De Landa is deliciously weird sort of scholar: an autodidact, a committed generalist, a erudite synthesizer, and...oddly enough...an ideologue. The axe he has brought to grind here is a rigorous materialism, and he uses it to hack telos out at the root. He seeks to collapse the distinction between "natural" history and "human" history, and the result is a "history" that is almost unrecognizable as such.
He asks us to imagine the last thousand years as a seething storm of material processes. The "great men," the "human events," the wars and values and struggle are all completely absent. To the extent humans interest De Landa at all, they appear here as crowds, organizations, markets, capital and labor.
Instead, De Landa gives us a plausible (if sketchy and somewhat speculative) account of the thermodynamic, geological, chemical, and biological processes of the past 1000 years. But the genius of this book is that it is not merely the history of rocks, chemicals, and plants. Instead, De Landa has boldly abstracted the logical processes underlying the natural sciences into what he calls "engineering diagrams." He applies these diagrams to the world we know, teaching us to see city walls as sea-shell-like "accretions", society as a stratified riverbed, economies as highly complex chemical reactions, and nations as parasitic superorganisms. Above all, he helps us to see "progress" as a perspectival illusion, resulting from human-centric narrative bias. Again and again, he demonstrates that the "triumphs" of the Western world were spontaneous physical processes; reactions between elements like "biomass" "carbon" "steel" "money" "genes" "population" and "germs." These reactions become interactions, feedback takes hold and wildly complex and diverse forms emerge. These forms bifurcate, find relatively stable states and then, inevitably, collapse again. And Delanda insists that he is NOT speaking in metaphors - the same "diagrams" that lay riverbeds also build empires.
De Landa's method is problematic, controversial, and likely to turn some readers off. Since his ideas are vague, abstract, and probably untestable, many will call them unscientific. But the book is a sketch, a manifesto, a prologomenon to a new way of looking at the world. And, as such, it is extremely thought-provoking. One does not have to agree with De Landa's neo-Marxism to be stimulated when he argues that there is no "clash of civilizations," that there will never be an "end of history," and that the fundamental factor which distinguished the "West" from the "Rest" is not religion or technology but...um..."autocatalysis."
Ultimately, this is a very fun and eccletic little read that is likely to tweak your perspective more than a bit. Hayekians, especially, will be intrigued (though perhaps not persuaded) by his discussion of "anti-markets" and the VIRTUES (!) of top-down decisionmaking.
As a final note, some prior familiarity with Ilya Prigogine's work is very helpful for full enjoyment of this book. De Landa relies heavily on Prigogine's thermodynamics (the concepts, not the math), and he does a fairly poor job of introducing them. As a layperson, I found Schneider and Sagan's Into the Cool: Energy Flow, Thermodynamics, and Life a very helpful introduction.