The Better Angels of our Nature
May 28, 2015 7:06 AM   Subscribe

An interactive look at the deaths of WWII and the relative peace that has followed Highlights include the sacrifices of the Soviet Union, the toll of past atrocities and a breakdown of holocaust deaths. Numbers are adjusted to world population at the end.
posted by laptolain (44 comments total) 37 users marked this as a favorite
 


Before we get sidetracked into a discussion of Pinker and Taleb (an argument that is inherently undecidable, because none of us knows when or if the next Big War will happen), can I just say that this video is amazing and should be watched? It's a tribute to the human mind's ability to mix symbolism with reality that I could feel my gut tightening as I watched the tower representing Soviet casualties grow... and grow... and grow...

Thanks for the post.
posted by languagehat at 7:40 AM on May 28, 2015 [8 favorites]


That was awesome. Very educational.
posted by Renoroc at 7:48 AM on May 28, 2015 [1 favorite]


At the end of the video the narrator hand-waves the question of the causes of the "new peace" away in order to shift to a reality-distorting comparison of the scale of recent conflicts relative to world population. But doesn't that cause tell the whole tale? The Great African War, the Iran-Iraq war, the Cambodian Genocide, the Syrian Civil War, Putin's Invasion of Ukraine, America's invasion of Iraq, and so on, all show that massive industrial conflicts are still perfectly within our capacity. Why have we resisted them? I know this has been said before about Pinker but it seems like it should be emphasized:

We haven't slaughtered each other en masse because of the constant threat, not just that millions might die horrible deaths in violent conflict, but that a renewed conflict would put an end to all human life on earth.

The "better angel" of our nature is the Nuclear Executioner, holding an axe to the throat of humanity, threatening to wipe out all human history. And that guy is twitchy, unpredictable, and one worries that his arms are growing tired.
posted by dis_integration at 7:48 AM on May 28, 2015 [11 favorites]


I'm not sure I can make it to the end. I may try again later.

This is a really thoughtful presentation of mass death. I have read a fair amount about WWII but have never felt I had a real feel for the scale of death. These huge numbers feel meaningless. This makes it more real.

I had the same reaction to you languagehat about the representation of the deaths of Soviet soldiers. All those boys...
posted by latkes at 8:02 AM on May 28, 2015


Really great. The way it morphs into a clock at the end is stunning.
posted by rlk at 8:05 AM on May 28, 2015 [2 favorites]


For what it's worth, it kind of "lightens up" at the end.

Excellent video.
posted by Alex404 at 8:23 AM on May 28, 2015


.
posted by Gelatin at 8:23 AM on May 28, 2015


Outstanding data visualization. And, yeah, the "plot twist" of the clock at the end gave me goosebumps. That simple idea puts the viewer right into history. Brilliant.
posted by mondo dentro at 8:49 AM on May 28, 2015 [1 favorite]


Regardless of the ideology being put forward, that's really amazingly well done data visualization.
posted by codacorolla at 8:49 AM on May 28, 2015 [1 favorite]


my first visualization gut punch was when they started showing German military casualties, and then segmented off the number of Wehrmacht casualties in Stalingrad and it was the same size as the entire whole of American or British military losses for the entire war.

Like, take the entire number of troops that the Americans lost in WWII -- Pearl Harbor, D-Day, Iwo Jima, the Bulge, Italy -- and just wipe them all out in a single five month battle. All of those fathers, brothers, sons, husbands who just went into the white expanse of Russia and never came back.

Then they started doing the Soviet losses and I was all, "oh, man, this is going to be SO horrible. That Stalingrad body count's going to be like a random month in the opening phases of Barbarossa."

Also, showing Yugoslav, Polish, Chinese casualties pre and post invasion was simple, effective, and chilling.

Thanks for the post.
posted by bl1nk at 9:24 AM on May 28, 2015 [1 favorite]


The Great African War, the Iran-Iraq war, the Cambodian Genocide, the Syrian Civil War, Putin's Invasion of Ukraine, America's invasion of Iraq, and so on, all show that massive industrial conflicts are still perfectly within our capacity.

Well there are two things going on here: one is that interstate wars are getting less severe in terms of casualty. The other is that civil conflicts are on the rise and have been for some time (aided in part by the fact that they last longer). There are two primary reasons why interstate war has become less violent: democracy and nuclear weapons. Democracies don't fight each other and nuclear weapons make war too costly to fight. Any study on trends on causalty figures from political violence should not disaggregate deaths from civil and interstate wars.
posted by MisantropicPainforest at 9:25 AM on May 28, 2015


Any study on trends on causalty figures from political violence should not disaggregate deaths from civil and interstate wars.

I don't think I understand your point. Surely we want to know both the separate totals for civil wars and interstate wars AND the combined totals? How does it help clarity of understanding to insist that they always be aggregated?

The video, by the way, gives totals for both post-WWII civil wars and interstate wars. Factoring in civil war deaths does not change the basic picture: we're living through one of the most peaceful eras in human history and we're mostly remarkably unimpressed by that fact.
posted by yoink at 9:32 AM on May 28, 2015


If you are going to make inferences about declines in political violence by looking only at interstate wars or only at intrastate wars, you will be making bad inferences.
posted by MisantropicPainforest at 9:35 AM on May 28, 2015


if the places where wars occur became strongly bifurcated after WWII i.e. no wars in Europe (and other places), then comparing the total deaths to the total world population is obviously meaningless...

Outstanding data visualization. And, yeah, the "plot twist" of the clock at the end gave me goosebumps. That simple idea puts the viewer right into history. Brilliant.

but it ends with a conclusion which is junk. if they had studied the total war deaths in Indochina after WWII, to the total population in Indochina, you would get a totally different story. Same thing for the Congo, the Yucatan, etc.
posted by ennui.bz at 9:46 AM on May 28, 2015


Then they started doing the Soviet losses and I was all, "oh, man, this is going to be SO horrible. That Stalingrad body count's going to be like a random month in the opening phases of Barbarossa."

That line goes up and up and up and up.

Having recently rewatched the excellent documentary The Atomic Cafe, I wondered again if we didn't avoid a nuclear war because the Soviets knew what it was like to lose a couple tens of million people. Both sides had their hawks and doves, yes, but I can't help but wonder if the cooler heads that prevailed weren't mostly on the Soviet side, and not just Stanislav Petrov and Vasili Arkhipov.
posted by Gelatin at 9:50 AM on May 28, 2015


then comparing the total deaths to the total world population is obviously meaningless...

Why? Why is the rate of politically violent deaths meaningless?
posted by MisantropicPainforest at 9:51 AM on May 28, 2015


if they had studied the total war deaths in Indochina after WWII, to the total population in Indochina, you would get a totally different story.

And if they ask an entirely different question, they will get a different answer.
posted by MisantropicPainforest at 9:53 AM on May 28, 2015


If you are going to make inferences about declines in political violence by looking only at interstate wars or only at intrastate wars, you will be making bad inferences.

True enough. But A) the makers of this video don't do that and B) you seemed to be saying that there was never any reason to study the two totals separately; which seemed an odd claim.

Perhaps we're just talking past each other, though?
posted by yoink at 9:54 AM on May 28, 2015


Haha yes I'm not sure who I was talking to.
posted by MisantropicPainforest at 9:59 AM on May 28, 2015


but it ends with a conclusion which is junk. if they had studied the total war deaths in Indochina after WWII, to the total population in Indochina, you would get a totally different story.

Yes: and if you look at war deaths per family-members-of-war-casualties you get an even higher percentage. I don't say that to be simply dismissive, either. This is part of the problem with approaching an issue statistically that, in human terms, is always brutally personal. It is no comfort to any of the people grieving lost fathers, brothers, sisters, mothers, daughters and sons in Afghanistan or the Congo or wherever that those deaths represent a tiny fraction of what we might have reasonably expected had we been looking forward from, say, 1905. There is a sense in which every war death is a tragedy and that any number is "too high."

But there is, surely, also another sense in which the 8 million dead Russian soldiers from WWII is a much more significant tragedy than the 2000 or so dead US soldiers in Afhanistan. There is a sense in which having, say, 10% of the world population dying in wars is worse than having 1% dying in wars. To acknowledge that truth is not to say that the deaths of that 1% are unimportant or should not have been avoided.

I find the response to these kinds of statistics on war deaths is curiously similar to people's reactions to statistics about crime rates--only different political factions tend to have dissimilar responses to them. On the left we tend to be very ready to be persuaded that crime rates have fallen substantially and that policies designed to "fight rising crime" are badly misguided. We tend, though, to be very dismissive of studies demonstrating that popular narratives about the world descending into military chaos are similarly unfounded. Like conservatives faced with statistics about the dropping crime rate we tend to zoom in on the anecdotal counterexample ("but what about this horrible crime in the headlines today" or "but what about this nasty battle going on right now") as a way of ignoring the larger statistical picture.
posted by yoink at 10:05 AM on May 28, 2015 [1 favorite]


if they had studied the total war deaths in Indochina after WWII, to the total population in Indochina, you would get a totally different story.

And if they ask an entirely different question, they will get a different answer.


Well, it's equivalent to saying that the US, over the last 30 years, has become increasingly wealthy, when one part of the population has grown very wealthy while another part of the population has grown poorer. It's true, in total, but of, at best, ambiguous meaning.

Of course words mean exactly what we say they mean, no more and no less, but talking about a "great peace" is an egregiously bad take on the latter half of 20th century history when looking at the scale, duration, and violence of conflicts in the localities where they occured.
posted by ennui.bz at 10:07 AM on May 28, 2015


So what should be said? the overall trend in political violence is decreasing, but where it does exist it is still bad?
posted by MisantropicPainforest at 10:10 AM on May 28, 2015


Yes: and if you look at war deaths per family-members-of-war-casualties you get an even higher percentage. I don't say that to be simply dismissive, either. This is part of the problem with approaching an issue statistically that, in human terms, is always brutally personal.

No, the problem is that it's a bad way to drawn conclusions from statistics. If the distribution has strongly localized variation, talking about things in terms of total averages is almost always going to be misleading, just from the perspective of data science. I mean, yes, you can draw very specific valid conclusions but it's hard, in a way that is guaranteed to be misleading when trying to talk about world history from the data.

It's an abuse of data science.
posted by ennui.bz at 10:10 AM on May 28, 2015


Of course words mean exactly what we say they mean, no more and no less, but talking about a "great peace" is an egregiously bad take on the latter half of 20th century history when looking at the scale, duration, and violence of conflicts in the localities where they occured.

But that would be true if the entire world were at peace except for, say, the island of Tonga, which had devolved into civil war. Are you really saying that it would be morally absurd to say that the world was better off in that case (all other things being equal) than during WWII?
posted by yoink at 10:11 AM on May 28, 2015


Do you have any evidence that the variance of the distribution of violence has changed in tandem with the mean? If you do, great! If not, then using the mean to describe how a distribution has changed is in no way misleading. In fact, its the most efficient way to transmit the change.
posted by MisantropicPainforest at 10:16 AM on May 28, 2015


But that would be true if the entire world were at peace except for, say, the island of Tonga, which had devolved into civil war. Are you really saying that it would be morally absurd to say that the world was better off in that case (all other things being equal) than during WWII?

I don't see how you could construe that from what I said. I don't know that there is much more, on a simple level, to say than there hasn't been a major war between the major industrial powers in a while.

There is a second problem which is that the population increase after WWII is historically anomalous i.e. the "Green Revolution." The growth in total world population after WWII due to advances in farming and medicine is pretty much unprecedented over the time scales they looked at, which adds further to the data abuse.
posted by ennui.bz at 10:18 AM on May 28, 2015


I don't see how you could construe that from what I said.

Because if the entire world was at peace except for Tonga, which was engaged in a drawn out and brutal civil war, it would remain true that while people might talk about a "Great Peace" that would have no bearing on "the scale, duration, and violence of [the] conflict[] in the localit[y] where [it] occured."

The growth in total world population after WWII due to advances in farming and medicine is pretty much unprecedented over the time scales they looked at, which adds further to the data abuse.

That's absurd. The reason for the high population numbers has absolutely nothing to do with questions about what percentage of that population are caught up in military conflict. Or, at least, if there were some connection you would need to demonstrate it separately to make that claim have any weight.
posted by yoink at 10:34 AM on May 28, 2015


What wouldn't be data abuse in this case?
posted by MisantropicPainforest at 10:38 AM on May 28, 2015


> We haven't slaughtered each other en masse because of the constant threat, not just that millions might die horrible deaths in violent conflict, but that a renewed conflict would put an end to all human life on earth.

You don't know that. Nobody knows that. It may be entirely random. Before Hiroshima, I'm sure most people would have said a nuclear weapon would never be used because it was just too awful. To quote William Goldman: Nobody knows anything.
posted by languagehat at 11:33 AM on May 28, 2015 [1 favorite]


It's very telling to me that the documentary-makers don't have the will to or are afraid to estimate the civilian cost of wars post 1945, just mentioning that that also went down.
posted by LucretiusJones at 12:09 PM on May 28, 2015


You don't know that. Nobody knows that. It may be entirely random. Before Hiroshima, I'm sure most people would have said a nuclear weapon would never be used because it was just too awful. To quote William Goldman: Nobody knows anything.

There are degrees of non-knowledge. Do you mean to say that the deterrent effect of nuclear weapons is complete speculation?
posted by dis_integration at 1:17 PM on May 28, 2015


I realize I'm getting hung up on a side-note here, but that figure he presents for the An Lushan Rebellion is just complete nonsense. My faith in a data visualization takes a hit when I notice a howler like that.
posted by Kattullus at 4:44 PM on May 28, 2015


Democracies don't fight each other

Depends on what you call a "democracy", I guess. Russia invaded Georgia and has been involved in the war in Ukraine, and all 3 of those countries are democracies at least in form. But I could understand quibbling with calling Russia a democracy.
posted by thefoxgod at 4:54 PM on May 28, 2015 [1 favorite]


> Do you mean to say that the deterrent effect of nuclear weapons is complete speculation?

Yes, I do, and if you think otherwise you're fooling yourself.
posted by languagehat at 5:49 PM on May 28, 2015




Yes, I do, and if you think otherwise you're fooling yourself.

Do you have any evidence? Because this is a well trod research topic and the general consensus is that nuclear weapons have a very strong deterrent effect. If they don't, please tell me why I should throw out all these books, articles, journals, disssertations, etc.
posted by MisantropicPainforest at 4:48 AM on May 29, 2015 [1 favorite]


> Because this is a well trod research topic and the general consensus is that nuclear weapons have a very strong deterrent effect.

So what? There's no way to test. "General consensus" means nothing. Note that I'm not claiming the deterrent effect of nuclear weapons is wrong, I'm saying it's speculation, which it clearly is.
posted by languagehat at 12:21 PM on May 29, 2015


The general consensus exists because there is a lot of evidence that nukes deter threats. Do you have evidence that they don't? It's absolutely not speculation.
posted by MisantropicPainforest at 2:36 PM on May 29, 2015 [1 favorite]


At best it's a theory, at minimum, conjecture which falls under the criteria of speculation. Semantics are important. Semantics helped create the bomb, it also fueled the Cold War. Wittgenstein thought,
"The hysterical fear over the atom bomb being experienced or at any rate expressed, by the public almost suggests that at last something really salutary has been invented. The fright at least gives the impression of a really effective bitter medicine. I can't help thinking: If this didn't have something good about it the philistines wouldn't be making an outcry. But perhaps this too is a childish idea. Because really all I can mean is that the bomb offers a prospect of the end, the destruction, of an evil, - our disgusting soapy water science..."

People think and say all sorts of things. Historians seeks patterns, these do not equate to fact. There are Foreign policy shifts in 1945 that are evident that the bombs centrality was its cause. I find it odd that Ike and and Wittgenstein agreed on one thing, "everyone feels insecure again" (Ike, Moscow, August, 1945)
Now speculative commodities are another matter but most historians would agree that 1945 is a "watershed" year for the planet.
posted by clavdivs at 11:47 PM on May 29, 2015 [1 favorite]


It's a theory supported by evidence. Formal models,case studies large N analyses. These things aren't made up out of whole cloth. Nukes exert a clear and strong deterrent effect. And you quote Eisenhower whose whole new look strategy was based on the massive deterrent effects of nukes. And it worked!
posted by MisantropicPainforest at 10:28 AM on May 30, 2015


I could counter argue the bomb fueled the Cold War causing covert warfare and proxy fights that hindered diplomatic solutions and commerical trade but it's not what I believe.
So the bomb would not be a good theory that supports diminished conventional warfare. It's lunacy to leave our fate to policy were lunacy is part of the policy itself.
posted by clavdivs at 1:08 PM on May 30, 2015


You could argue it, but do you have evidence to support the argument?
posted by MisantropicPainforest at 1:46 PM on May 30, 2015


This is were Dr. Strangelove comes in handy.

Sure, I could argue the bomb hindered postwar relations when evidence supported that a conventional threat (pre August, 1945) was viewed by the Americans as minimal.
The only thing that scared Stalin more then a re-armed Germany was the atomic dominance we held from 1945-1949. The years 47'-48' being prevelant to supporting data.
posted by clavdivs at 1:57 PM on May 30, 2015


« Older Questions about open source software, women, and...   |   Funny or Racist? Newer »


This thread has been archived and is closed to new comments