Impact of India-Pakistan nuclear war on x-risk?
Last month I was involved in a conversation thread about what the impact of a hypothetical nuclear war would be on existential risk.
There are many potential nuclear war scenarios which would have varying impacts on existential risk. It’s difficult to know where to start to gain an understanding of the long-term of nuclear proliferation.
For concreteness, consider the case of an India-Pakistan nuclear war.
According to Local Nuclear War, Global Suffering by Robock and Toon,
India and Pakistan, long at odds, have more than 50 nuclear warheads apiece; if each country dropped that many bombs on cities and industrial areas, the smoke from fires would stunt agriculture worldwide for 10 years.
[...]
1 billion people worldwide with marginal food supplies today could die of starvation because of ensuing agricultural collapse
Note that this would presumably cause some degree of chaos in the developed world.
I have not yet investigated the credibility of the papers’ claims. However,
Suppose that an all-out nuclear war between India and Pakistan were to occur and were to result in climate change killing 1 billion people. Then would the probability of a positive singularity increase or decrease and if so why?
This question seems very difficult to answer; maybe altogether too difficult for humans to answer. I welcome responses raising relevant considerations even in absence of a good way to compare the relevant considerations. Please read the linked conversation thread before commentating.
The question is not the right question to ask. Large scale war whether nuclear or not regardless of the countries increases existential risk in all forms. The more resources taken up dealing with such situations the less spent on preventing existential risks such as large asteroids, superbugs and very bad AI. The increased stress levels to societies will also encourage risk taking liking it more likely that people will try to develop major new technologies without adequate safeguards. Nanotech and AI both fall into this category. (To some degree this is the worst case scenario . If technological progress is halted completely this won’t be a problem. The really bad case is where technological research continues but without safeguards.)
The question as phrased also emphasizes climate change rather than other issues. In the case of such a nuclear war, there would be many other negative results. India is a major economy at this point and such a war would result in largescale economic problems throughout.
A slightly larger scale problem is that of total societal collapse, or human extinction. Both of these look unlikely in the Pakistan-India case but are worth discussing (although at this point seem very unlikely for any plausible nuclear war scenario). One serious problem with coming back from societal collapse that is often neglected is the problem of resource management. Nick Bostrom has pointed out that. to get to our current tech level we had to bootstrap up using non-renewable fossil fuels and other non-renewable resources. If the tech level is sufficiently reduced it isn’t obvious that such a bootsrapping can occur again.As more and more resources are consumed this problem becomes more severe. (This is in my view an argument for conservation of fossil fuels that is too often neglected- we need them in reserve in case we need to climb back up the tech ladder again.) But again, this situation doesn’t seem that likely.
Overall, nuclear war is an example of many sorts of situations that would increase existential risk across the board. In that regard it isn’t that different from a smallish asteroid impact (say 2-3 km) in a major country, or Yellowstone popping, or a massive disease outbreak or a lot of other situations. Nuclear war probably seems more salient because it involves human intent. This is similar to how terrorism is a lot scarier to most people than car crashes.
Agree with most of what you say here.
No, if technological progress is halted completely then we’ll never be able to become transhumans. From a certain perspective this is almost as bad as going extinct.
The Robock and Toon article estimates 20 million immediate deaths from an India-Pakistan war, less than 5% of the relevant population (although presumably among the most productive such people). This number is roughly consistent with extrapolating from data from Hiroshima and Nagasaki. In light of this, would you guess that the India-Pakistan specific economic disruption would be greater than the economic disruption caused by a billion deaths due to starvation?
Do you know if anyone’s attempted an analysis of the issues relevant here? On the most crude level we can look at the amount of oil that’s been used so far.
Agree, but I think that the probability of nuclear war is higher than the probability of the other possibilities that you mention. See http://lesswrong.com/lw/75b/link_brief_discussion_of_asteroid_nuclear_risk/ if you haven’t already done so.
A halt to technological progress would be temporary. Would you rather have a twenty year halt on new technologies or a massive rush of new technologies that end up destroying everyone?
There are a variety of different factors going on here. The immediate deaths are one problem. Subsequent further deaths will also occur among the refugee populations and will spread disease and the like. The resulting panic will also create economic damage. Most of the people who would starve due directly to the climate change are people in areas like sub-Saharan Africa who don’t have that major a role in the world economy. Their deaths would have a comparatively small impact on the world-wide economy.
Bostrom has mentioned doing this sort of thing in detail before I think, but if he has done it I haven’t seen the result. There are a variety of different factors that would go in. One obvious thing is that even as we have a fair bit of oil and other fossil fuels left, they are in much harder to reach locations. They are generally deeper in the ground, or farther out to sea, or simply harder to extract. So looking at the total reserves will underestimate the total problem. One related issue that would really need to be examined in detail would be the issue of metals. We’ve mined a large part of the world’s metal reserves. But for some of those, this is actually a good thing if a collapse occurs. Aluminum for example is very hard to extract from ores (it was at one point in the 19th century more expensive than gold). But although the technology to extract aluminum from ores is difficult, the technology to process pre-existing aluminum is much easier and we’ve helpfully left large quantities of it just lying around.
I agree that nuclear war is one of the more likely scenarios. The asteroid possibility is also much less of a worry now than it was a few years ago since WISE is now tracking most large near Earth asteroids and it looks likely that none are in really bad orbits for a few years. I don’t think that I have enough data to evaluate whether nuclear war is by itself more likely, but the chance of such a war specifically between Pakistan and India is the more relevant issue here than nuclear war in general. For example, if Iran and Israel get into a nuclear war, at most probably 10 or 15 bombs will be dropped, so the climate result will be unlikely to be that large, and for similar reasons most of the other problems will be on a smaller scale. It wouldn’t surprise me too much if Israel has explicitly but non-publicly precommitted to Syria and Iran that if any city in Israel is nuked they’ll nuke both Damascus and Tehran without waiting to find out who was responsible. If so, that would potentially increase the total problem but still not by that much.
There are more nightmarish scenarios, like a China-Russian war, or worse a US-Russian war. But it seems at this point that the US has first strike capability on China and has close to first first on Russia, and Russia has close to first strike on China (although obviously, nuking such a near neighbor could be a bad idea). I’d also estimate a not at all small chance (say 5-10%) that not all of the former Soviet countries have actually returned all of their nukes or nuclear material (Kazakhstan, Belarus and the Ukraine are the most obvious possible examples), but those situations are unlikely to lead to more than a single detonation.
The upshot seems to be that although I’m not sure of the numbers, I suspect that the a Pakistan-India nuclear war is more likely than many other large-scale disaster scenarios but probably not larger than the totality of other large-scale catastrophes.
Thanks for the excellent response. I’m familiar with much of the content but you’ve phrased it especially eloquently.
I’m not sure specifically about IndoPak nuclear war, but I think the general drift of this post is important—lots of Xrisk comes from factors we haven’t thought extensively about (I don’t think this particular area has come up on LW before).
There are two models for thinking about Xrisk. In the first model, which I associate with SIAI, most Xrisk comes from a small number of enumerable threats: uFAI, asteroid impact, etc. In this model, to prevent extinction, one should consider each threat in turn and design measures to protect against each.
In the second model, which I personally favor, the majority of Xrisk is distributed among a large number of threats. Each individual threat is highly improbable*, but because such threats are so numerous, net Xrisk is significant. In this model, it is impractical to design specialized defenses against specific threats. Instead, one should design countermeasures that protect against entire categories of threats.
* More precisely, the events may be probable but the likelihood of the event causing existential disaster is very low. IndoPak nuclear was is moderately probable, but it seems highly unlikely that such a war could pose a serious threat to human survival.
Marginal to significant decrease (in “probability of a positive singularity”). The recovery would monopolize global attention and effort.
ETA: The probability of “all-out nuclear war” between India and Pakistan is very low (and predictions of a nuclear winter will make a negligible difference to the odds of it occurring). Do you know the difference between strategic and tactical nuclear weapons? A strategic nuclear arsenal is there to deter other nuclear powers from destroying you, by giving you the means to destroy them in return, even after your own nation is in ruins. A tactical nuclear weapon is one which gets used on the battlefield. For example, Pakistan would potentially use nuclear weapons against the Indian army, in the event of a massive invasion. There are very strong incentives against escalation from tactical nuclear exchange to strategic nuclear exchange.
You seem to assume a very rational evaluation of the situation on both sides, even given a highly escalated situation like the destruction of the opposite army by means of nuclear strikes. If the people who run Pakistan and India are that rational, doesn’t that mean that people who are able to design AGI’s, capable of undergoing recursive self-improvement, shouldn’t be at least as rational and therefore take the risks associated with their work seriously? And even if the problem is simply that they don’t know about the associated risks, given that people in general do keep great care that they and their country are not destroyed, wouldn’t it be most effective to tell them about the risks as it would provide a strong incentives to make their AI friendly?
My personal opinion is that people are not as rational and estimating as you seem to believe. Take for example the German attack on Russia. The Soviets were surprised because their intelligence had no information that the German army was trying to acquire winterproof machinery and therefore concluded that the Germans wouldn’t attack them, that would be crazy. Or take the Yom Kippur War. Israel’s intelligence knew that nobody in the region could win a war with them at that time, it would be crazy to start a war with Israel, therefore they concluded that war is unlikely.
Everyone forgets that people are batshit crazy.
An even more important factor is that a disparate set of people in a state apparatus do not even combine into something as (in)coherent as a human being.
At present, both India and Pakistan have more than 100 nuclear warheads each. The relations are cold and (as an Indian living in India) there is much hatred. Militants based in Pakistan actively wage an insurgency in the Indian region of Kashmir.
To directly answer the specific question ask I think that that sort of thing IS a singularity in the sense that it’s a point in time that current rules and understandings do not let us see through. There is simply no amount of theorizing that will let you look from now through that sort of event to what is on the other side. There are climatological impacts beyond the article referenced, political effects, etc. Too much depends on the response from inside the developed nations. Do they focus on technological solutions to the problems? Do they waste precious time blamestorming and second guessing? Do they turn their back on that part of the world and “let nature take it’s course”?
Too hard to guess, the current practices and attitudes break down in the gravity of that sort of event.
It seems to me that answering the question about the impact of such a war on a singularity requires and extremely detailed model of such a singularity, to the point that I think this question is pointless. For example, would such a war impact Moore’s law? Does the rate of hardware advance impact the sign of the singularity?
Maybe discussion of the intermediate consequences of such a war would be useful, but no consequence claimed by any one on either thread is precise enough to use as input.
A lot of people seem to be jumping from delay to less likely singularity. While, ceteris paribus, delay is astronomical waste, and delay allows, say, asteroids to prevent the singularity, ceteris is not paribus. Anything that large event that causes a delay probably has a much larger effect on probability.
The lack of ICBM capacity for either side makes nuclear weapons in the hands of Pakistan and India effective as MAD deterrence due to the simple fact that any use of such weapons is likely to be nearly as destructive to their own side as it would to the enemy. Nations simply do not engage in suicidal behaviors. Not intentionally. Not even excessively religious groups.
That being said:
So you’re postulating that there would be a global—outside-of-India/Pakistan—death-toll of 1 billion persons? Well, I can see that regarding China… but as to the impact of “a positive singularity”; it seems that there would be a negligible impact on this. The areas where most research and development is ongoing for these possibilities are outside of the primary impact zones for this sort of thing. We already grow far more food than is actually needed to feed the planet; reducing the demand for food by nearly 25% would only make that problem far less significant.
In fact, the long-term impact would probably be beneficial if we assume the direct correlation of CO2 to GMT modulation is as the IPCC tells us; it would prevent both China and India from outstripping the US as primary CO2 contributors.
Can you substantiate this claim?
… that non-ICBM nuclear weapons would be ‘nearly as destructive’ to the user as to the enemy, of geographically adjacent nations?
India has roughly 80-100 weapons. India has been focusing on low-yield devices (with most apparently falling in the half-kiloton range.) Given how nuclear fallout works and the like… India and Pakistan launching their arsenals at one another would result in a large contaminated area affecting both nations.
Now, if you’re talking about the effectiveness of MAD deterrence … if MAD was ineffective the world would have glowed in the dark after the Cuban Missile Crisis.
Edited to ad: In this I was mostly just following my from the sub-post I was responding to in the general context of the OP, I sort of took the bit and ran with it, it’s not a really honest response to the original question.
Your first paragraph indicates that you may not fully understand the sorts of weapons systems that are available. I don’t have up to date information, and if I did I probably wouldn’t be allowed to share it, but there is a class of weapons called “theater ballistic missiles” that includes a sub-category of “Short range ballistic missiles”. These range from about 600 miles (1000k) to about 2300 miles (3500k), and are nuclear capable. At those ranges even large nuclear weapons are feasible.
Nuclear Weapons are a bad idea, especially the sort of “Primitive” stuff that (I understand) Pakistan and India have, but they are not the world destroying manifestation of Shiva the Destroyer that popular literature would have you believe. Yes, an all out “exchange” between the US and the USSR would pretty much be a civilization ender, and would stand a significant chance of wiping humanity and most of the rest of the species off the face of the planet.
Pakistan has, according to Wikipedia, about 80 to 120 warheads of unknown effectiveness. India’s is about the same size (same source says 80 to 100).
This is a serious concern because that’s a lot of extra radiation floating around, but in terms of primary effects (blast, EMP, the worst of the fallout etc.) it’s really not a whole lot, and it’s pretty far away from major centers of trade, major (for the world and for the “west” areas of critical production (we could easily spin up industrial production lost in other parts of the world). The terrain in Pakistan and much if eastern India (including the disputed Kashmir regions) is really rough terrain and would serve to limit blast, fallout and the spread of contaminated material through blowing winds and drainage—at least much more so than the rest of India.
When you look at the list of tests so far conducted (http://en.wikipedia.org/wiki/List_of_nuclear_weapons_tests) you see well over 2000 tests (admittedly spaced out over a much longer time). This suggests that even if they both emptied their arsenals you would not have a world-wide threat of extinction, just hard times. We in the west would largely be insulated from the worst of it because we have things like water treatment facilities, green houses etc.
The lost of life would be horrific, and the fallout—both nuclear and political—would take decades to deal with, but we would survive rather easily—some rough years, and those of use who believe in modern, pluralistic societies with freedom of speech and religion (or no religion as it were), as well as privacy right etc. would have to stand up and hold forth for all we’re worth, but it’s not a straight up existential risk.
Unless.
Unless it’s another Franz Ferdinand moment.
Now, you can label what comes next “bigotry”, or “prejudice”, or “experience” or whatever, but:
I believe that India is a fairly stable western-style democracy with several STRONG pressure groups that are at serious odds with each other. I believe that they acquired and field a nuclear arsenal because that’s what world powers do, and they believe in the deterrent effect of them. They aren’t eager to get in a fight, they want to build their nation up to be a modern, powerful country, sort of an eastern interpretation the modern pluralistic representative democracies we enjoy.
I believe that many people in Pakistan seriously want to join the rest of the world community, and want, as much as most Americans, Europeans and Asians to live a peaceful, commodious life. However there is clearly a large segment—maybe not a majority, but a powerful minority with strong influence in the ISI and in the Pakistan government that has a much stronger belief in Islam and in Islamic mysticism than they do in what we could call “modern, pluralistic societies”. This second group of Pakistanis is MUCH more loyal to Islam than to any government here on this planet.
Given those positions, my thinking is that it would be Pakistan to initiate the first strike, and even if they did not, they would not present the conflict as a battle between two nation states, but as a religious battle between the West and Islam—at least publicly. This seems to be their (Islam’s) standard response to anything that happened after the Reconquista.
This could:
Cause the muslims in India to rebel and and try to take over the rest of the country. This makes any sort of reconstruction MUCH more difficult, and could cause problems along the India/China border as well.
Cause one of the other Islamic countries in that region to attack, or to escalate their attacks on Israel.
Spread to other Islamic countries causing a more generalized initiation of hostilities against “The West”.
Spark riots in urban enclaves in much of Europe, the U.K. and Detroit.
(further) Destabilize Iraq, Libya, Lebanon, Egypt and Tunisia. and any other country (Syria?) that tries to liberalize/modernize (yes, Egypt is questionable at this point, but I’m being hopeful rather than realistic on that front).
This could lead to a serious all-in world war. Fortunately if it breaks out like this probably China and Russia will fall on the same side as the US—in this scenario it’s basically the Islamic World v.s. all comers. The problem here is that the vast majority of Europe’s energy needs are met from Saudi Arabia and Iran, with potential domestic sources being much smaller than the US. This means that Western Europe stands a very good chance of “going dark” under this scenario—we (the US) would have to scale up our oil and bio-fuel production rather rapidly, and since much of this requires infrastructure that isn’t amenable to throwing money at it.
None of those pose—as a first or second effect—an existential risk, but it’s going to re-arrange the world stage significantly, and third and fourth order effects could cause other flareups.
What would worry me more would be a Sino-Indian war—There is a history there, both countries are very male heavy at this point (male/female balance is very important for maintaining good social order. If it gets too far out of balance you get violence and related problems).
Should China and India come to blows over stupid shit we (US and Western Europe) wind up in a situation were our two biggest industrial centers (India and China) are in conflict and many nations might wind up taking sides. This could precipitate (as a side effect) the nuclear exchange this thread is based on, but would also have much wider consequences.
Then I suggest you read further. The Wikipedia link I gave specifically referenced ‘short range ballistic missiles’ as part of the Indian arsenal. SRBM is not ICBM. And even if it were; the targets a Pakistani/Indian war would engage in would still not be long-enough range for it to matter.
Given the kiloton yield of the weapons involved, and the persistence of radiation—not really. You’re looking at well under a million statistical deaths globally.
One might as well hypothesize that an Indian/Pakistani war cause the aliens on the dark side of the Moon to destroy us out of a fit of pique. It’s equally as likely.
Any nuclear interchange between Pakistan and India will almost certainly be isolated to those two nations. There simply aren’t any strategic mutual defense treaties that would extend to such an interchange. The history of warfare, furthermore, since the advent of nuclear weapons has been one of de-escalation of the scope of conflicts, not of broadening of them.
I discounted the possibility of this concern from the outset. I can see no rational basis for the assertion that Muslims of sufficient organization as to be in control of a military and nation would willingly engage in behaviors they knew would result in the destruction of their organization. Even al-Qaeda still exists, “ten years after”.
A historical footnote: black-market nuclear weapons have been available to whatever degree for almost fifty years now. There is simply no valid count of how many nuclear weapons the Russians ever actually made; and it is known that several of their weapons are ‘simply missing’. Furthermore, Russian scientists have actually been convicted of selling weapons-grade radioactive isotopes to criminal elements.
And yet, there has never once been a nuclear terrorist attack; there hasn’t even been a radiological one. Given the fact that multiple generations of such opportunity have existed, there has to be a reason why this is so.
I apologize for misinterpreting your first paragraph, however the wiki links you refer to were posted after I responded.
I may be over-worried about the actual fallout from an exchange in the range we’re talking here. Most of my understanding comes from military training in the context of a US/USSR exchange in the late 80s, and doctrine that weren’t re-written after that stopped being a concern.
However, I do think that political ramifications and third and forth order effects will be MUCH worse than you write about. Yes, in a certain sense nuclear weapons have saved millions of lives by lowering the intensity of conflicts, by making massing of troops dangerous, forcing people to the table rather than continuing to press the fight etc.
But all that happened after a VERY limited and one way nuclear exchange.
I think that the possibilities of a nuclear exchange between Pakistan and India are vanishingly small, but if the world winds up structured more than that becomes likely, then there have been significant changes in the geopolitical arena. We have no precedent for what happens when you get bombs going in both directions, and especially when you get a large quantity of bombs in both directions.
You have to be completely delusional or under immense pressure to order a launch of nuclear weapons. This means once it happens history is pretty much not a good guide to what comes next.
The Middle East is going to bear the brunt of the fallout, if not radiological then politically and economically. If the factories of India go offline, then they neither have the need for, nor the revenue to pay for the oil they buy from Iran, Saudi Arabia and other middle eastern countries. Those countries no longer get the revenue they need to buy food, mostly from India, Pakistan and China. India will have a HECK of a time producing enough food to feed itself, Pakistan might simply cease to exist as a nation (depending on how good our intel is about India’s nukes. Even 60 or 80 KT of nukes spread out properly in a country the size of Pakistan, and with it’s concentration of industry and agriculture might simple render it like Somalia was for much of hte last 2 decades—or worse. China, being an immediate neighbor, will wind up with the worst of the shadow effect trashing their agriculture (this could be good for the US, but only if we do the right thing and fast).
These are destabilizing forces, and they will happen at a time when the world will have more on it’s plate than it can really deal with.
I hope I’m wrong, and that you are right in that players in that region are more rational and thoughtful, but I don’t think our leaders are sufficiently rational and we at least make a passing attempt at vetting them for psychological problems other than narcissism (which seems to be a requirement of the job). Heck in the last decade we had a member of the House of Representatives that denied the existence of compound interest, a minor party candidate (libertarian) who was blue from drinking colloidal silver (no, really http://www.washingtonpost.com/wp-dyn/content/article/2006/11/11/AR2006111101004.html). Given the rather common beliefs in that area of the world, I don’t think it’s safe to assume de-escalation.
I screwed that up. Apparently retracted doesn’t delete, and if you keep typing it retracts all of it.
Here’s what I wrote, it should be easier to read:
I apologize for misinterpreting your first paragraph, however the wiki links you refer to were posted after I responded.
I may be over-worried about the actual fallout from an exchange in the range we’re talking here. Most of my understanding comes from military training in the context of a US/USSR exchange in the late 80s, and doctrine that weren’t re-written after that stopped being a concern.
However, I do think that political ramifications and third and forth order effects will be MUCH worse than you write about. Yes, in a certain sense nuclear weapons have saved millions of lives by lowering the intensity of conflicts, by making massing of troops dangerous, forcing people to the table rather than continuing to press the fight etc.
But all that happened after a VERY limited and one way nuclear exchange.
I think that the possibilities of a nuclear exchange between Pakistan and India are vanishingly small, but if the world winds up structured more than that becomes likely, then there have been significant changes in the geopolitical arena. We have no precedent for what happens when you get bombs going in both directions, and especially when you get a large quantity of bombs in both directions.
You have to be completely delusional or under immense pressure to order a launch of nuclear weapons. This means once it happens history is pretty much not a good guide to what comes next.
The Middle East is going to bear the brunt of the fallout, if not radiological then politically and economically. If the factories of India go offline, then they neither have the need for, nor the revenue to pay for the oil they buy from Iran, Saudi Arabia and other middle eastern countries. Those countries no longer get the revenue they need to buy food, mostly from India, Pakistan and China. India will have a HECK of a time producing enough food to feed itself, Pakistan might simply cease to exist as a nation (depending on how good our intel is about India’s nukes. Even 60 or 80 KT of nukes spread out properly in a country the size of Pakistan, and with it’s concentration of industry and agriculture might simple render it like Somalia was for much of hte last 2 decades—or worse. China, being an immediate neighbor, will wind up with the worst of the shadow effect trashing their agriculture (this could be good for the US, but only if we do the right thing and fast).
These are destabilizing forces, and they will happen at a time when the world will have more on it’s plate than it can really deal with.
I hope I’m wrong, and that you are right in that players in that region are more rational and thoughtful, but I don’t think our leaders are sufficiently rational and we at least make a passing attempt at vetting them for psychological problems other than narcissism (which seems to be a requirement of the job). Heck in the last decade we had a member of the House of Representatives that denied the existence of compound interest, a minor party candidate (libertarian) who was blue from drinking colloidal silver (no, really http://www.washingtonpost.com/wp-dyn/content/article/2006/11/11/AR2006111101004.html). Given the rather common beliefs in that area of the world, I don’t think it’s safe to assume de-escalation.
Have you tried retracting then following a link to your ‘retracted’ comment then clicking delete? Kind of annoying but it seems to work. I’m not sure if there are any limitations on it.
No, but I will keep that in mind.
Thanks.
No.
Without further information, I must assume that means 1 in 7 people will die, which means there is a 1⁄7 chance that EY will die (and a similar chance for each of us), which would do great harm to hopes of a positive singularity.
To put it another way, not only is there the issue Mitchell mentioned (that everyone will be focused on disaster recovery), but there is a reasonable chance that anyone who ever cared about, or even thought about, the Singularity would all be dead. And if not dead, much more likely not in a position to do Omega’s work, processing power be upon it (PPBUI).
Whoops, I guess I missed the “today” part.