought-provoking post.
This kind of reasoning is in line with how I think, but I have looser, more vague precedents than you have presented.
I remember Melvyn Bragg interviewing P.J. O'Rourke back in the '90s. O'Rourke mentioned the interesting fact that there are too many people in the wrong places, but there aren't too many people (yet).
The penny dropped. The scare-mongering by David Suzuki (whom I respect despite this) &c. was more guilt-ridden rhetoric than reality.
The short arguement: why should the Western world, a minority, feel guilty about having children if the bigger, less developed countries, whose citizens don't give a shit about David Suzuki, are going to have theirs? Let them have theirs, and let us not feel guilty about having ours (if we want them).
Neuroscience is a field where I just don't dabble. And I have an odd amount of respect for deconstructionism than you'd think for a spiritually inclined person (go to the TED talks site and look for Tim Ferriss' talk). Yet, echoing the concern of יאיר רזק, I wonder how useful the approach you describe would be.
When we reach the smallest possible size for a transistor, maybe we'll go to quantum computing. That would be a typical leap such as from the boat to the aircraft; from the propeller to the jet.
Oh, and you should expand on your comments about IQ and intelligence in a future post. Very interesting stuff indeed.
Pikeman Urge,
Why is a large population in the West a problem? Because we consume vastly more resources than the developing world does of course, not only on a per-capita basis but also on absolute numbers. I should think that was pretty clear if you'd read up on the issue a bit.
So what's with calling Suzuki a "scare mongerer"? Sounds like that penny dropped before you'd figured out the facts.
Tyro, it seems as if you don't think we can solve our resource problem. I think Richard at least implied that we could.
You don't count on the advantages of a solid population. If our population is larger, immigration becomes slower, allowing a more natural 'integration', for lack of a better word. Immigration if done at too fast a pace can have ill side-effects.
For the record, I'm 'pro' immigration. I am, after all, myself an immigrant.
Fall of Rome Stuff
Loren said... Any opinion of John Michael Greer's theory of catabolic collapse?
In general, he correctly describes the resource management problems societies will face and the directions they can take, but he seems to ignore the real factor, which is human organization. As long as you maintain a successful government, adaptation to any change in resource dynamic can regain equilibrium (for example, he expressly includes technological adaptation as an available solution, my point exactly). Conversely, tank the government, and all bets are off: it's hell-in-a-handbasket time, no matter what the resource management situation. So he's only got half of the picture, and as is often the case, he mismatches the historical facts to suit his theory, to wit...
He proposes that the Roman Empire was top-heavy with its army and its administration, and that it was short of the revenue source that the Republic had: conquest and plunder.
The top-heavy military and admin actually occurred after it's collapse in the 3rd century, and were desperate attempts to react to and quell that collapse, by instituting tighter control. One could perhaps point to the avariciousness of the governing class before that, which is not so much top-heaviness as inequity, but that was another symptom of the political defects I have often noted were the real cause of Rome's fall. Such inequity showed no immediate signs of injuring the system. It was the failure to establish a stable constitutional government ensuring the peaceful succession of power to capable leaders, that was the single linchpin that did the Empire in.
Also, he is relying on obsolete models of the ancient economy, greatly exaggerating the conquest revenue stream. In fact, the early Roman Empire hugely ramped up agriculture, manufacturing and trade exactly as it was reducing its conquests, and all archaeological markers show the result was greater success, not less. Had they continued in that direction, there would not have been any collapse at all, but very likely an eventual industrial revolution. All they needed was to keep the system working, which only required a working government that maintains law and order without resorting to constant civil war, and that pursues sound economic and social policies that increase rather than reduce public industry and confidence (instead of attempting to unfairly dink the system and instead cause a massive depression).
Hence what happened in the 3rd century had nothing to do with dropping revenues. It was entirely the product of inherent political instability and bad government.
Wind Power
Wind energy is already starting to become economically competitive, though solar still does not seem quit there.
Solar thermal is, it's just water intensive (so is nuclear), so it faces a competing resource problem (though that can be exaggerated, I'll discuss it in a later comment below).
I'm not sure about the expandability of wind as a solution, unless we start developing widespread local power generation (e.g. small backyard and rooftop turbines everywhere). We seem a ways from that now, since no one is developing it (i.e. aiming to scale down, simplify maintenance and production, solve safety issues, and thus ultimately make the equipment cost-effective for homeowners and landlords to invest in). This is happening for solar panels, though, and the trajectory is good. I expect affordable solar panels as standard roofing within ten to fifteen years.
Were I in a position to make a difference, though, I would pursue the development of cheap turbine DIY kits, foregoing any obsession with unnecessary efficiency, e.g. simple carpentry and wiring with little more than spare junk can produce a turbine capable of supplying a substantial portion of household power requirements. It's inefficiency would be irrelevant, as its cost would be so low the gain in energy savings would more than recover the costs (and since continued generation is nearly free, any inefficiency costs nothing). In fact, if we streamlined the idea with ultra-cheap, factory-made kits, efficiency would improve as well.
No one seems to be thinking like this. I expect that will change in ten to twenty years though, especially with computer home fabrication technologies on track for widespread availability. Just as PODs and printers and home binding kits, etc., have made publishing cheaply available to everyone, machinery that can make any plastic object, even sew, on command from an ordinary home computer is being scaled down and dropping in cost even now--within twenty to thirty years, a home fabricator machine will be as cheap and common as a home printer, and we will be able to literally manufacture wind turbines at home (and who knows what else), with everyday software, using parts bought at Radio Shack (or cannibalized from junk yards or even bought at Target).
That's just an example of the kind of thing we need to keep our eye on the horizon for. It's already happening, people just don't pay attention to the trends in technology, so they are unaware of what's in the pipe.
Physics Stuff
Guy with Hebrew Moniker I Can't Type: While it is true that quantum gravity should nullify singularities, we don't have a proven or even demonstrably plausible theory of quantum gravity yet, so that holding that there are singularities within black holes is still reasonable.
Only if you assume quantum mechanics is in some way false on current understanding, and false in exactly that respect rather than some other, which I'll grant is possible, but unlikely.
Hence I'll say on current understanding, QM makes singularities impossible (for the reasons given in that link). The only way to avoid these consequences is to speculate some new unproven theory of physics. Which scientists generally call "making shit up."
Is achieving enough resolution and amplification to see every synaptic firing using current imaging principles reasonable? Without crunching the numbers, my gut feeling is that the systems are too noisy and the signals too weak for this to be feasible.
I doubt that will be a barrier.
First, we already overcome similar conditions in existing instruments--e.g., adaptive optics in telescopes has now made ground telescopes superior to space-based telescopes, by simply eliminating the noise caused by the atmosphere; and many other imaging instruments take averages of many scans, thus locating a consistent signal and deleting the noise. The recent reconstruction of the Archimedes' Codex is a sterling example of the technique of combining numerous data streams to erase noise and detect a signal in spite of it. Computers can also calculate connections without having to see them. For example, an MRI only tracks blood, not synapses, but with enough scans you can actually begin to predict where neurons exist and where connections between neurons exist, simply by running computations on the observed activity detected. You don't have to actually see the things themselves.
Second, we are certainly nowhere near any such limit yet, thus our ability to "read minds" will only improve immensely long before we reach any limit on resolution, and as far as immortality technologies, even if we could show that a live scan won't work for all the noise, we would simply resort to a destructive scan (i.e. physically dismantle the brain cell by cell, nanorobotically or by any other method, mapping as you go), and rebuilding the brain with the resulting data, which you then have in store for future reproduction, or transfer to virtual environments.
Third, you might be thinking solely of the MRI. That is not the only scanning technology available, nor need it operate in isolation. For example, combining multiple MRI, PET and CAT scans can produce increasingly accurate pictures of internal structure (similar to what was done for the Archimedes' Codex in tracking iron atoms in its structure). We are also now developing other kinds of scanning technologies, which can combine with these. In fact, the history of imaging technology itself sets the pattern, and as in the past, so likely in the future: first we had just the X-Ray, then the CAT, then the PET, then the MRI, then the fMRI.
If I were to wager on where brain scanning technology is heading over the next fifty years or so, it would be an evolution of the MRI to use focused beams (instead of general fields) to detect spin corrections in the hydrogens of lipids and proteins using adaptive computerized noise reduction. The result: an accurate map of all cell surfaces in the brain.
The "Oh No The Breeders Are Coming!" Scenario
The Uncredible Hallq said... Can you poke any holes in the following semi-doomsday scenario? Some people are religiously opposed to birth control [and they have differential reproductive success] ... When that happens, the population will be kept in check by religious wars rather than birth control. It won't quite be doomsday, but it life on earth won't be very much fun.
Well in general, that would be an example of my category of "yes, it will suck, but it won't be doomsday," so it isn't a doomsday scenario to begin with (in fact, didn't your scenario just describe the last 2,000 years? :-).
But in particular, I think it's an example of Plantingian pseudo-evolution (on which see my remarks in We Should Attack Rocks? with Reliabilism and following sections, and Reppert-Plantinga's AfRF). As has been pointed out by others here already. Just as it is ridiculous to think one can evolve a fear of sticks because it keeps you away from snakes, so it's just not credible to imagine some sort of "anti-birth-control" gene. No such thing does or even could exist. Opposition to birth control can only be cultural, and therefore is subject to memetic, not genetic selection. And any gene that reliably promoted adoption of that meme would have a thousand other effects, and it's unlikely the net of those effects would be positive.
In all honesty, we cannot predict what will happen one way or the other, unless you can identify the actual gene you have in mind. But just looking at the basic facts, culture is like an environment: your success is mediated by your ability to adapt to that environment. And it's not enough to just have a lot of kids. You also have to raise them. Hence just as it would actually be maladaptive to have large families in a desert, where resources are scarce (hence smaller families have more net success), it appears maladaptive to have large families in a modern first world society, as indicated by the fact that the clear adaptive trend within such environments is toward smaller families. This is usually explained by the fact that in such societies children are a net drain on resources rather than (as once was the case) a net gain (through child labor among other things), hence it is so resource-exhaustive to have a large family, smaller families have an overwhelming advantage.
I suspect this would show up if you studied the number of people you have any connection with who hail from large families vs. small, compared to that ratio as it was for either of your parents or grandparents (especially if you only include in these counts persons born to native citizens). Most likely, the ratio will have shrunk with every generation, indicating that, contrary to your prediction, large families are actually under-competing small ones, not the other way around. Maybe the results would come out the other way, but from my experience that seems unlikely. In any case, someone would have to run the study to see. Maybe someone already has. Let me know if anyone here finds any.
By contrast, IMO, rationality and truth-valuing seem highly adaptive so far, and the past historical trend has been toward both. I don't see a contrary trend.
But speculation is moot, as in any case, it will take thousands of years for such a development as you propose to dominate the gene pool, yet barring catastrophe, in five hundred years we won't have gene pools anymore. Even if we still use genes to build our bodies then (which is not necessarily likely), we will have complete control over our genetic architecture and thus will no longer be subject to natural selection. So in all probability your scenario will never transpire.
Irrational People Are Lame: Case In Point
Solon said... That's not realism, that's ... Irrational, hyper-optimistic faith in reason.
It's that very statement that is irrational. It rejects all empirical evidence of both the present and the past ten thousand years, and rests on no valid logic.
Our technological approach to life - domination of life - has resulted, amongst many benefits, in an environment that is increasingly toxic to our being and fundamentally endangering our survival.
That exaggeration is not in accord with the facts. Mismanagement is indeed endangering our pocketbooks and making many lives in varying degrees miserable, but it has not and will not endanger the survival of the species. And we already are aware of the problem and taking steps to remediate it, and have the technologies to do so, and those nations and communities with the political will, are implementing them. It's not technology that is hurting us, it's human stupidity. Exactly as I said: if we actually used the technologies we already have, and soundly, none of the degradation you refer to would exist.
This is illustrated when comparing any third world port like Rio Haina in the DR with a well-managed U.S. port like the ports of San Francisco or Seattle: the difference in pollution and damage and harm is vast, yet not because of any difference in available technologies, but solely because of a difference in socio-political wisdom and will. And yet the fact that these differences exists is proof that humans are not so inevitably stupid they can't use technology and manage environments wisely. San Francisco does a reasonably good job. It could do better, but compared to Rio Haina (or, just imagine, Mogadishu), it's lightyears improved.
Further confirming my point, China is now instituting massive programs to clean up and de-pollute its ports, which if carried through, will outdo any U.S. policy by far. All that's needed is the will. Technology isn't the problem.
Remember Socrates, the original man of science and hyper-optimism...
Whoa, who are you reading? Socrates was one of the most cynical and anti-scientific philosophers of antiquity. He was the godfather of Cynicism, which was begun by his pupil Antisthenes and was the only sect that remained most true to Socrates' original teachings.
...and his final pessimistic word on the scientific approach to life: "I owe a rooster to Alcepius." (Life is a disease.)
Huh? Are you saying you agree with him, that life is a disease? So, why are you still here? Drano is cheap.
In any case, that remark for Socrates had no connection at all to "the scientific approach to life," which did not yet exist when Socrates was alive. Aristotle was the first to propose anything approximating any such thing, and to articulate its methods and principles and goals (and arguably the modern worldview is an evolution and perfection of his original program).
So you are just making shit up.
All of this irrational blather about science and morals, science and perfection, is on the same self-deluded course.
Said the most deluded man here.
I don't heed loons.
Nuclear Power Stuff: Part I
Tyro said... Nuclear power has had a lot of hype but despite advances still continues to be one of the most expensive forms of power available. Nuclear power plants have always been heavily subsidized to bring the costs down to that of other forms and the government is rightly reluctant to continue this.
That's not really true. I think you are being duped by anti-nuke propaganda. The sites you link to make many dubious claims, and contradict more reliable sources. International, U.S., and EU assessments (see summary) do not give many of the figures your sources do. And indeed, your claims contradict the fact that China is also ramping up nuclear facilities, which would make no sense if it wasn't cheaper, since they control all capital power investment and thus obviously would build what will cost them less, not what would cost them more.
More tellingly, they misrepresent even the studies they cite, such as the Keystone Center's Nuclear Power Joint Fact-Finding Dialogue, neglecting to mention, for example, that their estimate of kwh cost was based on assumptions of escalating construction and materials costs of plant creation, which would affect all power generation industries, not just nuclear, and also neglecting to mention that that study calculated that if the government built (or merely guaranteed the loans to build) the plants, the cost drops 25% below the figure your source reports.
Governments have subsidized nuclear power because governments wanted it, not because power companies needed the money. Nuclear power brings vast returns on investment (especially now; Entergy Co. in particular is making a killing). Companies are chomping at the bit to build more of them because they make so much money--only laws stand in their way, not capital. The fact that governments often footed the bill (usually for construction, not maintenance) is just another example of corporations stealing tax dollar pork (or, if you want to be charitable, another example of positive socialism at work). Although coal (up until recently) was competitive with nuclear, since nuclear costs about the same, and coal is a non-renewable, highly-pollutive resource, coal really has nothing to recommend it (even so-called "clean coal" which actually costs more).
Nuclear Power Stuff: Part II
Tyro said... Smart and rational people avoid nuclear power because it's expensive to build and produces expensive power [and] time-consuming to build
It is expensive to build, that's true, but it does not produce expensive power. In fact, in most states, it's cheaper than any fuel-based alternative. As for time, that's a factor of politics, not technology. Nuclear plants can be built as quickly as any other. The only disproportionate delay comes from all the slow inspection bureaucracy and public resistance that generates endless red tape. In other words, human irrationality is what makes nuclear plants slow to build.
...relies on hard to obtain materials...
That's not really true. Uranium is no harder to obtain than any of thousands of materials we depend upon daily. In fact, in terms of actual process, it's easier to obtain than oil, and about comparable to coal.
...consumes large amounts of water (which is becoming increasingly rare and costly)...
That's equally true of all power sources: solar thermal uses as much water as nuclear, and coal and oil and gas plants almost as much. Only wind and solar (and technically hydroelectric, since the water used would be lost anyway) use minimal water, but they are not cost-competitive, and even when they are, cannot achieve the required scale. I agree we need to expand use of them far beyond current levels (and thus keep improving their cost efficiency as well), but there will come a point when we've maxed out their expandability (regardless of their cost), and that point will be far short of national power demand. I think considerable room to expand even more will be available in small-scale local power generation (as I've said elsewhere in comments here), but unless someone really supports that avenue, it won't materialize fast enough to meet demand. We need the big stuff.
However, the "cost" of water is not as major an issue here as elsewhere. Even with rising costs due to a shrinking supply-demand ratio, the cost of water is a minimal component of power generation costs, especially since a substantial amount of the water that powerplants use is recovered. Agriculture uses vastly more water, without recovery, hence food production has far more to worry about on this score. Moreover, water is actually not in short supply--fresh water is in short supply. Agriculture needs fresh water. Powerplants do not. In fact, powerplants essentially do the same thing to water that desalination plants do, so it is really a wonder no one has yet developed a combined plant, which consumes seawater for powerplant steam, the condensed run-off of which would be fresh water. The only tech hurdle is what to do with the brine stream, but there are many avenues for innovation here (I can think of several just off the top of my head).
...and still has unresolved issues of waste disposal.
Only, again, because of human irrationality. Politics alone has prevented simple and obvious solutions. In any real rational analysis, the waste issue is actually vastly less a concern for nuclear than oil or coal, which are polluting the entire atmosphere of the earth, causing global warming, and killing millions of people from carcinogens and contaminated air and water. Exxon Valdiz alone did more environmental harm than the entire nuclear waste industry in its entire history has ever done or is ever likely to do. But people don't do rational math. They hear radiation and they freak out. They hear Valdiz and global warming and don't give a shit. That's not technology's fault.
Whale Oil!?
Tyro said... Had the transition from whale oil to petroleum happened a decade later, we may have exterminated the whales and suffered rough times with very little oil.
I hope you're kidding. Before petroleum, the primary source of oil was olive and corn, not whales, and the primary source of energy was coal, not oil. Your imagined scenario never existed.
Your time scale is also weird. China had been using petroleum as a fuel source since the 4th century, the Middle East since the 9th century, and Europe since the 13th century. Petroleum refining began circa 1850, and the Americas had reached hundreds of millions of barrels of production annually by 1900. So what narrow "decade" of transition are you referring to?
Now too, while oil is not going to run out, the cost could make it unavailable to many people. Fortunately I agree that we already have the technologies to ease that transition, provided we act quickly.
Neither is true.
First, if you read up on the actual way oil is being hoarded and sold, you would know the price of oil is inflated possibly an order of magnitude above it's actual value, because OPEC limits production to limit supply and thus maintain profits. If OPEC didn't collude to do that, oil might be a tenth it's current cost. And although that would much more rapidly exhaust supply, the fact that OPEC is sitting on vast supplies specifically to inflate prices tells you one thing: we are nowhere near close to facing any supply crunch on oil. OPEC has enough oil to keep its cost wherever they want it to be, for centuries to come. It's just we will have replaced oil by then, with technologies already in the pipe (e.g. clean algal fuel).
Second, even if we removed OPEC somehow and started ramping up distribution of OPEC's existing oil reserves, we would have decades of steady supply, which is more than enough time to develop an effective replacement (should we be aware of needing one), and no matter what happens, if vanishing oil supply raises its cost, eventually any replacement will be cheaper than oil, and we already know the thresholds when that will happen, e.g. existing hydrogen economies can replace oil at two to four times the current cost per barrel (by equivalent energy output). That's not very high. And with enough demand (which guarantees big returns on investment), we could effectively replace the current oil economy in twenty years (the transition from coal to petroleum seafaring is a good model, and that was slower solely because the demand differential wasn't as steep--increase that differential, and the profit margins for a more rapid roll-out soar). Hence I foresee no difficulty even in the worst-case scenario, which is already very improbable.
Should We Be Hopeful? Part I
Tyro said... I think we must remember that the transition periods between energy sources or technologies can be rough.
Exactly. That's my "it can still suck" category, which I mentioned frequently. I'm not arguing we shouldn't endeavor to make things better and prevent things getting worse. I'm just saying no matter what we do (or don't do), the human race isn't in any real danger of extinction, and societal collapse isn’t much more likely.
But generally, just because it has always happened in the past doesn't mean it will always happen in the future.
Yes, it does--to a high probability when the facts so warrant, just like anything else. Basic induction. Just as we can be confident the laws of gravity won't change in ten years, so we can be confident that the next million technologies will progress at much the same rates and impact society in much the same ways that the past million have. There is, in fact, no evidence at all warranting any other conclusion. Realism must draw its conclusions from reality, not extremely improbable what-ifs.
It's certainly possible that the environmental damage could become severe enough that the world human population gets cut back to a small fraction.
As I said, too. My whole point was that this is very improbable (and extinction even more so), not that it was impossible. But you have to take probabilities seriously. The earth could be wiped clean of all life by a local supernova gamma ray burst tomorrow. That's actually a very real possibility. It's just so fantastically improbable there is no point in fretting about it. So mentioning that it (or anything else) is merely "possible" seems a rather pointless exercise (unless you have an affordable way to do something about it).
Bounce back? Eventually, probably, but what a rough time it'll be.
Hence my "it will still suck" category of possibilities. Already covered that in my blog.
Should We Be Hopeful? Part II
It's trivial to create a thought experiment where you give a population of people some massive amount of technology and some horribly degraded environmental condition (such as the desertification of the entire planet or the heat death of the universe) so that they all die.
I agree. That's my point: such scenarios are extremely improbable and therefore of no interest to realists like me.
Except we've already used up all the easily accessible resources when WE did all those things.
That's not only untrue, it's not even possible. All we can do is move chemicals around. We can't destroy them. That's why even Chernobyl's surrounding territory is thriving biologically: even the worst disasters don't take long for nature to filter, especially when you measure by centuries. Trees grow back. Rain refills lakes. Erosion dissipates toxins and replaces soils. Metals oxydize back into ores. We couldn't stop any of this if we tried. And it doesn't require "geological" time scales. It happens in mere centuries (in fact, in most cases, much less, as Chernobyl attests).
Hence a global collapse would indeed cause mass death (because without the present system, we can't support 95% of the people living here), but once population is normalized to what can be naturally supported (probably a few dozen million even at the present state of environmental depletion), everything starts advancing all over again, just as happened after the collapse of the Roman Empire. Based on that example, it took about a thousand years to get back to the same level of science and technology, and then only five hundred to get from there to here. Give any environment the same thousand years to regenerate, and almost all effects of human interference will have been erased and resources completely reset.
These guys will be working with sticks and stones (and whatever resources we may have left over from when we stepped up to something new).
First, that caveat kills your point. Retasking existing supplies in a collapsed civilization (metals, plastics, magnets, wires, dams, pots, knives, pumps, books, etc.) and knowledge (chemistry, metallurgy, engineering, medicine, etc.) would give a future dark age society an enormous leg up over previous dark age societies. Sorry, no sticks and stones.
Second, as I noted originally, the whole point is moot. Our ancestors started with sticks and stones. Look at what then ensued in just six thousand years. Hit the reset button and you have no reason to expect anything less will happen again. And eventually one of these cycles will get further than any previous, and so on again, just as the last cycle (1200-2000 AD) has gotten further than the previous (1200 BC-200 AD). Thus, as I said: to a very high probability further progress is inevitable, you can only quibble about how long it will take.
I do reserve a few eggs, though, for the possibility of social disruption, food shortages, etc. that may accompany something like a difficult transition to non-fossil fuels. It's entirely possible, or at least, in my opinion, possible enough to take into considerations.
I agree. But that falls in the "it still might suck" category of options I surveyed already. It's not human extinction. And it's not very likely. But I do agree affordable measures should be taken against such things. Since wise risk management always hedges bets. You just have to make the cost proportionate to the risk.
The "Manifest Destiny" of Human Technology
Shane said... I've never been one to believe this inevitable, "manifest destiny", of human technology.
Note that I did not say it is inevitable. I said it is inevitable given maintenance of current conditions (e.g. in the short-term, the persisting conditions of the past four hundred years; in the long-term, the persisting conditions of the past ten thousand years), which at present is highly probable, because the forces that could end it are (even collectively) highly improbable.
But if you focus on the low-probability alternative outcomes, you have that remaining probability-space split between two fields of alternatives: dark ages (which we've had many of in just the past ten thousand years) and extinctions (which have so far occurred every hundred million years or so), the latter being (as just indicated) vastly less probable than the former (by several orders of magnitude).
Of dark age events, all so far have not ended technological progress (in the long term), and therefore the probability that any future dark age will end such progress (other than merely delay it) is extremely low, even lower than human extinction.
Of extinction events, they are (as I argued) very improbable. Which does not equal impossible. But as far as risk-management, there is call perhaps for making affordable provisions against such threats, but no call for excessive worry. Especially since extinction events (which happen several orders of magnitude less often than dark ages) are very unlikely to drive humanity extinct precisely because of our technological knowledge and ingenuity (e.g. if dumb mammals could survive the last extinction event, intelligent, technologically ingenious mammals will surely survive the next one).
[you say] technology will inevitably continue to improve--no exceptions.
Just a reminder: I very specifically did mention exceptions. My argument is that they are either extremely improbable or can only delay the inevitable.
"technology can overcome any environmental loss" : I think this is, at best, too strongly worded. It is a mistake to simply conflate technology and resources. Technology is just knowledge, techniques, or tools that you ultimately have to apply to resources.
Actually, economically, these are identical. A lever that allows a man to load twice as much weight, in basic terms of resources, has doubled a resource: manpower. One man now equals two. Similarly, a new engine that gets 50 mph, compared to an engine that gets 25 mph, now allows the same distances (or kwh or whatever) to be produced from half as much fuel, which now has doubled a resource: fuel. One gallon of gas now equals two. Similarly agriculture: a field that generates twice as much food from the same resource input has doubled a resource: food. And this follows even more abstractly, e.g. plastics vastly increased a resource we used to identify as "wood."
Technology does a great deal more than that, but one of its most prominent effects, as we can empirically observe over the past 4,000 years, is the magnification of nearly every important human resource, by several orders of magnitude to date, with little sign of stopping. That is not conflation. It's fact.
Predicting the Future from the Past: Part I
Shane said... You can soften [your claim] to some lesser degree and it becomes a probability calculation of how likely some given technology is
Tosh. Give me any length of time (at least five hundred years from the current start date) and follow the overwhelming probabilities (i.e. no significant catastrophes), and the predictions are solid:
Supercapacitors: ~100%.
True AI: ~100%.
Cyclotron-manufactured basic elements: ~100%.
Solar-thermal desalination: ~100%.
Clean algal fuels: ~100%.
Genetically engineered people: ~100%.
Immortality: ~100%
Synthetic mind transfer to simverses: ~100%
Faster-than-light travel: ~0%
And so on.
We have more than enough information to make these predictions, especially given the stated conditions. It is simply folly to cover your eyes and claim we don't already have ample data for this.
...technology, by its very nature, is impossible to predict
No, it is not. It is difficult to predict, but not impossible. Though some things (by being presently unimagined) will not be predicted, others are entirely predictable.
Though people often focus on the failures of past futurists, they ignore all the successful predictions they also made. Yet most of them were not working from scientific models, like the current progression laws in computation and diode efficiency, and did not have the knowledge we have now, e.g. regarding the widespread game-changing social impact of increasing efficiencies in the information economy, miniaturization, specialized AI, spacefaring, etc.
It is also easier to predict near-term developments (like the impact of supercapacitors) and get their timelines reasonably correct (based on past timelines of similar innovations in the present industrial economy, from development to dissemination, e.g. the history of the solar panel, the computer, and the jet turbine all provide models), as well as long-term end-point developments (replacement of natural selection with genetic engineering in human reproduction, simverses, space colonization), as long as you forego attempting precise timelines, since the relevant technologies already exist and need merely be economized, and the past teaches us the ubiquity and timescales of economization in all technologies.
Simply extrapolating forward may be realistic, but it's by no means a guarantee of anything.
That's exactly what I said: timescales are more variable, and small improbabilities can intervene, and every technology will have an end-point limit, and often we can only talk probabilities, etc. But if you allow large timescales, or track smaller developments on shorter scales, and in either case stick to highly probable outcomes (e.g. by excluding improbable catastrophes), some extrapolations forward are guarantees. What can defy our predictions are not the failure of the predicted developments to transpire (the past teaches us we will almost always be right about those), but the effect of unforeseen technological advances (i.e. yet more progress than even we dreamed of).
For example, the full impact of the rise of the internet was largely unpredicted and has changed the game from what past futurists projected, and yet if you take that into account, their predictions were actually far less inaccurate than is supposed: they just didn't foresee the integration of what they did predict, with what they hadn't foreseen would also be invented.
Predicting the Future from the Past: Part II
Another common mistake, as I already noted in my blog, is to foreshorten the timescales for developments. Hence you get a film like 2010 in which the size and portability of computers falls far short of the actual pace of advance (in the film they are comically large and clunky compared to actual computers and devices of the year 2010, just as their projected monitor technology fell humorously short of actual), but at the same time predicted the development of true AI by now, even though all realistic, informed projections do not anticipate that being even possible before 2030. But errors in timescale are not errors in inevitability. There will be true AI. It's just a question of when.
Yet another common mistake is simply failing to consider practical realities that should have been known even in the futurist's day, and thus aren't a failure of technological prediction, but of political or economic prediction. Flying cars will never happen not because the technology won't be there, but because their safety can never be secured to any degree that would satisfy political opposition to their employment (beyond specialized use--and in fact we have those: we call them helicopters), and because they will never be able to compete with ground vehicles in efficiency (for obvious reasons, more energy will always be needed to keep a car aloft than to roll it along the ground, hence ground vehicles will always out-compete aerial ones in cost effectiveness no matter how efficient the latter get, except, again, when flying machines can do what ground cars can't, hence the uses we see make of the modern helicopter).
But if you are aware of these errors, you can avoid or compensate for them, and even with them, predictions are most often only in some degree inaccurate, not outright incorrect (e.g. we do have flying cars, we just call them helicopters).
I don't see how it is inevitable that we are going to reach any pinnacle of technology. We may get to a technological dead-end where we used up all the technologically accessible resources necessary to reach some further step.
If you are saying the potential for technological advance is unlimited, I have no objection to that in general. Or if you mean there will be, e.g., quantum mechanical limits to development in various specific areas, I already said exactly that--but those limits are vastly far away. So the room for intervening development is almost beyond conception, and therefore those limits are largely moot to any realistic futurism.
If you mean we’re close to some general technological dead-end, all past history, involving thousands of examples, and present in-progress developments, affording thousands more examples (e.g. home fabrication units, algal fuel production, supercapacitors), are uniformly against you.
Indeed, when you combine threads, you get exponential advances: compact nuclear power (as is housed now in submarines) + standard desalinator = unlimited fresh water (that whole process is already standard on nuclear submarines, and can be scaled up to serve a nation if we had the need and the will); solar power + cyclotron + space station + computerized fabrication = unlimited conversion of all solid and liquid waste into raw elements like gold or hydrogen or diamond, anything you want—again, all it requires is the scaling-up of already-existing technologies.
Human will might stand in the way. But ability does not.
Tyro said... Why is a large population in the West a problem? Because we consume vastly more resources than the developing world does of course, not only on a per-capita basis but also on absolute numbers.
This fact is misleading if you don't break it down into what's actually being measured. For example, most electrical power consumption in the U.S. is entirely generated with local resources (gas, coal, nuclear, solar, wind, hydroelectric, etc.) and therefore it does not take resources away from the developing world. To say we consume ten times as much solar-generated electricity, for example, would be a meaningless complaint.
Another way the comparison can be exaggerated is where high Western consumption is of benefit, not harm, to developing nations. This is most obvious in imported product (because we consume more, we buy more, and thus flood developing nations with cash) and exported labor (because we consume more, and outsource the labor to produce it, we actually expand third world employment, which boosts every sector of local economies, e.g. India's information sector and Mexico's manufacturing sector provide some of the largest and most crucial bases for their respective economies, entirely depending on high Western consumption).
And, not to be forgotten, exports as well. Consider food consumption: we actually produce far more food than we consume, so the developing world actually obtains a net benefit from our high food consumption. For example (just to pick arbitrary numbers), if we sell a surplus of 10% and consume 100 units of corn, the developing world gains access to 10 units of corn it otherwise would not have available, whereas if we cut our food consumption to a fifth (20 units), the developing world would lose 8 units of corn. The devastating effects of such a drop in consumption have already been mimicked by our wasteful diversion of corn product to useless biodiesel, which has caused a spike in the global cost of all food (not just corn), which reveals how dependent the third world is on our high levels consumption. If we lost that corn not to fuels but to just lowering on consumption, the effect on developing nations would be the same, i.e. just as bad.
Of course we could instead divert reduced consumption to increased surplus for export, which would be of even greater benefit, but we could only do that using the same high-consumption technologies we presently use. Hence again, in the end, the technology isn't the issue, human choice is.
Nick Bostrom
Haukur pointed me to Nick Bostrom's extensive analysis of the possibilities of future collapses or extinctions (Existential Risks: Analyzing Human Extinction Scenarios and Related Hazards). I've already commented on most of what he does, and we're in agreement on some things, and our disagreements will already be obvious from my main blog entry and throughout this thread. But some things he raises I didn't touch on, I can touch on briefly here:
(1) Nanobots and manmade germs (or even AI run amock): Extremely unlikely. (a) Any such threat would require massive sustained replication, and thus would exhaust its own resources before exhausting ours. (b) We already have superbly designed replicators, called viruses and bacteria, and they aren't killing the world. In reality, it's very unlikely we could build one much better. (c) Doomsday fiction aside, killer germs are impractical, precisely because once released into the wild you cannot control their genetic stability, nor anticipate all the biological defenses of the germ's targets, and let's be honest, once we devote suitable attention and resources to the problem, they are easy to isolate and kill. (d) And even in the extremely unlikely event all of those obstacles are overcome, we have the technology to evade the end result: we have the means even now to construct isolated self-sustaining environments on earth (the "bubble city" scenario) and in space (the "space colony" scenario). So you can't kill mankind this way, even if you tried.
(2) We’re living in a simulation and it gets shut down: intrinsically the least probable threat of all we can imagine. Bostrom suggests “a refutation...of [his previous] argument" that we're in a sim has yet to appear. Yet I wonder why, as it’s obviously fallacious (see next comment).
(3) Everything else he proposes: is either already addressed in my original blog (directly or indirectly), or just as extremely improbable.
But his advice (in the second half of his existential threats paper), for shoring up our preemptive defenses against risk factors, is sound, and should be heeded, since it would be of benefit regardless of the low extinction probability, since many comparable scenarios have a much higher probability of making things at least merely suck, and that should always be of real concern (as it is in more obvious cases already, e.g. global warming, germ warfare, nuclear proliferation, and asteroid detection).
Are We in a Sim?
In Are You Living in a Simulation? Bostrom sneaks in an implausible premise that a civilization capable of generating unlimited sims would use any of that resource-space to generate sims exactly like our world, rather than worlds more overtly intelligently designed (like game sims and other managed simverses are now), and even if it would run a sim like this universe instead (for some nefarious and unintelligible reason), it would have no reason to run that sim more than once (since once you've completed a sim, you don't need to run it again), yet his argument requires that it be run millions of times (for no explicable reason).
In mathematical terms, Bostrom hugely overestimates N, by ignoring the fact that almost all, if not all, sims run by an average civilization will not be ancestor-sims, and if any were, very likely only one would be, or at best a few, by analogy with the ability to replace a moon with a ball of cheese: merely having the ability to do that (even at a tiny fraction of the cost of the total resources available) does not entail a civilization will do that even once, much less millions of times. Even more in the sims case, given the severe immorality it entails, i.e. any Argument from Evil against the existence of God would establish all the civilizations Bostrom imagines as running millions of ancestor sims as evil, and therefore he implausibly assumes that most posthuman civilizations will be evil. And not only evil, but gratuitously wasteful of resources (why a million sims when you only need one?), and bizarrely motivated (why run any ancestor sim?).
Which is heavier? A pound of feathers or a pound of lead? Um, if there are infinite worlds and 10 simulated worlds in each of those worlds, there are not more simulated worlds than natural worlds. They are numerically identical since they'd both be infinite. It's 50/50 chance of ending up in either, so for people to assume that it is more probable to end up in a created world is fallacious.
Ben
War_on_Error said... ...if there are infinite worlds and 10 simulated worlds in each of those worlds, there are not more simulated worlds than natural worlds. They are numerically identical since they'd both be infinite. It's 50/50 chance of ending up in either, so for people to assume that it is more probable to end up in a created world is fallacious.
Well, I'll be honest, that's unfair to Bostrom's argument. You are making the mistake of applying arithmetical rules to transfinites, but arithmetic is invalid for transfinites. Pertinent to this query, no one to my knowledge has worked out how probability theory would apply to transfinites, so we actually can't say your conclusion is correct. One could just as easily argue that there are still ten times as many sims than reals, it just also happens to be the case that both quantities are infinite. As there is no axiomatized system that deals with this problem, your debate would go nowhere, as there is no available deductive proof for either position (that is, which is demonstrably consistent). Craig makes this same mistake with his arguments against an actual infinite.
But to give you an example of a counter-argument, if you are falling toward an infinite platform, and one out of every ten planks on that platform is missing, and the odds of hitting any one plank (missing or not) across that whole infinite expanse is equal, then it doesn't seem plausible to argue there is a 50% chance you will fall through the platform. Surely the odds are 1 in 10. If you filled in 90% of those gaps, the odds of falling through the platform would surely decrease (presumably to 1 in 100), not stay the same. And so on.
I could then go on to show there is a mathematical problem with this argument, but that would only show we don't really understand how to cope with transfinite ratios.
Sweet,
So glad you responded. I often run across this situation in thought experiments and was wondering if I was getting something wrong.
Anyway, I was under the impression that WLC's main problem was applying the rules of finite numbers to infinite sets. Maybe that's what you are saying, but it looks to me like your "counter-example" is contriving a finite set up to refute an infinite scenario. If someone is falling from a finite height, there's only so much surface area of an infinite plane you are actually dealing with. And the density (and consistency) of the plank ratio matters. However, in the simulated worlds scenario, it is much more (as in exactly) like spawning into a video game and no such finite parameters apply at all. You can put all the planks to one side of the infinite plane or do whatever, and it shouldn't effect the probability in any way shape or form. Any other counter-argument that sneaks in some finite aspect is going to fail for the exact same reasons. As I see it, it is straight 50/50 and the idea that one infinite can be "more" than another is incoherent.
So if you are really going to tell me that a pound of lead weighs more than a pound of feathers, I'm going to have "am I crazy" issues. :D But then again the Monty Hall Problem almost made my brain implode (I got better), so I understand I might just be totally wrong from some reason I don't see.
Please help!
Ben
WAR_ON_ERROR said... I was under the impression that WLC's main problem was applying the rules of finite numbers to infinite sets. Maybe that's what you are saying, but it looks to me like your "counter-example" is contriving a finite set up to refute an infinite scenario.
Exactly my point (hence my closing paragraph just above). We have no math to deal with this. Or so I assumed. Since I posted I had a nagging doubt and looked through the textbooks a bit more and realized I was wrong: arithmetic might actually be demonstrably valid for probability functions on transfinites (in a way that, for example, transformative operations like Hilbert's Hotel are not). But if so, then my last example is the correct analysis. I'll discuss this in a moment.
If someone is falling from a finite height, there's only so much surface area of an infinite plane you are actually dealing with. And the density (and consistency) of the plank ratio matters.
Actually, you can just change the laws of physics any way you want to make the stipulated conditions hold, e.g. have it so people don't fall in a fixed direction but any direction at random that is below the parallel. After all, it's just a thought experiment, not anything we're actually going to try and set up. You could even structure it as a problem in QM where there is a probability the photon will land at any location as soon as it is emitted, and then put film only at 1 out of every 10 locations, and calculate the odds you'll get a photo of a photon (which is close to the very situation I just described). The point is not the incidentals of physics, but what happens to probabilities when you change ratios in an infinite series.
You can put all the planks to one side of the infinite plane or do whatever, and it shouldn't effect the probability in any way shape or form.
And it wouldn't in the stipulated case (because, if you need a physical model to establish the parameters, it would be something like either of the two models I just proposed).
As I see it, it is straight 50/50 and the idea that one infinite can be "more" than another is incoherent.
Actually, that is demonstrably untrue. One of the few things that has been proven (deductively, formally proven) for transfinites is that there are infinities that are objectively bigger than others. Cantor demonstrated this, and thus originated the concept of cardinality. It's just that what makes an infinity larger than another is something more bizarre than any of the examples Craig, for example, discusses. You don't make it bigger by just adding things to it, not even adding infinities to it, but rather by multiplying it by another infinity, which reveals one clue: infinities are not in fact quantities as we normally understand them, (e.g. there is no "infinity" on any number line, much less all the higher cardinality infinities after that).
I'll follow this up with an analysis of the Craig case and why it's different from the probability case.
First, why Craig is wrong. Craig does things like this: imagine a library with infinite red books and infinite black books, you have infinite books, now remove the black books, you still have infinite books, in defiance of logic which tells us (supposedly) you should have less after subtraction, not the same quantity. He does the same with even and odd numbers.
The argument requires the following premise, which I’ll call Claim (A):
If set N is an actual infinite, and set S is an actual infinite, and S is a proper subset of N, then quantity N (QN) equals quantity S (QS).
Claim (A) results in contradictions when included among the basic axioms of mathematics, and in consequence of this it cannot be included as a mathematical premise without rejecting at least one fundamental axiom of mathematics--and since it is far less likely that the basic axioms of mathematics are false with regard to real quantities, it is far more likely that Claim (A) is false with regard to real quantities.
In demonstration of this:
N = All natural numbers
S1 = All even natural numbers (or, properly, S1 = N - S2)
S2 = All odd natural numbers (or, properly, S2 = N - S1)
And the corresponding (Q)uantities, QN, QS1, and QS2 are, by definition, all actual infinities.
According to (A): QN = QN, QN = QS1, and QN = QS2
p1) A - A = 0, Therefore: QN - QN = 0
p2) If A = B, Then B = A
Therefore: If QN = QS1 (as per claim (A)), Then QN - [QN] = QN - [QS1]
And: If [QN - QN] = 0, Then [QN - QS1] = 0
And: If QN - QS1 = [QS2] (which is true by definition), Then QS2 = [0]
Consequently, when Claim (A) is added to the basic axioms of mathematics, it allows that something can equal nothing (QS2 = 0). Since this is by definition impossible, one or more of the premises involved in arriving at this conclusion must be false. But which one(s)?
Statement [QN - QS1 = QS2] is necessarily true and therefore cannot be false. In fact, to deny it is to deny the existence of the set of all odd natural numbers, which entails denying the existence of at least one odd natural number that can in fact be produced, thereby refuting any attempt to deny its existence.
Likewise, statement [QN - QN = 0] cannot be denied without denying [A - A = 0], which is not only self-evidently impossible, but it would entail that all mathematical conclusions are false (since all depend on this axiom being true). In fact, proponents of the Kalam require it to be true in order to carry their argument against an actual infinity, so they cannot deny it without destroying their own argument.
Statement [QN - QN = QN - QS1] is also necessarily true, if [QN = QS1], unless you reject the law of commutability [if A = B, then B = A], but no Kalam proponent can deny this axiom.
Therefore, statement [QN = QS1] must be false, as it is the only proposition that remains (except the premise that N, S1, and S2 are all actually infinite, which cannot be in dispute because it is true by definition—as even Kalam proponents must allow without engaging in circular argument).
Since statement [QN = QS1] is Claim (A), then Claim (A) must be false. Since Craig’s argument requires Claim (A), his desired conclusion cannot be demonstrated. Since Kalam requires demonstrating it, Kalam cannot be demonstrated. QED.
Note that my argument here has little to do with the universe. Rather, it simply shows that a commonly held assumption about infinity must be false, if the basic axioms of mathematics are true, and that any argument based on that definition cannot apply to real quantities, because it cannot be reconciled with proven mathematics.
Most importantly, note that your own argument (against Bostrom) requires Claim (A) as well, and thus collapses just as surely.
Now to the next bit...
Bostrom relies on probabilities applied to transfinites. As it happens, probability theory requires this application: probability itself is defined as frequency extended to infinity (e.g. the true probability of a die roll equals the frequency of that roll after hypothetically infinite rolls, and using calculus with infinite trials as the limit this can be formally proven to be correct). This fact is central to modern statistical mathematics (polls rely on the proofs working for even infinite population sizes).
There are two ways to prove this.
The first is the same way mathematicians prove the axiom of infinity itself:
Start with a finite case: suppose there are 1000 universes, and 900 are sims. Bostrom's probability argument is then valid (it is, I argue, merely unsound, i.e. he cannot establish that 900 are sims, but we're not taking issue with that now, so I'll just grant he's right instead for the sake of argument). Now add 1000 universes in which, again, 900 are sims. Same ratio results. Do this again and again. Same ratio, it never changes. Since adding another 1000 never changes the ratio, repeating the operation to infinity can never change the ratio. Therefore, the ratio obtains even for infinite universes. QED.
This is the same argument as the proof for the axiom of infinity (which is the seventh of the nine fundamental axioms of mathematics in standard ZFC set theory): add one empty set to one empty set, you get two empty sets. Repeat the operation, and you get still more. Repeat and repeat, same and same. Repeat it indefinitely, and you get an infinity. Since there is nothing to stop you from adding and adding, there is nothing to stop you from reaching infinity. Therefore, infinity exists. Craig tends not to tell anyone the existence of an actual infinity has been formally proven. When he's called out on it, he makes some lame argument about abstract objects being different than real objects (thus revealing he actually is aware of the proof and knows better than to deny it when he gets caught). Or, if you leave an opening, he'll press the distinction between actual and potential infinities, but with simultaneity the latter always equals the former (hence he needs to assume the existence of time between additions).
The second is more direct, and I’ll put that in next...
The second proof starts with the premise and then tests the contrary and finds it false:
Probability = frequency. Frequency = the ratio of x(f) to ~x(f). Given any set N = {x1,x2,x3,...,~x1,~x2,~x3...}, there is a ratio of x(f) to ~x(f) equal to the number or fraction of ~x(f) that can be paired with every x(f).
Your contention would be that this would not hold for an infinite set (which we'll suppose N to be), because you can line up every x(f) and put it into a one-to-one correspondence with all ~x(f) no matter how many there are, therefore the ratio will always be 50:50 (except when either the number of x(f) or ~x(f) is finite, of course, in which event the ratio will always be infinitesimally approximate to zero).
However, producing a one-to-one correspondence is not by definition the same operation as producing a ratio, thus it is a conflation fallacy to convert the one into the other. If in N for every x(f) there are ten ~x(f)'s, and you extract one item from N and it is a ~x(f), you can pull one x(f) out of N to pair with it. If you repeat the operation indefinitely, you will exhaust all the x(f)'s by the time you've exhausted only a tenth of the ~x(f)'s, but you cannot do the reverse, i.e. even after infinite operations, there will still be ~x(f)'s left over, but no x(f)'s. This can be formally proven the same way as above.
Since you will only have the set completely empty after repeating the procedure ten times, this is what a ratio of 1:10 means for transfinite sets. That it's the same as for finite sets should be unsurprising, given that the previous proof just demonstrated that this is exactly what we should expect.
Though you can also perform an operation that lines every x(f) up with a ~x(f), given that both are infinite (an operating we might call Function P), a subtraction operation requires applying the axiom of separation (the third axiom of ZFC), which requires selection, which you can't conduct by performing Function P on set N. It requires performing a different function instead, which entails the 1:10 pairing. That is, pairing every x(f) up with a ~x(f) establishes an operation that cannot undergo subtraction, because subtraction requires application of the axiom of separation, which requires the 1:10 pairing procedure, not the 1:1 pairing procedure. You can show the 1:10 subtraction function to be formally valid, but you can't show the 1:1 subtraction function to be (for essentially the reasons I showed earlier: doing this requires Claim (A) to be true).
For example, you can select enough ~x(f)'s to pair with all the x(f)'s one-to-one and still have ~x(f)'s left over, but you can't do the opposite, i.e. you can't select enough x(f)'s to pair with all the ~x(f)'s one-to-one and still have x(f)'s left over. Work it all out, and you find every operation is similarly forbidden except the one that gives us a 1:10 ratio. You might think to cheat the result by splitting the x(f)'s into two groups (each infinite) and pairing one off with all the ~x(f)'s, thus removing all ~x(f)'s and having x(f)'s left over, but this violates the axioms of mathematics...
It can be proven that the set HN = {half x(f)} is a subset of FN = {all x(f)} and the axiom of choice (ninth in ZFC) entails that HN cannot be larger than FN, which entails if pairing FN with XN (the set of {all ~x(f)}) produces parity (i.e. subtracting these pairs empties N completely), pairing HN with XN can only produce parity (because the pairing operation FN:XN includes the pairing operation HF:XN, because HF is inside FN, so FN:XN is HF:XN, i.e. for every FN,XN pair, there will be exactly one HF,XN pair, because there cannot be more without violating the axiom of choice). I'll add one last post below explaining why this is.
But in short, a lot of the tricks Craig pulls on infinities are actually formally invalid, blatantly violating the axioms of set theory. But applying probability theory does not.
Okay, last bit: the axiom of choice and transfinites (if I now understand this correctly)...
In finite sets, if we have 8 cats and 80 dogs, we have a ratio of 1:10; if we split the cats between black and white and there are as many of each, we have 4 cats and 80 dogs, so we have a different ratio (1:20) and cats left over. It works the same for transfinites, as long as you use proper definitions and rules, which actually forbid tricks using "one-to-one" correspondence as a device to dink the quantities.
Where we slip is that the finite case assumes application of finite arithmetic, but it is a category fallacy to apply that to the transfinite case (and vice versa), because transfinite arithmetic is different. The two situations are not sufficiently identical for the same rules to hold. This has partly to do with our mistaken intuition that infinity is a quantity in the normal sense (e.g. that it has a location on a number line). But it is not, thus we can't treat it like one.
To see this more clearly, imagine TN = {one tenth of all x(f)}, and putting TN in one-to-one with XN, hoping to end up with ten times as many x(f) as ~x(f) even though it should be the reverse. Yes, doing this you can show that the ratio of TN to XN is 1:100 (and prove it by pairing TN members with XN members until TN is deleted and seeing how many times you have to do this to delete XN as well), but all that proves is that there are a hundred times as many members of XN as TN, it doesn't prove there are ten times as many x(f) as ~x(f), because the set of all x(f) is not the same set as TN. The error is in conflating the two (on the one hand, quantities in the abstract, and on the other hand, the contents of the sets) and then attempting to jump from the one result to the other, which works on finite sets, but is invalid on transfinites.
You can do the subtraction function for TN on FN and end up removing TN from FN and still have some of FN left over (and you could remove all of FN by repeating the same function ten times). You can even thus change the ratio now between FN and XN (as a tenth of FN is now gone). But you can’t use this trick to put the elements of TN in one-to-one with XN and claim now there are ten times as many x(f) as ~x(f). That’s an illegitimate move (i.e. it has no support from sound definitions and axioms). Because due to the axiom of choice, there are still more ~x(f) than x(f).
In the HF example, splitting the x(f)'s into two groups (each infinite) and pairing one off with all the ~x(f)'s won’t result in removing all ~x(f)'s and having x(f)'s left over, because you can’t legitimately perform subtraction this way. A valid subtraction function would simply result in deleting half of the x(f)’s in N (the HF set) and a twentieth of ~x(f)’s in N. Because for each member of HF, you can draw out one ~x(f), then remove all those pairs, and still have lots of ~x(f)’s left over, but HF will be empty. And in fact if there is one x(f) for every ten ~x(f)’s, the ratio of half the x(f)’s to all the ~x(f)’s will be 1 in 20, as you’d expect, so pairing up half the x(f)’s with as many ~x(f)’s will draw only 1/20th of the ~x(f)’s out of N.
Anything else is just cheating, using a method of subtraction that has no basis in sound logic.
Post a Comment