This book looks at the collapse of the Roman empire, the Mayan civilization, and those Chaco folks you heard about from the X-Files. Why do civilizations collapse? There's a bunch of theories running around. This book's theory: civilizations tend to get more complex. Sometimes that's a good thing. E.g., you'll have better picnics if you're sufficiently well organized such that not everybody brings potato salad. But complexity accretes; it's hard to dismantle. Years after your civilization has stopped having picnics, don't be surprised to find out that there's still a Federal Department of Potato Salad Coordination somewhere. After a while, you reach a point of diminishing returns, perhaps negative returns. E.g., you invent an internet such that your society can operate more efficiently, yay! But if most of your society's energy was already going into obsolete Salad Management Best Practices, that internet won't help you that much.
It's a fine theory. He points out a difficulty in really proving it though—or in proving a competing theory. When a civilization goes into the final stages of collapse, the record-keeping tends to suffer. So when the theorists poke through the wreckage centuries later, you can't really tell what happened. And of course everybody wants to project their favorite theory onto this pile of vagueness.