A scientific theory makes predictions in experiments in which the causal influences are controlled. No theory can make a prediction about the world at large, with its seven billion people spreading viral ideas in global networks and interacting with chaotic cycles of weather and resources. To declare what the future holds in an uncontrollable world, and without an explanation of why events unfold as they do, is not prediction but prophecy, and as David Deutsch observes, “The most important of all limitations on knowledge-creation is that we cannot prophesy: we cannot predict the content of ideas yet to be created, or their effects. This limitation is not only consistent with the unlimited growth of knowledge, it is entailed by it.”
Our inability to prophesy is not, of course, a license to ignore the facts. An improvement in some measure of human well-being suggests that, overall, more things have pushed in the right direction than in the wrong direction. Whether we should expect progress to continue depends on whether we know what those forces are and how long they will remain in place. That will vary from trend to trend. Some may turn out to be like Moore’s Law (the number of transistors per computer chip doubles every two years) and give grounds for confidence (though not certainty) that the fruits of human ingenuity will accumulate and progress will continue. Some may be like the stock market and foretell short-term fluctuations but long-term gains. Some of these may reel in a statistical distribution with a “thick tail,” in which extreme events, even if less likely, cannot be ruled out. Still others may be cyclical or chaotic [...] For now we should keep in mind that a positive trend suggests (but does not prove) that we have been doing something right, and that we should seek to identify what it is and do more of it.
When all these objections are exhausted, I often see people racking their brains to find some way in which the news cannot be as good as the data suggest. In desperation, they turn to semantics.
Isn’t Internet trolling a form of violence? Isn’t strip-mining a form of violence? Isn’t inequality a form of violence? Isn’t pollution a form of violence? Isn’t poverty a form of violence? Isn’t consumerism a form of violence? Isn’t divorce a form of violence? Isn’t advertising a form of violence? Isn’t keeping statistics on violence a form of violence?
As wonderful as metaphor is as a rhetorical device, it is a poor way to assess the state of humanity. Moral reasoning requires proportionality. It may be upsetting when someone says mean things on Twitter, but it is not the same as the slave trade or the Holocaust. It also requires distinguishing rhetoric from reality. Marching into a rape crisis center and demanding to know what they have done about the rape of the environment does nothing for rape victims and nothing for the environment. Finally, improving the world requires an understanding of cause and effect. Though primitive moral intuitions tend to lump bad things together and find a villain to blame them on, there is no coherent phenomenon of “bad things” that we can seek to understand and eliminate. (Entropy and evolution will generate them in profusion.) War, crime, pollution, poverty, disease, and incivility are evils that may have little in common, and if we want to reduce them, we can’t play word games that make it impossible even to discuss them individually.
This is an extract from Steven Pinker's Enlightenment Now published by Viking