Enlightenment Now: The Case for Reason, Science, Humanism, and Progress

No, changes over time may be statistical, with unpredictable fluctuations, without being cyclical, namely oscillating like a pendulum between two extremes. That is, even if a reversal is possible at any time, that does not mean it becomes more likely as time passes. (Many investors have lost their shirts betting on a misnamed “business cycle” that in fact consists of unpredictable swings.) Progress can take place when the reversals in a positive trend become less frequent, become less severe, or, in some cases, cease altogether.

How can you say that violence has decreased? Didn’t you read about the school shooting (or terrorist bombing, or artillery shelling, or soccer riot, or barroom stabbing) in the news this morning?

A decline is not the same thing as a disappearance. (The statement “x > y” is different from the statement “y = 0.”) Something can decrease a lot without vanishing altogether. That means that the level of violence today is completely irrelevant to the question of whether violence has declined over the course of history. The only way to answer that question is to compare the level of violence now with the level of violence in the past. And whenever you look at the level of violence in the past, you find a lot of it, even if it isn’t as fresh in memory as the morning’s headlines.

All your fancy statistics about violence going down don’t mean anything if you’re one of the victims.

True, but they do mean that you’re less likely to be a victim. For that reason they mean the world to the millions of people who are not victims but would have been if rates of violence had stayed the same.

So you’re saying that we can all sit back and relax, that violence will just take care of itself.

Illogical, Captain. If you see that a pile of laundry has gone down, it does not mean the clothes washed themselves; it means someone washed the clothes. If a type of violence has gone down, then some change in the social, cultural, or material milieu has caused it to go down. If the conditions persist, violence could remain low or decline even further; if they don’t, it won’t. That makes it important to find out what the causes are, so we can try to intensify them and apply them more widely to ensure that the decline of violence continues.

To say that violence has gone down is to be naïve, sentimental, idealistic, romantic, starry-eyed, Whiggish, utopian, a Pollyanna, a Pangloss.

No, to look at data showing that violence has gone down and say “Violence has gone down” is to describe a fact. To look at data showing that violence has gone down and say “Violence has gone up” is to be delusional. To ignore data on violence and say “Violence has gone up” is to be a know-nothing.

As for accusations of romanticism, I can reply with some confidence. I am also the author of the staunchly unromantic, anti-utopian The Blank Slate: The Modern Denial of Human Nature, in which I argued that human beings are fitted by evolution with a number of destructive motives such as greed, lust, dominance, vengeance, and self-deception. But I believe that people are also fitted with a sense of sympathy, an ability to reflect on their predicament, and faculties to think up and share new ideas—the better angels of our nature, in the words of Abraham Lincoln. Only by looking at the facts can we tell to what extent our better angels have prevailed over our inner demons at a given time and place.

How can you predict that violence will keep going down? Your theory could be refuted by a war breaking out tomorrow.

A statement that some measure of violence has gone down is not a “theory” but an observation of a fact. And yes, the fact that a measure has changed over time is not the same as a prediction that it will continue to change in that way at all times forever. As the investment ads are required to say, past performance is no guarantee of future results.

In that case, what good are all those graphs and analyses? Isn’t a scientific theory supposed to make testable predictions?