As regular readers of this blog know, I'm a longtime Neal Stephenson fan. I've read nearly every book he's published, (sorry, The Cobweb), and excepting his early books The Big U and Zodiac, I've read them in the order they were published. So I've been attuned to his growth as a writer as he went from Snow Crash to The Diamond Age to Cryptonomicon to The Baroque Cycle. While this last one -- being a three volume, 2,700 page historical fiction epic about scientists in the late 17th century -- let down many of his sci-fi fanboys, it is probably my favorite work of his, combining interesting ideas and historical facts with swashbuckling adventure and his best prose to date. (You can read my review of it in the Harvard Law Record archives.)

Anathem, his latest novel and clocking in at 900 pages, is similar to his pre-Cryptonomicon work -- it has fascinating speculative ideas but there's also a slight but noticeable decline in his writing craft. There's something about his historical writing that improves his prose -- perhaps it's the Charles Dickens influence. And without giving away any spoilers, I'll mention another problem that Anathem struggles with and sometimes conquers that I'll call the "Slow Learner" problem, after Thomas Pynchon's introductory essay to his short story collection of the same name.

In the essay, the overly humble Pynchon criticizes some of his early short stories for a mistaken sense of priorities:

It is simply wrong to begin with a theme, symbol or other abstract unifying agent, and then try to force characters and events to conform to it.

He was talking partially about his famous story "Entropy", where the characters at a party act as a representation of the third law of thermodynamics. I actually think that Stephenson was fully aware of this issue while crafting Anathem, and the book steers mostly clear of it by avoiding the metaphorical approach, but I do think that the last third of the book sometimes puts its philosophical ideas above structure and character. (I think this is partially why the New York Times reviewer says he isn't sure Anathem is a proper novel.)

Let me be clear: I hold Neal Stephenson to a high standard, and there are only a handful of writers who I could read a 900-page book by and be constantly entertained and intellectually stimulated. But what it meant for me is that, in the end, I had less fun exploring his wild ideas because they weren't fully earned, whereas here is what I wrote four years ago about the second volume of The Baroque Cycle:

[The Confusion] is Stephenson at his finest, intermixing swashbuckling and piracy with surprisingly engaging digressions on such things as the antiquated timber trade centered in Lyon, France.

The good news is that Stephenson at his finest is to be found in Anathem, but it's mostly in the first third. While the beginning of the book is slow to start due to the heavy load of a made-up vocabulary, much of the fun is learning about the "maths" of the fictional world of Arbre. (If you're interested in details, try the NYT review I linked to above.) The experience of reading this is a lot like a Harry Potter story starring philosophy graduate students. Indeed, in one of his recent interviews, Stephenson explicitly cites Harry Potter as a reason why he thinks readers are comfortable enough with made-up words to take on his latest.

If you're already a fan of Stephenson, then you'll likely enjoy Anathem despite its shortcomings. If not, I highly recommend you start with Cryptonomicon -- if you like what you see there, you'll eventually make it to the philosophically (and sociologically) fascinating ideas of Anathem.