The book is not inane. That is, there's some real weight to the philosophy it's working through. In effect, Harman has taken the worst of Heidegger (everything a thing-in-itself, with Being receding between all these things and placing them thereby in constitutive but unsatisfying relation) and given it a weird, Leibnizian twist (he presents the twist as Leibnizian, anyhow; to call it that, he ends up having to twist Leibniz quite a bit already). He's finished this off by subtracting any necessarily human participation (or so he thinks he's done).*
As an intelligent representative of a current trend in philosophy (object-oriented ontology, speculative realism, or whatever they're calling it presently), Harman is worth reading. The book is speckled with moments of insight and the occasional nicely turned phrase. There's a feeling of creativity running through it.
That said, it's also stylistically wearying. Harman is one of the most affected, Henry-James-on-the-wrong-side-of-the-bed writers I've grumped my way through in years. Don't get me wrong, please: I love a good myth. And I even like a certain amount of coy self-historicizing (though not all that much--and there's *lots* here). The trouble is that Harman doesn't do either well. The myths have a tendency to fall apart at the seams, and the self-historicizing is not so much coy as frankly self-indulgent. Harman neither fictionalizes enough to make the stories sing nor edits enough out to make them fly by. I was repeatedly left feeling he thought I should care about his personal life enough to want to know more. I didn't.
In sum: of interest and short enough that you probably won't feel you've wasted your time, but far from special.
* Disclaimer: Obviously, I don't find Harman's philosophical positions convincing. If this were an academic review, I'd go into why that is. I will, however, note that I quite like what he's done with Leibniz, since in effect he ends up with more or less what Badiou's been saying since the '80s. Even then, though, I suspect he'd do better to engage Badiou directly than to doll Leibniz up.
