Pinker, Foucault and Progress

Panopticon (1)

As readers may know, a little while back I wrote a piece on Steven Pinker’s Better Angels of Our Nature a book that tries to make the case that violence has been on a steady decline throughout the modern era. Regardless of tragedies such as the horrendous school shooting at Newtown, Pinker wants to us know that things are not as bad as they might seem, that in the aggregate we live in the least violent society to have ever existed in human history, and that we should be thankful for that.

Pinker’s book is copiously researched and argued, but it leaves one with a host of questions. It is not merely that tragic incidents of violence that we see all around us seem to fly in the face of his argument, it is that his viewpoint, at least for me, always seems to be missing something, to have skipped over some important element that would challenge its premise or undermine its argument, a criticism that Pinker has by some sleight of hand skillfully managed to keep hidden from us.

I think an example of this can be seen in Pinker’s treatment of the decline of torture and fall in the rates of violent crime. Both of these developments, at least in Western countries, are undeniable. The question is how are these declines to be explained.  What puzzled me is that Pinker nowhere even mentions the work of the late philosopher, Michel Foucault,  a man who whatever the flaws and oversimplifications of his arguments, thought long  and hard about the questions of both torture and crime. In fact, Foucault is the scholar whose work is most associated with these questions.  It is a very strange oversight, for Pinker does not bring up Foucault even briefly to dismiss his views.

It seems, I am not the only person who asked this question for on his website addressing frequently asked questions Pinker gives the following explanation for ignoring Foucault:

Questioner: You obviously must discuss Michel Foucault’s Discipline and Punish, the book that explains the decline of judicial torture in Europe.

Pinker: Actually, I don’t. Despite being a guru in the modern humanities, Foucault is not the only scholar to have noticed that European states eliminated gruesome punishments, and his theory in particular strikes me as eccentric, tendentious, and poorly argued. See J. G. Merquior “Charting carceral society” in his book Foucault (UC Press, 1985), for a lucid deconstruction.

I wanted to see what this “lucid deconstruction” of Foucault by  Merquior (Pinker is nothing if not clever- Foucault is a patron saint to literary deconstructionist), so I checked it out.

Here is how Merquior introduces Foucault’s Discipline and Punish:

 Foucault once called it ‘my first book’ and not without reason: for it is a serious contender for first place among his books as far as language and structure, style of organization and ordering of parts go. It is not a bit less engrossing than Madness and Civilization, nor less original than the order of things.  Once again Foucault unearths the most unexpected of primary sources; once again his reinterpretation of the historical record is as bold as it is thought provoking.” ( Foucault p. 86)

This is the guy Pinker asks us to turn to for a rebuttal of Foucault?

Merquior does have some very valid arguments to make against Foucault, more on that towards the end, but first the views that Pinker does not discuss- Foucault’s view of the rise of the prison.

The theory that Foucault lays out in his Discipline and Punish which provides a philosophical history of the modern prison is essentially this: The prison emerged in the late 18th and 19th centuries not as a humanitarian project of Enlightenment philosophes, but as a disciplinary apparatus of society in conjunction with other disciplinary institutions- the insane asylum, the workhouse, the factory, the reformatory, the school, and branches of knowledge- psychology, criminology, that had as their end what might be called the domestication of human beings. It might be hard for us to believe but the prison is a very modern institution- not much older than the 19th century. The idea that you should detain people convicted of a crime for long periods perhaps with the hope of “rehabilitating” them just hadn’t crossed anyone’s mind before then. Instead, punishment was almost immediate, whether execution, physical punishment or fines. With the birth of the prison,  gone was the emotive wildness of the prior era- the criminal wracked by sin and tortured for his transgression against his divine creator and human sovereign. In its place rose up the patient, “humane” transformation of the “abnormal”, “deviant” individual into a law and norm abiding member of society.

For Foucault, the culmination of all this, in a philosophical sense, is the Panopticon prison designed by Jeremy Bentham (pictured above). It is a structure that would give prison officials a 24/7 omniscient gaze into the activities of the individual prisoner and at the same time leaves the prisoner completely isolated and atomized. In the panopticon Foucault sees the metaphor for our own homogenizing conformist and totalizing society.

What Foucault succeeded in doing in Discipline and Punish was putting the horrific judicial torture of the pre-Enlightenment era and post-Enlightenment policy of mass imprisonment side-by-side. In doing this he goads us to ask whether the system we have to today is indeed as humane, as enlightened, compared to what came before as we are prone to believe.

This is exactly what Pinker responding to a question on imprisonment does not allow us to do:

Questioner:What about the American imprisonment craze?

Pinker: As unjust as many current American imprisonment practices are, they cannot be compared to the lethal sadism of criminal punishment in earlier centuries

Okay, true enough, but for me, this answer misses the point of the question. The underlying assumption behind the question seems to be “yes, violence might have declined, but isn’t locking up millions of people - six million to be exact – a number larger than those of Stalin’s gulag archipelago, 60 % of whom are there for nonviolent offenses, a form of violence?” Or perhaps “might the decline in violence be the result of mass imprisonment?” Admitting either would force Pinker to accept that the moral progress he details is perhaps not as unequivocal as he wants us to believe.

Here, I think, is where Pinker’s attachment to the Enlightenment idea of progress leads clearly to complacency. Pinker loves graphs, so here’s a graph:

U.S._incarceration_rates_1925_onwards

Source: http://en.wikipedia.org/wiki/United_States_incarceration_rate

It seems frankly obtuse to not connect the decline in crime with the sheer number of people now being locked up. It is tragic, but the connection between rising rates of imprisonment and declining crime rates can be seen even in Pinker’s vaunted Western Europe where the rate of imprisonment rose- though to nothing like the obscene rate it rose in the United States- and the crime rate fell in tandem.

Yet, unless the scale of imprisonment is put in context we are likely to see imprisonment of nonviolent offenders as less than morally problematic, and merely as an unfortunate consequence of the need to protect ourselves from violent crime by throwing the net of criminal justice as wide as it can be thrown, something Pinker seems to do when he states:

A regime that trawls for drug users or other petty delinquents will net a certain number of violent people as a by-catch, further thinning the ranks of the violent people who remain on the streets. (BA 122)

The strange thing here is that the uniquely American practice of locking up every law breaker without distinguishing between the risks posed by the accused is not only clearly disproportional and unjust it has makes no apparent effect on the actual rate of violent crime. The US incarcerates a whopping 743 persons per 100,00 whereas Great Britain lock up 154 per the same amount  and the US still has an intentional homicide rate 4 times higher than in the UK.

By seeing modern history almost exclusively through the lens of moral progress, Pinker blinds himself to the question of whether or not our own age is engaged in practices that a more progressive future will regard with horror.

The question of imprisonment and its relationship to the decline of crime is not the only place where Pinker in his Better Angels of Our Nature dismisses a messy, often harsh, reality in the name of a simplified Enlightenment notion of progress. This can be seen in Pinker’s notions regarding contemporary slavery and war.

In a strange way, Pinker’s insistence that we recognize the reality of moral and social progress might short-circuit our capacity for progress in the future. You can see this in his treatment of “human trafficking” a modern day euphemism for slavery. As always, Pinker wants to let us know that current figures are exaggerated, as always, he reminds us that what we have here is no comparison to the far crueler reality of slavery found in the past. But this viewpoint comes at the cost of continuity. Anti-slavery advocates such as those of the organization Free the Slaves assume a moral continuity between themselves and the earlier abolition movements- and well they should. But Pinker’s rhetoric is less “we have almost reached the summit” than one of undermining the moral worth of their struggle with his damned proportionality- that things are better than ever now because “proportional to world population” not as many people are murdered, die in war, or are enslaved.

Numbers off or not- anywhere even in the ballpark of 25 million slaves today- the high estimate- still constitute an enormous amount of human suffering- such as innumerable rapes, beatings, and forced labor (no doubt Pinker would try to put a number on them)- suffering Pinker does not explore.

What holds for slavery in Better Angels holds for war as well. He is at pains to point out the casualty figures of the most savage conflict of the last generation- a conflict most westerners have probably not even heard about- The Great War of Africa- are grossly exaggerated, that the war only killed 1.5 million- not the 5 million human beings often reported.

Pinker’s right about one thing- wars between the world’s most powerful states have, at least for the moment disappeared.Wars between the great powers have always been the greatest killers in history, and we haven’t had any of those since 1945, and the question is- why? Pinker will not allow the obvious answer to his question, namely, that the post 1945 era is the age of nuclear weapons, that for the first time in history, war between great powers meant inevitable suicide. His evidence against the “nuclear peace” is that more nations have abandoned nuclear weapons programs than have developed such weapons. The fact is perhaps surprising but nonetheless accurate. It becomes a little less surprising, and a little less encouraging in Pinker’s sense, when you actually look at the list of countries who have abandoned them and why. Three of them: Belarus, Kazakhstan and the Ukraine are former Soviet republics and were under enormous Russian and US pressure- not to mention financial incentives- to give up their weapons after the fall of the Soviet Union. Two of them- South Africa and Libya- were attempting to escape the condition of being international pariahs. Another two- Iraq and Syria had their nuclear programs derailed by foreign powers. Three of them: Argentina, Brazil, and Algeria faced no external existential threat that would justify the expense and isolation that would come as a consequence of  their development of nuclear weapons and five others: Egypt, Japan, South Korea, Taiwan and Germany were woven tightly into the US security umbrella.

Countries that face a perceived existential threat from a nuclear power or conventionally advanced power (and Argentina never faced an existential threat from Great Britain, that is Britain never threatened to conquer the county during the Falklands War) would appear to have a pretty large incentive to develop nuclear weapons insofar as they do not possess strong security guarantees from one of the great powers.

Pinker believes that Kant’s democratic peace theory (that democracies tied together by links of trade and international organization do not fight one another) helps explain the decline of war, but that does not explain why the US and Soviet Union did not go to war or India and Pakistan, or Taiwan and China, or South and North Korea. He pins his hopes on the normative change against nuclear weapons found in a Global Zero a movement that includes an eclectic  group of foreign policy figures including realists such as Henry Kissinger that hopes to rid the world of nuclear weapons.

While I find the goal of a nuclear weapons free world laudable, the problem I see in this is that weaker powers lacking advanced conventional weapons could very well understand this movement as a way for the big powers to preserve the rationality of war. In fact, the worse thing imaginable would be for great power war to regain its plausibility. If the recent success of Israel’s  “Iron Dome” is any indication we may end up there even without the world abandoning its nuclear weapons. Great powers, such as the US and China, may be more likely to engage in brinkmanship if they start to think they could survive a nuclear exchange. Recent confrontations between China and its neighbors and East Asia’s quite disturbing military buildup do not portend well for 21st century pacifism.

Global Zero might with tragic irony prove more dangerous that the current quite messy regime if it is not followed in parallel with an effort to solve the world’s outstanding disputes and to build a post- US- as- sole- superpower security architecture-not to mention efforts to limit conventional weapons which while we were sleeping have become just as deadly as nuclear weapons as well. Where everyone feels safe there is no need for everyone to be armed to the teeth.  

Pinker recoils from messy explanations or morally ambiguous reality because he is wedded to the idea that the decline in violence was driven by a change in norms- a change that he thinks began with the Enlightenment. In his eyes, we are indeed morally superior to our predecessors in that we have a more inclusive and humane moral sense. Pinker turns to the ethical philosopher- Peter Singer- and his idea of the “escalator of reason” for a philosophical explanation of this normative change. Singer thinks that overtime human generations reason their way to inclusiveness and humanity by expanding our “circle of empathy”. Once only one’s close kin sat in the circle of concern, then fellow members of one’s state or faith, now perhaps all of humanity or, as Singer himself is most famous for in his Animal Liberation, the circle can be extended to non-human species.


Singer, however, is an odd duck to peg yourself to as a kind of philosophical backdrop for modern moral progress. A reader of Better Angels who did not know about Singer would be left unaware of just how controversial Singer’s views are. If memory serves me correctly, this fact that his views are something less than mainstream is tucked away in a footnote at the back of Pinker’s book.

Here is Singer from his  Writings on an Ethical Life:

When the death of a disabled infant will lead to the birth of another infant with better prospects of a happy life, the total amount of happiness will be greater if the disabled infant is killed. (189)

I should be clear here that Singer is not talking about abortion, but infanticide, indeed he sees both practices as acceptable and morally equivalent:

That a fetus is known to be disabled is a widely accepted grounds for abortion. Yet in discussing abortion, we saw that birth does not mark a morally significant dividing line.  I can not see how one could defend the view that fetuses may be “replaced” before birth, but newborn infants may not be” (191)

If this is the escalator of reason I want to get off.

Much as with the case with Foucault, Pinker doesn’t spend even a page or two engaging with these ideas. With 802 pages to its name a few more pages would seem a small price to pay, but again they are ignored, perhaps largely because they detract from Pinker’s Enlightenment notions of moral progress. Even briefly grappling with these ideas, for me at least, seems to lead to all sorts of interesting and often quite disturbing possibilities that are outside the simplistic dichotomies of progress and anti-progress set up but Pinker and Foucault.

Our society has certainly made progress morally over past ages in its abolition of torture and slavery, in it’s extension of rights to the formerly oppressed , its inclusion of women in political, economic and intellectual life, its freedom of speech and thought,
not to mention the vast increases in our standards of living,  and yet…

May be our society has not so much progressed morally in the sense of empathy as it has become squeamish about violence, and physical coercion (real violence that is- media and video games seem to reveal an obsessive bloodlust).  What we have done is managed to effectively conceal violence, and wherever possible to have adopted social and psychological methods of manipulation and control- including surveillance- in place of, to use military speak, “kinetic” methods. Our factory farms kill and confine more animals than have ever suffered such a fate- only we never see it. (Perhaps that is part of the explanation for why our urbanizing world has become so squeamish about violence, the fact that so few of us are engaged in the violence against animals found in agricultural life). We do not physically torture but confine and conceal far more persons than were ever caught in the cruel but paltry nets of pre-modern states. Chattel slavery and its savagery are a thing of the past, but what we have now are millions of invisible slaves, kidnapped, locked in houses, people who are our very neighbors suffering the cruel tyranny of one human being over another.


Our wars are fought in regions deemed too dangerous to be covered by mainstream media and our images of them sanitized for prime-time viewing.  Our bloated and growing militaries represent bottled up potential energy that could level whole civilizations, indeed destroy the human species and the earth, should circumstances ever sweep us up and call it to burst forth. Yet, even our soldiers are averse to killing, so we are building machines capable of murdering more effectively and without conscience to replace them.

We do not expose our newborns on the rocks because they are girls or are disabled, but select against them in the womb so that 100 million girls have “gone missing” and whole categories of human beings are disappearing from the world, and some, such as the geneticist Julian Savulescu argue it is our “moral duty” to perform this “redesign” of the human species.

This returns me to the critic Merquior. Merquior makes the valid critique of Foucault that he is a sloppy historian, that he wants history to neatly fit his theory, which history can never do. Above all Merquior sees the flaws in Foucault’s argument stemming from his a prior position that the Enlightenment was less a humanitarian than a proto-totalitarian movement. This makes it impossible for Foucault to see the movement against torture and the creation of the modern prison system as anything more that an expression of a Nietzschean “will to power”.

But Merquior asks:

Why should the historian choose between the angelic image of a demo-liberal bourgeois order, unstained by class domination, and the hellish picture of ubiquitous coercion? Is not the actual historical record a mixed one, showing real libertarian and equalizing trends besides several configurations of class power and coercive cultural traits?  (98)  

Pinker might have done better had he employed Merquior’s critique of Foucault to himself, for, by seeing in modern developments the hand of progress from savagery to civilization, Pinker ends up blinding himself to the more complex historical picture as much as Foucault who saw in modern trends little but the move towards social totalitarianism. Indeed, Pinker could save his Better Angels by adding just one chapter as an afterward. The chapter would look at not where we have come from, but where we are and the struggles still left to us. It would provide a human face to the modern day suffering of those in our progressive age who are still enslaved and who continue to be killed and maimed by war. Those murdered and raped and those suffering behind bars for crimes that have harmed no one but themselves and those who love them. It would be a face Pinker had taken from them by turning them into numbers. It would seek to locate and avoid the many cliffs that might just plunge us downward, and say to all of us “we have just a little ways to go, but for the sake of our own enlightened legacy, we must have the forward thinking and endurance to climb onward, and above all, not to fall.”

* An earlier version of this post was published on December 30, 2011
About these ads

17 comments on “Pinker, Foucault and Progress

  1. Charles says:

    Fascinating reading on a New Years Eve! :)

  2. I agree that Pinker doesn’t really deal with Foucault’s analysis, and that the primary thrust of modernity is, as you say, “to effectively conceal violence.” We currently have a nice big metaphor for that concealment in Guantanamo Bay, where the US freely and openly suspends due process rights, detaining them indefinitely, often in torture conditions. In another example, one of the US hikers detained in Iran came back, toured a US “supermax” facility, and said that the US prisoners were subjected to much worse solitary confinement than he was as a suspected spy.

    That said, I don’t think the ad hominem attack on Peter Singer is necessary. I respect Singer precisely because his arguments make people — including myself — uncomfortable, but they’re not easily dismissed except through ipse dixit assertions of morality.

    • Rick Searle says:

      Thanks much for your comments, Max.

      I totally agree with you on the Guantanamo Bay and solitary confinement issues. I am especially concerned about the former given the current assault on the 5th Amendment.

      http://www.theatlantic.com/politics/archive/2012/12/scandal-alert-congress-is-quietly-abandoning-the-5th-amendment/266498/

      I hope I didn’t engage in an ad hominem attack on P. Singer. If I assume (ipse dixit) that most people find what he says regarding ethics (and I am especially thinking of his justifications for the infanticide of the disabled here) are morally abhorrent: is this much different from the fact that Pinker assumes (rightly in my view) that we will naturally find abhorrent past practices such as torture or slavery, domestic violence or war, and therefore their decline something to be thankful for?

      The end of willful infanticide and decline of discrimination against the disabled should be considered part of the moral progress Pinker so lauds. For Pinker to connect his theory with a thinker who justifies such practices without any mention of the fact, and especially without disavowing such statements, leaves me baffled.

  3. It seems that at least part of the decline in unacceptable violence (like crime) has been replaced by an uptick of acceptable violence (like some cases of imprisonment, but there could be others). And what about violence in a place such as prison? And the intensity of violence as opposed to its amount (did he address this, or is it even something to address?) On net, there may still be a decline in violence (after all, Pinker’s book deals with much more than prisons, and it is a substantial book), but as you said he may have not subjected his findings to all of the objections, so the decline might be exaggerated. To what extent?

    • Rick Searle says:

      Hey Hank, hope you had a nice holiday.

      All good points. Pinker, from what I can recall, doesn’t really address the question of intensity in the way you phrase it. To the extent he does, his views in terms of state violence or the potential for massive state violence which he sees primarily as a question of nuclear weapons, is that states have given or are giving
      up such WMDs up and that global norms are moving against such weapons. In so far as an INDIVIDUAL having the potential for intense mass violence- something again that he sees primarily as an issue of WMDs- he thinks the issue is largely hyped.

      The strange thing is I largely agree with Pinker that violence in many different forms has declined in the modern era. The questions for me are- why? is this really “the best of all worlds” or is there a darker side to this? Is there not a danger in not judging our own age as harshly as others have judged theirs? And is this decline in violence likely to persist regardless of what we do?

      • Maybe the violence is in some cases more intense (isn’t it more intense for 1 person to kill 23 people than it is for 23 people to kill 23 people?) but committed less and by less people and is thus seen as an outlier, and in other cases is so very much less intense (at least apparently) and so well spread out that it doesn’t even register.

        I think if that second part is correct it helps answer the question, affirmatively, “Is there a darker side to this?” because among those things that can be seen as less intense or less obvious but more widespread violence are indeed things like prison (and in my opinion abortion), as well as possible things that aren’t considered violent at all but are still types of “undesirable” behavior (even someone who is not necessarily pro-life shouldn’t find it hard to put abortion in this category, assuming they don’t see population stagnation/decrease and psychological problems for young women as good things) that may be on the increase to take up the slack of violent behavior.

        I’m not saying that there is a set amount of violence per person in the world from the beginning of time, but I am saying that when it is decreased, it seems evident that — I don’t know what you call it — “human will”, is channeled into something else. Some good (for example, where it once may have been practical to wipe out your neighbors and steal their food and mineral sources, it is now much more practical to trade with them for their finished products), some bad (such as less obvious violence, more “acceptable” violence, or other “undesirable” behavior).

      • Rick Searle says:

        In light of Pinker’s theory the issue of intensity holds for the World Wars, but not after. The idea is even though the potential for one person to murder a large group of people- such as the recent horrendous mass murders at Sandy Hook Elementary- exists and is sometimes tragically acted out- an individual’s chances of being murdered or committing murder are substantially less than in any other era. The homicide rate for pre-agricultural societies- if I remember correctly from Better Angels is somewhere near 15%. (It’s only a fraction of a % today). The reason for this is that in pre-agricultural societies disputes both within and between peoples are primarily settled by force. Most crimes today occur in places such as inner-cities where the kinds of systems for the arbitration and control of force are not in effect. Pinker makes the case, and I think he is right in this, that many of the murders we have to day are actually expressions of this ancient system of justice which continues to exist at the margins of society.

        There are people, such as Robert Wright in his book The Evolution of God, who tend to see something like a “divine plan” in the decline of violence- though he is not thinking of a personal God in the Christian sense. Pinker sympathizes with the temptation to turn to such explanations, but remains an atheist/humanist.

        Pinker himself brings up your point about abortion. I don’t have the book in my possession to cite it exactly, but he reasons that if one considers abortion actual homicide then there hasn’t been a net decline in violence at all. The estimate on the number of “gendercides” which has usually meant abortions performed to select against girls is estimated at 100 million which is a larger number of people than were killed in WWII.

        The way he avoids the conclusion that abortion negates his thesis regarding the decline of violence is to argue that the fetus is not a person based on neurological development. He also, however, points out that the number of abortions has long been in precipitous decline, so even considering abortion on an equal plane with other forms of violence still leads to the conclusion that violence is in decline, and I would add that I recently heard a talk by Hanna Rosin that pointed out that the trend to prefer boys over girls at the root of gendercide in non-Western countries is, thankfully, rapidly giving way:

        http://www.ted.com/talks/hanna_rosin_new_data_on_the_rise_of_women.html

        Though, what remains troubling for me, as I mentioned in the post, is that Pinker connects his views to those of Peter Singer who justifies not merely abortion but selective abortion and not merely selective abortion but even selective infanticide. I find this especially dangerous given the advances in genomics which may allow parents to cheaply screen for “good” and “bad” traits.

  4. There are two points I wanted to make, and keimh3regpeh2umeg/Henry/Hank (covering all the bases, just in case!) got there before me for one in its essence. We have managed to relabel certain practices in such a way that they are no longer regarded as the same act (i.e., of coercion or violence) simply because the name has changed. We don’t have imperialism any more, we have globalization, and that is a Good Thing. Most recognize that the situation is more complex than the Washington Consensus view of things which was too all intents and purposes, common sense for the past two decades. I don’t know how that picture stands in the U.S. today (not so great) but in Europe certainly it is well on the way out, never mind for the rest of the world. I append a link to an article in The Atlantic about this.

    http://www.theatlantic.com/international/archive/2012/12/slaverys-global-comeback/266354/

    My second point is that to hold up the non-use of nuclear weapons as a proof of Kant’s view of democratic peace is, again, to focus on a simplified view of things. To go from Pinker to another one idea, big-theory hedgehog, Niall Ferguson maintains in “The War of the World” that there was in fact a third world war, and it was fought conventionally, ruinously, and more or less out of sight. Civil wars tore apart every region of the world after WW2 (and before, Ferguson points out), for ethnic and nationalist ‘reasons’. Not to draw attention to this is a white-wash (in multiple senses perhaps?).

    Thanks for another great post. Have returned to The Signal and the Noise. Working on something related to it now.

    • Rick Searle says:

      Great point about relabeling! I can’t believe it didn’t come to me as well- it’s straight Orwell after all. And your point about Ferguson’s view in his War of the World is spot on as too.

      I am very concerned that what we are living in might just be a lull between storms- although many in Iraq, Afghanistan, Syria large sections of Africa etc wouldn’t believe there even was a lull. After all, it was 99 years between the end of the Napoleonic Wars and WWI. We should probably not be complacent and fool ourselves into thinking we are living Kant’s dream just yet.

      Looking forward to your piece on Signal and the Noise.
      Shoot it my way whenever you’re done.

  5. [...] Still, if Pinker’s Better Angels can be read from this utopian standpoint it be approached from a dystopian viewpoint as well. My subject next time…. [...]

  6. [...] the ancients. At least that is the impression one gets from reading books such as Steven Pinker’s The Better Angels of Our Nature, or Robert Wright’s The Evolution of God. Both Pinker and Wright, each in their own very [...]

  7. […] The worst case scenario that I can think of is that we are witnessing an early stage of the shift away from the non-violent society, we, despite our wars, created over the 19th and 20th centuries. A decline that Steven Pinker so excellently laid out in his Better Angels of Our Nature.   […]

  8. […] of these perspectives have been countered by Steven Pinker in his monumental The Decline of Violence. Pinker’s is the example par excellence for the argument that the Enlightenment wasn’t a failure […]

  9. […] Sadly, I have my doubts regarding Pinker’s argument. Here’s me from an earlier piece: […]

  10. […] this idea of prehistoric and primitive pacifism has come under sustained assault. Steven Pinker’s Better Angels of Our Nature: Why Violence has Declined is the most popularly know of these challenges to the idea of primitive innocence that began with […]

  11. […] the reality of slavery in our supposedly advanced and humane 21st century with there being more people enslaved today than at the height of 19th century chattel slavery. It’s just the proportions that are so much lower. Many of the world’s working poor, especially […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s