Can we survive technology? (1955) [pdf]
drive.google.com
drive.google.com
"The carbon dioxide released into the atmosphere by industry's burning of coal and oil-—more than half of it during the last generation—may have changed the atmosphere's composition sufficiently to account for a general warming of the world by about one degree Fahrenheit." (Page 512 / 666 / 9)
“The ever accelerating progress of technology and changes in the mode of human life give the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue.”
which some people worry about.
To me, his thoughts on the weather are some of his most interesting things to read just from the fact even someone with the brain power of Von Neuman could be utterly and completely wrong.
It is the ultimate lesson in the fallibility of what you believe to be true.
One measure of the progress of a civilization is efficiency with which ideas can be communicated. And the crux of what von Neumann perceives as danger is "runaway" technology. We attempt to solve one problem, like controlling the weather, by introducing remedies according to our current state of the art capabilities. And accidentally introduce cataclysms that cannot be reversed.
Now that capability is accelerating, the same fears arise anew. But it's a different world today. Tech is not developed in secret large scale government run labs of yore. Its too pricey. The culture has shifted as well. From ICBMs to private space enterprises. And it's that cultural feedback loop, that civilizing progress as ideas are shared instantly about the globe, that very few were able to foresee.
Real technology (advancement of science vs software+marketing) is still very much developed by governments, because they are the only entities willing to spend billions on possible dead ends. Fusion reactors, CERN, nuclear technology, and those are the ones we know about because we're allowed to know. The US alone spends about 80-90 billion per year on "Special Access Programs." [1]
> From ICBMs to private space enterprises.
"China is building more than 100 new missile silos in its western desert, analysts say" [1]
I'm also trying to be an optimist about the future, but part of that is understanding where we actually are. Technology is still firmly in the control of governments and a handful of corporations.
[1] https://www.thedrive.com/the-war-zone/29092/special-access-p...
[2] https://www.washingtonpost.com/national-security/china-nucle...
Fusion power itself will likely not have much practical application anytime soon, because other power sources will be far more economical.
But if we consider fusion power to be a backdoor way to get funding to research superconductors, plasma physics, and so on, then there is something valuable in the manner of that von Neumann quote, even though their proximate goal is probably all but useless for now.
While I agree, isn’t the flip side of this that _bad_ ideas can be much more efficiently communicated as well? I know the counter is that the best remedy for bad information is better information but at least when poor ideas are amplified by human cognitive biases, is transmission efficiency a good measure of progress?
* If iterations of will in the end come out "correct" then speeding it up will help progress in process up until a limit. * Conversely if we cannot come across the "correct" answer eventually as a species we were already doomed and it isn't like we are making matters worse. (Putting aside the definitional issues of technically already being doomed in an entropic universe.)
I stick to binary outcomes here by design due to "optimal" being not only non-computable but non-definable.
There is a popular argument about "bad information traveling faster than good" but it doesn't quite convince me. The speed may also be its own undoing via overexposure undermining its credibility. There is a notable historic pattern seen in England, Boston, Ireland of theocrats getting in charge and so taxing everybody's patience they cause declines in religiousity athiests could only dream of. Respectively Cromwell and the puritans being so bad people went back to nobility, "Banned in Boston" censorship, and theocratic abuses including the Magdalyne Laundries.
My reading of this article is that von Neumann was talking about carrying capacity, and not some runaway or rogue technology from exploding. We are just too big for this planet. He says this over and over.
I find it harmful to equate mere proliferation with democratic participation. Sure, you have many terminals, but most of the information processing is still lopsided towards the centralized "server". When recommendation algorithms shove content down people's throats, only consequential participation expected from them is to engage with the ads, and they can't talk back to ads. The rest is almost as "democratic" as watching TV.
> One measure of the progress of a civilization is efficiency with which ideas can be communicated.
And it is a bad measure.
A psychotic person's mind also communicates tons of ideas back and forth, in fact too "efficiently". Propaganda posters from an airplane also communicates very efficiently. The idea that mere communication makes progress is what we've been thought by the engagement-maximization culture.
Real measure of progress is the extent those ideas converge towards the truth. How they can form a pattern of meaning that increasingly conforms to the reality. Current tech cares about none of those, and no wonder there is a crisis of meaning making and acceleration from departure from reality, just like a psychotic person.
> Now that capability is accelerating, the same fears arise anew. But it's a different world today. Tech is not developed in secret large scale government run labs of yore.
It is just developed in secret by large scale internet companies now. For most of the tech most of the people have the most screen time with, what content is recommended why, what data is processed how etc are the biggest secrets.
Running mainly a few apps/websites tightly controlled by a few companies strongly incentivised (by greed and their governments) to keep a tight leash in the type of content that can be consumed AND created. I think Orwell and Huxley foresaw many of these perfectly clear even before the term "cold war" existed.
- More censored.
- More commercialized.
- Less open, more walled-gardened.
- More banal (For example, academics used to write long Usenet posts, then they started to write blogs, now they mostly bicker on twitter)
- Less sincere and personal, now people mostly use it as just another PR tool.If you dont conform, you will be cancelled, downvoted,banned, etc.
Now, before someone comes to accuse me of "Old-man-yelling-at-the-clouds syndrome", I fully accept that in absolute terms we have more good things now, but the signal/noise ratio is also a very important thing, or as papa Stalin used to say: "Quantity has a quality on its own" , for every thoughtful post,blog,site we are drowned in thousands of crappy sites and it is not always very easy to distinguish them, because it is the sad reality that quality publications will little by little become crap just to survive or fit-in.
That internet node not only isn't an equal peer in the global information network (thanks to simple things like NAT, but also more insidious problems like increased battery drain when P2P apps are active and the OS killing background apps "for your convenience"), but it is also a telescreen that you carry at all times and even pay for its upkeep.
You can keep this democratisation of technology, if that's what it looks like.
> Since most time scales are fixed by human re- action times, habits, and other physiological and psycho- logical factors, the effect of the increased speed of technological processes was to enlarge the size of units— political, organizational, economic, and cultural — affected by technological operations. That is, instead of performing the same operations as before in less time, now larger-scale operations were performed in the same time.
We call von Neumann architectures to the very devices that have allowed to operate--by proxy, so far--in different timescales. And we do tons of stuff these days in less time, thanks in part to von Neumann.
Had he lived today, I would imagine he would be still concerned with timescales and sizes of physical places, but from an entirely different perspective.
https://en.wikipedia.org/wiki/Gustafson%27s_law https://en.wikipedia.org/wiki/Parkinson%27s_law
And if we’re being precise, aren’t we all on modified Harvard systems these days? :P (I am not a computer scientist)
“While in college, he had worked on a military base in Albuquerque, and he had showed his superiors how to run certain computer programs a hundred times faster; instead of saving time and money, the bureaucrats ran a hundred times more equations. He concluded that the goal of government officials was “not so much to get answers as to consume the computer budget.”’
This as some justification for becoming a libertarian-activist billionaire, but it seems more likely this is a universal experience than Mercer plagiarizing von Neumann.
[1] https://www.newyorker.com/magazine/2017/03/27/the-reclusive-...
I would think someone who goes on to found a hedge fund would understand that if you drastically lower the cost of something which produces a valuable result, the effect should be to increase the demand for that value, not just shrug and say "no, we only do 1 of those per day".
So I think rather talking about whether these small states will use nukes, we should talk about when this will happen. 30 years? 50? 100? 500?
MAD policy appears to be effectively over, or at least there is no longer any mention publicly, perhaps it is just assumed. While it can appear crazy, hence the name, it was a rather ingenious way to solve the problem, albeit we know now we have come dangerously close at times, at least we knew the risks and the lines that had to be crossed.
The next point and somewhat related is the change in nature of nuclear weapons. The variable ability of tactical nukes to change their explosive size, has resulted in a modern range of weapons that aren't as destructive as their elder brethren. It may seem like a good idea but the issue is it also opens the possibility of their more active use. A state may consider it acceptable. This could result in the reverting to an older policy of nuclear weapons. The MacArthur plans to use them in active combat, it becomes too easy to dial the Nuke down, target a military base and say "sure it's ok it's a legitimate military target".
That to me is the danger in the modern world. Attitudes, at least Western are still as a weapon of last resort but I can't predict the future and say that will always be the case or that other nations will share the same sentiment.
Nuclear weapons in the modern world are useful deterrants, and that's it: a dictator with nuclear weapons has no external enemies they could consider targeting without guaranteeing total annihilation in response.
The one exception here is that using a nuclear weapon on your own country and people might be a useful exercise of power...but even then, it's abundantly clear that the most likely outcome is still total annihilation - the world won't tolerate a nation lighting off nukes against targets in it's own borders much either.
Why do you think so? Famines seem to work just fine within borders.
I'd argue that nuclear weapons are no less safe in representative democracies. The only country to use nuclear weapons against another nation was a representative democracy. The only country to use nuclear weapons against another nation multiple times was a representative democracy. The only country to use nuclear weapons against civilian targets was a representative democracy.
Perhaps accidentally stepping over the line is the most likely use scenario.
MAD is still going strong. It's not military based, but economical.
It's the reason why theories about intentional release of COVID-19 are bunk.
China is worse off than it would have been without the virus.
Same goes for nukes. If Xi or Putin decide to nuke the US , people will assault their palaces in 3-6 months for food and water. Even if the US somehow fails to respond.
The global economy is interconnected to a degree that makes it impossible for a leader to order a nuclear attack and not have his own people remove him violently
Only if you measure progress in economic terms. Nothing I've seen contradicts my belief that they are now geopolitically better off than before the pandemic.
Quality of life of the population
If you leave quality of life on the table because of geopolitical games you are shooting yourselves in the foot.
Even during massive state sponsored hate campaigns (such as the one promoted by Hitler against the Jews) the population spends just a bunch of minutes per day actually fuming and hating...the rest of the 24 hrs they live their lives.
Quality of life is a much more potent motivator to support the leader/party than hatred.
That's why any sane country and leader would never attack an economically important country intentionally.
Violent wars are relegated to Africa and the economically unimportant areas of the Middle East such as Yemen. With some small outbreaks here and there in ethnically hot areas such as Ukraine
Exponential tech and social peace of a basically unaltered species do not go well together. Atomic capabilities are basically just one of the thousands of facets of this problem.
Imagine air-borne-electric taxis being prevalent in a future city. A 9/11 event might just look like a hack into traffic control and rerouting a thousand flying batteries to smash into buildings. With great power, comes great incapability for responsibility.
I wonder if there isn't a massive difference of likelihood between remote (IT) and physically deployed terrorism. For example, it may be easy to mess with smart cars by tampering with electronic billboards or leaving devices by the road to "fake out" cameras. But it's always been easy to leave dangerous things on a hidden part of a road (over a hill, around a sharp turn). It doesn't happen because very few people want to hurt random people, and those that might do it impulsively have plenty of time to rethink while going to the location, waiting for a chance, and then placing the trap. So I'm not worried the future will have more location-based terrorism or dangerous pranks. Even biological.
I'm worried about easy and remote terrorism. Where people don't have to see or even think about the victims, and there's little time for reconsideration between planning and acting.
Needs citation: I wonder if the need to be sneaky in a location is itself a guard rail. It requires the would-be terrorist to imagine how other people might catch them and react, which maybe triggers the social part of our nature. And maybe it only works if they imagine people being present and with faces, instead of disembodied actors on a network.
The armageddon scenario might be much easier to trigger than von Neumann believed.
Even 'just a few' 'small' nukes could kick up enough dust to the stratosphere to severely disrupt sunlight and weather, where 'severely disrupt' means (probable) total global social collapse.
The result was at most an cooling of the global temperature of 0.1-0.2 Celsius. But the "multiple uncertainties mean we cannot say for sure".
https://www.carbonbrief.org/guest-post-did-bombing-during-se...
So we need about 5-10 times larger bombings when ww2 to cool the global temperature with one degree. But we have increased the global temperature with one degree thanks to global warming. So you need at least 10-20 times larger bombings then ww2 to get a tiny atomic winter. And probably a lot larger to get any catastrophic cooling.
To get to nuclear winter levels of ash and dust you probably need dozens to hundreds of detonations and burning cities.
[1] https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1002/2017...
These days the threat model is broken because its preciously not needed to be a nation state to develops some major biological/chemical/drone etc attack with new technologies that on the battlefield would be called force multipliers.
So for those of you who also care about surveillance, I try to explain this is part of the reason for its increases... now when the threat model has evolved but you havent caught up, the rational response is (not saying right, but it makes logical sense on the surface to MICCIMAC) to surveil everything.
I have a great book on the post nuclear world I keep picking up to reread sections of that talks about the political transitions that will happen because of this that is worth a look: "The Shield of Achilles" tldr - corporations are going to take over the gaps left by increasingly ineffective government.
It is interesting the JvN starts off on a political-economy note and proceeds to geopolitics considerations, yet doesn't really address the drivers of "industrial revolution" or "empire". He opines that "[a] much more satisfactory solution than technological prohibition would be eliminating war as "a means of national policy.""
There are alternatives, John. Maybe we should work on technology to temper the desire -- construed as "divine right" -- of the few to lord it over the many. That's one alternative. War is rarely a "national" concern. Historically it is the means employed by "princes" and "kings" and "queens" to assert dominance over other egomaniacs. "Technology" of course is used to whip up the masses to serve as canon fodder.
Another alternative is developing robust ethical systems that drill into the heads of budding technologists that their humanity is not worth a paycheck.
Can we survive technology? (1955) [pdf] - https://news.ycombinator.com/item?id=20724363 - Aug 2019 (48 comments)
The implication then is of progression from countries to blocs, from blocs to super-blocs, from super-blocs to an inevitable world state when the use, and potential use, of weapons becomes too great for even a continent size union to bear or credibly defend against.
I imagine that a future with a substantial space population will look very much like the past when the Steppe nomads could destroy more advanced civilization by just deciding they should.
The difference is that repopulating 10 billion people takes rather longer than the tens of millions that were usually killed in the past.
A future where the equivalents of ‘Steppe nomads’ could throw around significant of mass would be a future where every habitable planet and fixed space facility would be slagged in short order, i.e. even more unstable and untenable.
If you want to read a better exploration of the larger issues and possible answers to this question I recommend: http://gator.uhd.edu/~williams/downloads/skinner82.pdf
"Most thoughtful people agree that the world is in serious trouble. A nuclear war could mean a nuclear winter that would destroy all living things; fossil fuels will not last forever, and many other critical resources are nearing exhaustion; the earth grows steadily less habitable; and all this is exacerbated by a burgeoning population that resists control. The timetable may not be clear, but the threat is real. That many people have begun to find a recital of these dangers tiresome is perhaps an even greater threat.
Why is more not being done? Within a single generation, we have made extraordinary progress in the exploration of space, genetic engineering, electronic technology, and many other fields, but little has been done to solve what are certainly more serious problems. We know what could be done: We could destroy all nuclear weapons, limit family size, and adopt a much less polluting and less wasteful style of life. The mere listing of these steps is enough to show how far we are from taking them."
Overpopulation wasn’t solved by intelligently overcoming evolutionary sexual urges but by invention and distribution of contraceptives combined with personal entertainment technology (“electricity is the most powerful contraception”), etc.
I find documents like that still highly influential but in many ways orthogonal to how we need to solve our current problems. Some lessons still apply but other points proved painfully wrong and sometimes backwards or at least irrelevant.
People need to update their thinking from 70s era ideas like those of Limits to Growth. Degrowther mindset in many ways is a rebirth of those 70s era ideas, and I think Degrowth as a movement is largely counterproductive for addressing climate change (although the sense of urgency is important!).
Last I checked, the threat of nuclear war is not over. Sure it doesn't seem as likely today, but the bombs still exist and if our current prosperity was to halt, it seems their use would become more and more probable. By capitalism do you mean globalization? I agree governments have lost much of their power due to how powerful current corporations are, which was enabled by technology that allowed globalization. So there is less incentive for war nowadays. However when things get though.. cue all the impending changes that climate change will bring... I'm not so sure the threat won't reemerge.
> Overpopulation is no longer a threat in most of the world.
It seems overpopulation is as much of a problem today, as it was back then. It is not mentioned directly, but every time you hear about global warming, water scarcity, over-fishing, the great pacific garbage patch, etc. All of these things are problems of scale. Or do you believe that with 100 thousand humans we would have these problems?
I agree with the oil, but also notice your omission of pollution. As for the larger point of resource depletion, I'm not so sure we are any better than we were in 1980.
Overall I think you are missing the forest for the trees. The purpose of the paper is to highlight how our behaviors are selected for a past when they were beneficial, but currently we are changing our environment at an increasingly faster rate due to technology (and this is the point of the von Neumann paper also). Our methods for dealing with the problems that arise are not good enough.
Taken literally, then the human race does not currently have any problems, otherwise there would be at least one of the problems we have ever had which is not solved.
But what I think you are actually arguing sounds tautological to me: for every problem that has been solved, it has been solved by "progress". I don't know if you mean technological advancement or social change or something else, but it appears as if progress is defined in terms of solving a problem.
All that is left is for you to show that every problem has an achievable solution to warrant the optimism in "progress". Otherwise, it just seems like your optimism is based in the premise that the human race has solved problems in the past; persistent, long-standing, and current human problems notwithstanding.
2 years ago https://news.ycombinator.com/item?id=20724363
When Watt made his steam engine, coal use went up. Uber was supposed to lower congestion and miles driven but they went up. We repeat the pattern over and over. The "father" of the Green Revolution, on winning the Nobel Prize said
> "The green revolution has won a temporary success in man’s war against hunger and deprivation; it has given man a breathing space. If fully implemented, the revolution can provide sufficient food for sustenance during the next three decades. But the frightening power of human reproduction must also be curbed; otherwise the success of the green revolution will be ephemeral only.
> Most people still fail to comprehend the magnitude and menace of the “Population Monster”. . . Since man is potentially a rational being, however, I am confident that within the next two decades he will recognize the self-destructive course he steers along the road of irresponsible population growth…"
He understood the value of technology as well as anyone in theory and practice. He recognized that we have to change our values from growth to enjoying what we have, from externalizing costs to responsibility, and so on or technology will keep exacerbating and augmenting our problems. That pattern includes nuclear, electric vehicles, and space travel. They may solve local problems, but they exacerbate the systemic effect. Electric vehicles, for example, make sense as a tactic under the strategy of reducing vehicles. Nuclear makes sense only as a tactic under the strategy of lowering power consumption. But we keep valuing producing more cars and energy.
This community seems over and over again to miss the systemic and unintended side effects. If we had clean fusion under our current values, we wouldn't live as we do today only cleaner. We would grow again until we hit more limits and had to live again under scarcity or natural calamity. Borlaug worked the second half his career to help people see the consequences of helping solve only part of our system but not the system.
If we have to stop growing at some point, at least acknowledging the laws of thermodynamics, the sooner we do it, with the smaller number of people, the more abundance we can live with. The more people we have, the harder to limit ourselves. If we don't, nature will, as we all know.
> If we don't, nature will, as we all know.
In a way, it's already happened, it's just that there is a delay effect and it takes time for things to... unravel ("the future is already here - it's just not evenly distributed yet")
I recommend this lecture https://www.youtube.com/watch?v=7WRojM9_uAI by Pablo Servigne (agricultural engineer, PhD in ethology and behaviour of complex systems).
I'd say we've basically known since at least https://en.wikipedia.org/wiki/The_Limits_to_Growth (but in many ways much earlier than that) and have simply been in denial. Yes we adapt and leverage technology and innovation, but there are rather hard time limits and we've exhausted our window quite some time ago. The correction (to use a friendly word) will be very painful.
We sometimes do so protect. We have national parks and protect endangered species, for example. We can do plenty more. Jevon's Paradox and rebound effects don't have to happen.
Improving efficiency of coal burning and extraction can increase use of coal, especially if no alternative exists and it is the bottleneck for economic development. But if you develop a more efficient technology that doesn’t burn coal, it’s not going to automatically mean more coal will be used. Natural gas, wind, and solar have all increased in absolute terms in the last two decades in the US while coal burning has halved in absolute terms, reducing overall absolute grid emissions in the US.
Anyway, just wanted to point that out. (But I will say Jevons Paradox is a good argument against just relying on improved internal combustion engine efficiency in hybrid cars and buses vs using another energy source for transportation entirely, like renewable and nuclear electricity for trains, electric buses and electric cars.)
In a tightly interconnected global economy it is disingenuous to disconnect the energy usage of one country or region from another.
US coal usage goes down because the things we used to build with that energy are now built in China. Solar panels in the US are cheaper than ever because manufacturing has switched to being 80% Chinese production, and guess where the energy to build those solar panels comes from? Coal.
The only way to understand energy is at a global scale as you can see here: https://ourworldindata.org/energy-production-consumption
As you can clearly see no source of energy has decreased overtime globally.
Looking at local energy production is just an accounting trick, but the planet doesn't really care. You can't combat global warming by saying "look, technically we didn't produce that energy on our shores!"
If you forced China to decrease its use of fossil fuels you would immediately feel the impact on the global economy. Just because the power plants aren't within our borders doesn't mean that's not energy being consumed for our benefit.
It's almost as if a population of apes that evolved to live in groups of 200 individuals maximum and has survived only by constantly spreading to new areas once resources are depleted can't run a sustainable planetary society.
Every single trait that we need to run a sustainable planetary society is selected against in the local incentive system of both evolution, capitalism, and technological advancement.
Frankly, our greed is eating this planet. And we are greedy because the greedy strategy worked best (evolutionarily) up until now, and greed was baked into our genes by all the selection pressures of the many systems we are a part of and set up ourselves.
Efficiency is our only way out of it. Farm acreage shrinking isn't from shortage of housing but an abundance of productivity.