The End of Productive War
If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:
As I mentioned in my last post I just finished War! What Is It Good For?: Conflict and the Progress of Civilization from Primates to Robots by Ian Morris. As I mentioned in the review he ends up combining a lot of other books I’ve read into another history of progress and at times it seemed like he wasn’t covering much in the way of new territory, but he did introduce me to one new idea, which I thought was pretty interesting: the idea of productive war. Though I should also mention here, at the beginning, that he also acknowledges the existence of destructive wars as well. He doesn’t think all wars are productive
In Morris’ view productive war is war which consolidates nations and people into larger units with greater scope for cooperation, and, according to the central claim of the book, less chance of violent death. Morris’ assertion is that the chances of someone dying violently is in large part based on the size of the community they belong to. And that it’s an inverse relationship, the bigger the community the smaller the chance. So, for a member of a small tribe of hunter-gatherers their chances of dying violently was between 10 and 20%. If, on the other hand, they were a citizen of the Roman Empire or Han China then their chances of dying violently were in the 2-5% range, and for someone living in a modern, developed nation their chances are around 1%.
Accordingly as wars of conquest created larger communities, deaths went down, and beyond that as trade and commerce expanded, living standards got better as well. So while empires had to begin with a series of bloody wars in order to be created, in the end, through these productive wars they created zones of stability within the borders of the empire where everything was better. This has progressed on down through the ages until now we no longer have regional hegemons, we have global hegemons (Morris actually calls them globo-cops), first with the United Kingdom and then with the United States. Of course in between those two hegemonies there was the cold war where the Soviet Union and the US vied for dominance. And it is also in this period where we start to see the beginnings of the problem I want to talk about.
Historically, when two civilizations competed, eventually one of them triumphed over the other. When that happened the victorious empire absorbed the losing empire and created a new larger empire. Think of Rome and Carthage or even the United Kingdom and India. But lately such absorption, or it’s less brutal offspring, colonization, has fallen out of favor. When the US won the cold war we didn’t absorb Russia and create a new, expanded empire where cooperation, trade and lower violence flourished. Nope, we basically left them alone (though some would argue we wrecked their economy and then left them alone.)
This is not how it has generally worked historically. Generally when the victors conquered, they Conquered! And we certainly could have done that, particularly at the end of the World War II. (Though I’m not saying it would have been easy.) But we didn’t. By not doing that was World War II less productive in the sense Morris describes than it could have been? Is it possible that over a long enough time horizon that we might actually put it in the destructive column? To come at things from another direction, if gobbling up vanquished foes is no longer an option, how do we expand the zone of cooperation?
Morris asserts that having a globo-cop/hegemon works much the same way, but does it? Sure, a US hegemony definitely contains some elements of the imperial cooperation of the past, but, first, no one would look at current events and say that things were going well with Pax Americana. And second there’s a big difference between ensuring the continuance of global trade or acting as a policeman when nations get out of line and entirely absorbing a nation and its culture.
Modern morality has made this sort of absorption unthinkable. The US was the first empire to (mostly) eschew colonies. And since that time the idea of colonies and colonization has only become more taboo. Arguably there has been no shortage of American force projection, but it definitely doesn’t lead to colonies, nor is it practical in places much larger than a small failed state. It’s impossible to imagine the US invading and pacifying China or Russia in the same way that Rome pacified Gaul or the British pacified India. Meaning that, as the tide of US power flows out, it reveals entirely intact nations with more lingering animosity than lingering desire-to-compromise.
And, if some nation did want to go back to the “old way” of doing things and start absorbing other countries into a Greater East Asian Co-prosperity Sphere, like the Japanese of World War II, then that becomes a lot more difficult in a world where nukes exist. There was a time when you might have imagined India conquering and reabsorbing Pakistan. It was unlikely but not inconceivable, but with nukes as part of the equation that will never happen. Or it will happen, which is even worse. Meaning the good guys won’t do it for moral reasons and the bad guys are welcome to try, but it’s likely to end in mushroom clouds.
The way productive wars used to work is that there would be an initial, short-term spike in deaths, but that would be followed by eventual assimilation leading to integration and cooperation which raises the standard of living for everyone in the new empire. This sort of thing is no longer possible between two nuclear powers because there won’t be any assimilation after the initial spike of deaths because there won’t be anything after that initial spike.
I don’t want to overstate my case. I suppose it is possible to imagine a limited nuclear exchange, where there is still something left of both the conqueror and the conquered, but if this is the best case scenario, we’re in a lot of trouble.
More likely the presence of nukes and the reluctance to colonize might lead to a situation where unity actually starts heading backwards. If a part of a nuclear armed nation manages to secede while hanging on to some of those nukes, is there any scenario where the mother country would go to war to reclaim its lost territory if it knows those nukes might be used? Meaning that if nukes continue to spread we may end up with more countries and less cooperation.
All of this is to say, that the historical process of unification through the means of productive wars which Morris mapped out in the book appears to have stalled. We may have run out of steam right before the final sprint to the finish (a unified world).
Thus far we’ve assumed that achieving unity and cooperation can only be accomplished by means of productive war. And that seems to be Morris’ thesis, but might there be another way?
Certainly most people hoped that international cooperation would grown through peaceful means. That was the goal of the original League of Nations and the current United Nations, but is there anyone who still thinks that the UN will eventually create the level of cooperation we’re talking about? A true world government? Certainly I don’t. From where I sit the UN appears to be getting weaker with each passing year. Indeed, this decline makes a certain amount of sense. In the aftermath of World War II even the most bellicose nations could see the need for an international body to resolve disputes in a less bloody manner. But after 70 years without a great power war, the need for something like the UN is less and less obvious.
In the absence of nations voluntarily unifying, you could imagine that US influence continues to grow until we have a de facto world government. Or at least you could have imagined that at the end of the Cold War. Lately the idea seems laughable. At a minimum we would need some sort of motivation. As I pointed out in a previous post, external threats seem to help. Would Rome have been Rome without Carthage? How much of what the US did was because of the USSR? (space race anyone?) But at this point it seems that regardless of how Russia and China behave our taste for empire is gone, and it’s not even clear that we can keep the “empire” we have, to say nothing of continuing to expand it in the way Morris imagines.
Which leaves us with a couple of possibilities:
As I mentioned in my review of the book, the possibility Morris favors is that we’ll pass smoothly from an American hegemony to an AI singularity. That Pax Americana will become Pax Technologica. Here’s how Morris describes it:
Everything will hang on the relative timing of the shift from the Pax Americana to a Pax Technologica and the mounting difficulties that the globocop will face—if current economic trends continue—in doing its job. I suggested earlier that in the 2010s and probably the 2020s too, the United States will remain largely unchallenged, but as the 2030s, 2040s, and 2050s go on, it will find it harder and harder to overawe rivals. I also noted that the majority opinion among the futurists is that merging with the machines will reach the Singularity stage in the 2040s. If all of these guesses are right, we perhaps do not have too much to worry about. The world will become increasingly troubled, polarized, and tense as we head through the 2020s, but the globocop will remain strong enough to handle the stresses. As we enter the 2030s, the globocop will be feeling the strain, but it will by then be pulling back anyway as the Pax Technologica begins to make violence irrelevant to problem-solving; and in the 2040s and 50s, just at the point that the globocop ceases to be able to cope, the world will no longer need its services. All will be well.
It would be nice if “all” was truly “well” and things proceeded exactly as Morris describes, but I think he underestimates the number of things that need to go “right” in order for this to happen:
America has to maintain the peace until an AI or something similar is ready to take over. Morris estimates they’ll be able to do that until sometime in the 2030s or maybe a little later. Given current events I’m not sure I’d agree with him that the US is “largely unchallenged” even now, and I’m even more doubtful that will be the case over the next decade.
Pax Technologica, whatever it’s form, has to be ready to step in as soon as the US starts “pulling back”. Morris has said it will “[begin] to make violence irrelevant to problem-solving” in the 2030s. This also seems far too optimistic, particularly since we appear to be headed in the opposite direction. Thus far, our best guess is that machine learning and AI are actually making problem-solving of all strips harder.
Perhaps technology will get better and it will switch to lessening rather than creating conflict. That’s still a long way away from replacing everything that goes into making America the lone superpower. Which includes, among other things, the $639 billion dollars we spend on defense. To replace that we not only need the singularity, we need a rather impressive singularity.
Morris says that the “majority opinion” is that we’ll reach the “Singularity stage” in the 2040s. This is by far the most optimistic of his predictions. Even Kurzweil, who’s optimistic to the point of being delusional, is saying it won’t happen till 2045. Perhaps in 2013, when the book was written, the majority opinion was the 2040s, but these days most experts are predicting later than that. And these are not predictions of “When will AI be able to take over as the world’s super power?” But more along the lines of, “When will AI be able to replace human surgeons?” (Average answer: 2053)
Which takes me to my final point. What does it mean to “take over”? As I pointed out, Morris appears to have a very specific idea of what that means, and it’s very different from what most people imagine when they talk about AI. But even if we end up with an AI exactly as powerful as Morris hopes, and it happens soon enough to step in for Pax Americana before it collapses. He’s ignoring the whole field of AI risk, which makes the very salient point that we can’t be sure a superintelligent AI will be benevolent.
If we reject the Pax Americana Pax Technologica transition for the reasons I just listed, and we accept Morris’ thesis. Then that tosses us back into the realm of war. We’ve currently got a globo-cop keeping that war at bay, but many people, including Morris, think we’re getting near the end of that. Meaning that the other possibility remaining to us is actual war. Actual war is bad enough in the short term, particularly since, for all the reasons I’ve laid out, this actual war is unlikely to be one of the ones that’s eventually productive. We’re much more likely to see destructive wars, similar to what followed the collapse of the Roman Empire.
Morris doesn’t spend much time on this second possibility. Probably because he thinks it’s unlikely to happen. In many senses despite his different outlook he’s still very much in the same school of thought as Steven Pinker. And both appear to believe that the arrow only points in one direction. In particular Morris claims that the 500 years of European colonial expansion from 1415 to 1914 were the most productive wars in the history of humanity. That Hitler was something of an aberration, and that in any case since that time we’ve had the long peace, which is further evidence that we’re in the final act and there will be no more destructive wars. And indeed, the finish line does seem really close, but that doesn’t mean we’re going to cross it. In fact for all of the reasons I mentioned above it feels like the very progress that has gotten us to this point won’t work for the final five yards.
One of the frightening things Morris points out is that a period of destructive wars often follows a period of constructive wars. That a particularly confident nation will conquer all of the surrounding territory unifying it into a larger area where trade and cooperation flourish, but that at some point the nation/empire(/ideology?) runs out of steam. Whether this is because of exhaustion, over-expansion, bureaucratic bloat or something else, the empire can no longer defend all of its territory. When that happens, whatever unity it achieved is lost to the destructive wars which inevitably follow as a consequence of this exhaustion. If Morris is accurate and we just finished 500 years of constructive wars, then even if we didn’t have nukes and an aversion to expansion through colonization it might be time for the pendulum to reverse itself in any case. Also, while it seems difficult if not impossible to have constructive wars if nukes are involved, they’re perfect for destructive wars.
All of this would mean that Pinker and Morris are wrong. (And indeed I’ve asserted that very thing.) And I’d rather not jam a second book in here, right at the end, but I just started reading Only the Dead by Bear F. Braumoeller which was written as a direct refutation of Pinker’s thesis, going so far as to say that it may end up having the opposite effect from what he intended. In support of this claim he includes an excellent quote from one of the reviews of Better Angels:
[T]here is something deeply unsettling about the argument of this book. While I began reading without either smug comfort in my own circumstances or indifference to the violence that remains, by Pinker’s final sentence on page 696 it was impossible to muster any other reaction. Indeed, I want to suggest that Pinker’s book produces the type of reaction that conceivably could stop this important trend dead in its tracks. A world of elites and foreign policy decision makers well-schooled by Pinker in the causes of the decline in violence would be a world unmotivated to work to sustain it.
The logic laid out in the quote seems straightforward enough, but Only the Dead goes on to cite studies which show that as nations become less willing to go to war they actually end up going to war more often. I’ll go into this more when I get around to reviewing it, but add everything together and we seem unlikely to have seen the end of war. And when it does return it appears unlikely to be productive war either, even if we can look past the terrible near term costs.
To be fair to Morris the book was written in 2013, and a lot has changed since then. The election of Trump has made a lot of things written beforehand seem quaint and even naive. Which is not to say that things are that much worse now then they were in 2013, just that we appear to have had significant movement on the catastrophe track without that much movement on the singularity track. This is important, because Morris, unlike Pinker, acknowledges that there will be war. He just thinks having a globo-cop can keep those wars productive. He’s also more realistic than Pinker about how long the US can serve in this role. Where his optimism is equal to or greater than Pinker’s is with what comes after. And it all hinges on the next couple of decades.
Morris hopes that the 2030s will be a decade where the US can still mostly “overawe” its opponents while at the same time “every year will see more [technological] change than happened in the whole period between the 1980s and the 2010s.” And that’s what brings us the Singularity. That rather than descending into destructive war, we’ll narrowly thread the needle between all the potential catastrophes. As I said this is what Morris hopes will happen. I hope it happens this way too, but I would bet a lot of money against it. Anyone want to take me up on that bet? We’ll know who’s right in just 10-20 years.
You have to wonder if there’s any similarity between war and blogging. Is there also a productive phase of blogging? Do bloggers eventually get exhausted? Perhaps running out of things to say? Does the blogging then become destructive? Was my blogging ever productive? If you think it was or still might be, consider donating.