LevV wrote:You could buy the book or you could go to the site below to get an excellent detailed summary of the book.
I read about 80% of the book. It was an effort at times, but worth it. He certainly supports his statements with an exhaustive amount of statistical evidence.
newbooksinbrief.com
80% is quite good, I'd say. It'd be hard to say that Pinker isn't exhaustive in his treatment of the topic, but that can lead to exhaustion, too. After I read about 200 pp and had to give the book back to the library, I read Aaron's great summary on newbooksinbrief.
saffron, you say you're not convinced about the decline in violence. I won't use the word 'indisputable' here, but the data look pretty convincing. I think what's important to realize and very significant is that the decline is proportionate to population size, so that in most cases we wouldn't be talking about a decline in absolute numbers. We'd be seeing more rapes, homicides, assaults, etc. currently than in the past, but the rate is lower. Taking a page from Daniel Kahneman (whose book I still owe you), I note that we humans are poor intuitive statisticians, which is going to mean that absolute numbers will impress us while rates won't; we won't even care much about rates. 10,000 murders sounds like a lot to us whether we're aware that the population of the U.S has burgeoned or not. And it is a lot, nothing to shrug off. It's just that we can't logically support a conclusion that the society is more violent and dangerous, because rates are the controlling statistic here.
When it comes to certain kinds of violence, there will be a decrease in the absolute numbers as well, which of course means a plummeting rate. Racially motivated violence is probably an example; violence against women also (again, despite the horrors that reach us from places like India); violence against children (again, despite the impression we have of a crisis, and of course excluding the abortion question); and violence against animals.