Steven Pinker in Edge.org:
Today the vast majority of the world's people do not have to worry about dying in war. Since 1945, wars between great powers and developed states have essentially vanished, and since 1991, wars in the rest of the world have become fewer and less deadly.
But how long will this trend last? Many people have assured me that it must be a momentary respite, and that a Big One is just around the corner.
Maybe they're right. The world has plenty of unknown unknowns, and perhaps some unfathomable cataclysm will wallop us out of the blue. But since by definition we have no idea what the unknown unknowns are, we can't constructively worry about them.
What, then, about the known unknowns? Are certain risk factors numbering our days of relative peace? In my view, most people are worrying about the wrong ones, or are worrying about them for the wrong reasons.
Resource shortages. Will nations go to war over the last dollop of oil, water, or strategic minerals? It's unlikely. First, resource shortages are self-limiting: as a resource becomes scarcer and thus more expensive, technologies for finding and extracting it improve, or substitutes are found. Also, wars are rarely fought over scarce physical resources (unless you subscribe to the unfalsifiable theory that allwars, regardless of stated motives, are really about resources: Vietnam was about tungsten; Iraq was about oil, and so on.) Physical resources can be divided or traded, so compromises are always available; not so for psychological motives such as glory, fear, revenge, or ideology.
Climate change. There are many reasons to worry about climate change, but major war is probably not among them. Most studies have failed to find a correlation between environmental degradation and war; environmental crises can cause local skirmishes, but a major war requires a political decision that a war would be advantageous. The 1930s Dust Bowl did not cause an American Civil war; when we did have a Civil War, its causes were very different.
More here.