A more peaceful world
May 17, 2017 1:54 PM   Subscribe

Since it was recommended by Bill Gates as the "most inspiring" book he has ever read, Stephen Pinker's thesis on the decline of violence, The Better Angels of Our Nature (prev), has been a number one best seller again. Pasquale Cirillo and Nassim Nicholas Taleb (that one) had previously critiqued the thesis in an paper [technical pdf, though well explained], concluding that "humanity is as violent as usual." Economist Michael Spagat joined Pinker in writing a response, leading to a reply by Cirillo and Taleb, provoking yet another letter. While the argument isn't violent, there are some nasty feelings on both sides. Very recently, and in an easy-to-follow set of slides, Spagat has laid out a final (for now) and fairly convincing argument that we really are seeing a decline of violence in the world. For a visual tour of declining violence, there is this terrific set of infographics.
posted by blahblahblah (12 comments total) 27 users marked this as a favorite
 
The "final and fairly convincing argument" seems to have been rather an argument that Cirillo and Taleb's argument isn't really that good, which is an argument in favor of rejecting their criticism, not an argument in favor of the claim that we really are seeing a decline in violence.
posted by kenko at 2:16 PM on May 17 [3 favorites]


Edward S. Herman accused Pinker's book of amounting to imperialist apologetics; here's the incredibly long exegesis of that which I found a while back, but haven't gotten around to reading. Both he and Chomsky have voiced disagreement with Better Angels.

It's definitely a curious debate, but even since Pinker himself suggested that his disagreement with Chomsky is in part due to political and philosophical differences, that's what I tend to attribute these ongoing debates to as well. The sides can argue statistics and interpretation of evidence as much as they like, and it's a fascinating exercise/rabbit hole, but if there are deep differences in worldview, then that's something that could be addressed to help close the gap.
posted by polymodus at 2:24 PM on May 17 [2 favorites]


I haven't read through all of the links, but it also seems that all of the arguments, as Spagat himself points out, are about Pinker's claims about war. When I read BAoON, the parts where he uses evolutionary psychology to explain rape seemed way more conjectural and less convincing.
posted by tofu_crouton at 2:25 PM on May 17 [1 favorite]


The focus in this debate is indeed about the decline in violence, rather than Pinker's (speculative) reasons for the decline. Pinker can be wrong about the reasons, but right about the trendline.

And the fact that there is a decline in current violence seems pretty clear from almost all studies: war, murder, battle deaths, etc. The question is whether it represents a true change or just random variation. Cirillo and Taleb's argument is statistical - that we are not really seeing a trend at all, but rather some standard distribution of rare events. If there is a trend, which is what Spagat shows, then the reasons might be different than Pinker suggests, but the phenomenon is important in any case.
posted by blahblahblah at 2:31 PM on May 17 [7 favorites]


Pinker's good news hasn't gotten around as much as it should because it's bad news for politicians of both the left and right, whose stock in trade is convincing everyone that times are bad and only they can save us. The decline of violence supports into nobody's crisis-driven ideology, it doesn't suggest any policy, and nobody's exactly sure why it's happened, so it's pretty much an orphan concept in the political marketplace. And for that reason, I love it.
posted by Modest House at 4:11 PM on May 17 [7 favorites]


Specifically, we can estimate the probability of heads based on the first 500 flips (which will be 1⁄2) and then test whether the last 50 flips seem to be governed by this fairness parameter.
Sure, if you have explicit external reasons to believe the last fifty flips are unique. Otherwise you're just fishing for significance with no constraints, and you'd better take that into account when claiming to have observed a difference. Testing against a model that assumes no break and claiming consistency seems like a perfectly reasonable approach a badly posed question. It's far less questionable than arbitrarily deciding the year 1945 is special and that historical mechanisms are continuous for thousands of years and suddenly discontinuous there, as Spagat does.

I don't know how many free parameters you run through before deciding "WWII" is the one unique turning point in history at which to place your break, but I'd bet it isn't terribly far removed from the number of wars that have been recorded.
posted by eotvos at 4:39 PM on May 17 [1 favorite]


Pinker's "good news" has nothing to do with crises, since a decline in violence is completely compatible with an uptick in all sorts of nasty things, and IIRC, on Pinker's account mass incarceration gets a bye (and of course mass immiseration and impoverishment isn't violence either). I wonder if we even have accurate numbers of beatings or deaths incurred in prisons for him to ignore in the first place.

(You'd think the big obvious explanation for why WWII-scale hasn't occurred, if indeed it hasn't, in the course of a single war, is that following WWII and the rather dramatic demonstration of the destructive power of the new weapons developed in its course in the bombings of Hiroshima and Nagasaki world power was more or less divided between two global hegemons each of which had the power to utterly destroy the other, and which therefore confined themselves to more minor proxy wars and avoided testing the limits of mutually assured destruction. It's a potted story but it's the one I'd look to falsify if I had a big theory about the decline of violence-construed-predominantly-as-killings and also believed that its decline on a global scale was mysterious.)
posted by kenko at 4:43 PM on May 17 [5 favorites]


I'm not hopeful. People haven't been around that long and people get the weirdest ideas about what makes sense with regard to violence. We didn't get a good solid demonstration of what industrialized killing was until WW1 despite having machine guns campaigning around the globe and the hinting of the American Civil War. To me the way we use space weapons against against 'freedom fighters' suggests a ghostly parallel to the Maxim gun in Colonial exploits leading up to the great carnage and I dread a 'real' war between industrially competent parties with modern firepower. I think a few years of that would erase all the peace gains we are seeing right now. And will it be seen ahead of time? No, at least not by most people. Sure it is nice to bask in a present with out ancient round the clock warfare and think ourselves civilized and modern but remember our capacity for killing is now pretty unfathomable and our DNA hasn't changed all that much.
posted by Pembquist at 5:25 PM on May 17 [4 favorites]


Again, I am not defending Pinker's reasoning for why the world might be changing (I am an economic sociologist who has problems with evolutionary psychology), but I really think there some selling short of good news in the actual data. Specifically, in response to some of the objections:

Pinker's "good news" has nothing to do with crises, since a decline in violence is completely compatible with an uptick in all sorts of nasty things, and IIRC, on Pinker's account mass incarceration gets a bye (and of course mass immiseration and impoverishment isn't violence either). I wonder if we even have accurate numbers of beatings or deaths incurred in prisons for him to ignore in the first place.

But mass impoverishment is down (by insane amounts) - almost 75%(!!) in 30 years. A billion people have been lifted out of poverty since 1990, in what might be humanity's most amazing achievement.

Mass incarceration is a big problem in the US, but it isn't generally thought to be a primary driver in the drop in violence. It also isn't an issue in Europe or Asia, which have low crime rates that continue to decline without mass incarceration, and Africa and Latin America are also stabilizing.

So I am not sure what the nasty things that are responsible for the drop in violence are, they may be there, but they aren't immediately obvious.

It's far less questionable than arbitrarily deciding the year 1945 is special and that historical mechanisms are continuous for thousands of years and suddenly discontinuous there, as Spagat does.

I broadly agree that the argument over whether wars are follow a Poisson distribution or have experienced a discontinuity is still an open debate. I do think this sells the argument of Pinker a bit short. The argument is that, if wars alone showed a rapid change after 1945, we would have reason for suspicion. However, broad global changes unknown on historical scales have occurred in the same period (see my answers on poverty above, and general murder rates), but also increases in literacy, democracy, and even light. These changes might represent a fundamental shift in how the world operates that began during the post-war era. It is not an open-and-shut case, but it is not unreasonable.
posted by blahblahblah at 5:27 PM on May 17 [13 favorites]


Blahblahblah, to be clear, the issue with Pinker et al discounting mass incarceration isn't that it's believed to be a cause of the crime rate. It's that in and of itself it is a massive act of violence and the violence that occurs within prisons goes pretty much unchecked and untracked. If you say rapes are on the downturn but you don't count prison rapes as rapes, you're not counting the real numbers. (That's why I'm so heartened by you pointing out that mass incarceration is not a global phenomenon.)
posted by peppercorn at 5:33 PM on May 17 [9 favorites]


I realize y'all are arguing this on a philosophical level that I can't really speak to at all, but I thought the rise and decline in violent crime around the world (between ~1960 and now) was directly correlated to the rise and decline in use of that sweet, sweet leaded ethyl?
posted by carsonb at 7:43 PM on May 17 [1 favorite]


(Reading links now...) I see this goes back a bit further than the age of the automobile.
posted by carsonb at 7:49 PM on May 17


« Older When Your Child Is a Psychopath   |   Notes from an Emergency Newer »


You are not currently logged in. Log in or create a new account to post comments.