The Better Angels of Our Nature: The Decline of Violence In History And Its Causes Kindle Edition

I've struggled a bit to write this review because I have mixed feelings about The Better Angels of our Nature by Steven Pinker.

I started reading The Better Angels of Our Nature: Why Violence Has Declined with the attitude that Pinker needed to first convince me violence had declined before getting into explaining why. To be perfectly honest, given the world we currently live in, it's hard to imagine that violence has declined.

While I finished the book convinced that violence has declined, I felt like the explanations for why seemed more hypothetical than proven. Pinker explored violence quite thoroughly beginning his book at the beginning of human existence and moving to modern times in the almost 700 pages of The Better Angels of Our Nature. He explored historical myths as well as historical documents to arrive at his conclusions. He used archaeological finds to disprove mythical battles. He described how the development of etiquette and the creation of government helped quell violence and change our norms about violence. He used a combination of statistics, anecdotal evidence, and archaeological studies to present his case.

Yet, the more I read, the more my college corrections statistics professor's words haunted me. He always warned our class to be careful when writing papers not to allow our biases and our desires to prove our points to affect the weight we gave the studies we used as evidence.

Pinker seems less objective in some areas of The Better Angels of Our Nature than in other sections. He seemed to excuse violence against some people while unequivocally condemning it against others. This bias felt incredibly out of place in a book on why violence has declined.

For example, when talking about things like the FBI's crime report and other such studies on crime, Pinker never mentions the effect of police discretion and biased court results on crime rates or how the statistics for individual areas are sometimes skewed by reporting or not reporting data. My assumption is he believes the numbers wouldn't be enough to skew the overall results, and a simple paragraph could have addressed that issue. Maybe even just a few sentences; however, if those sentences existed I couldn't find them.

His inconsistent handling of anecdotal evidence and research surveys deemed certain groups of people more credible than others without giving a clear reason why.

As I read The Better Angels of Our Nature, I found myself wanting it to be better than it was yet I still think it's a book worth reading. Pinker obviously studied violence in great depth. He explains the statistics in an easy to understand, straightforward method, and he tells the story of violence quite well. He makes violence the main character, for better or worse, in a story that is ongoing and relevant and important. In fact, Pinker tells the story so well and brings up such important points, facts, and conclusions, that I am tempted to dismiss the things that bothered me about the book. Yet, I can't do that in good conscience. Pinker drives home the fact that violence is much less acceptable than it used to be for a variety of reasons and that unacceptability has come about as humans have developed civilization and sought out ways to live together more peacefully. The Better Angels of Our Nature left me hopeful that we can continue to rise above violence and find nonviolent solutions in spite of my skepticism about certain sections of the book.

This books uses data to impress readers, as smokefront; but full of statistical errors. Aside from his central claim being wrong, there are flaws in the data analysis.
For instance:
1) Pinker accepts that violence follows a power law distribution with an exponent around 1, hence no stable average, but then makes statement about the drop in mean, not realizing the problem. (Under these conditions one cannot draw scientific conclusions from descriptive statistics. The closer the exponent is to 1, the lower the validity of claims of drop in violence.)

2) Pinker uses Richardson (1960) as back-up. The latter states that there is no sufficient evidence to reject the null hypothesis of a homogenous Poisson process, which denies the presence of any trend in the belligerence of humanity (Cirillo and I verified the claim). Nevertheless, Pinker interprets Richardson as saying the opposite, and is not responding to critiques.

3) Other metholodogical flaws such as the estimation of the An Lushan rebellion or the Mongolian invasions without an error rate for past biases.

4) Marketing (there is little, very little data, but a lot of graphs and words giving the illusion of thorough empiricism).

Being a recent college graduate, the moment I saw that Bill Gates had recommended this book, I bought it and began reading. This book has some important, intriguing, and profound themes such as how far we have come collectively as humans from even the middle ages to now. Though we are by no means close to being angels, (as evident by the chapter about our inner demons), progress is undeniable. I also am a fan of Pinker's style of writing. He adds humor intermittently, has some interesting anecdotes, and does a good job at explaining most of his reasoning and data. I found the length of this book to be excessive though. This is a big book, make no mistake about it. At times, when Pinker is backing up his claims with copious amounts of data and graphs in the middle sections of the book, it can be a chore to read. However, I think I am the better for having read this book, and though I personally would not say that this book changed my life, I am glad to have read it.

I would also like to mention that if you buy the paperback version, towards the beginning, the text on the pages is uniformly at an angle, and not entirely horizontal across the page.