Cryptonomicon


I can't actually recommend it for leisure reading. It's over 900 pages, and not a page-turner like Grisham or Clancy or especially Crichton. It took me almost two months to wade through it.

Neal Stephenson has a vast control of off-beat words, many of which I found myself looking up in the dictionary -- and often not finding. But Wikipedia has numerous entries keyed to Cryptonomicon, including many made-up words Stephenson invented, like "nefandous".

It was not a complete loss. Stephenson  has a wonderful way of getting inside his characters. One of his major characters was a Japanese (he consistently calls them "Nipponese", which I believe is what they call themselves) military man in WWII, and you could actually believe that is how they think.

The primary character -- almost a hero in the classic sense, with super-human abilities -- is Lawrence Pritchard Waterhouse. Waterhouse -- he is always referred to by his last name, never Larry, and by Lawrence only when he is introduced into the story, but not thereafter -- is some kind of mathematical genius, who together with (or independently from, it's never quite clear) Alan Turing, invented modern digital computers. OK, it's fiction. Alan Turing was real, but Waterhouse is fiction. But he's a credible geek, and Stephenson does a very good job of helping the reader experience how geeks think. I know, I are one. Stephenson has the right combination of geekiness and literary skills, so that he accurately portrays how we process data, but does it in a readable way. If I ever wrote a novel (I keep thinking about it from time to time), it would have to be about a geek like Waterhouse, because that's all I know.

Except Waterhouse and most of the other Americans in Cryptonomicon that Stephenson tells us what they are thinking, all of them are sexually obsessed. Maybe people think that way. Stephenson obviously does: one of his chapter titles is "The Spawn of Onan". I prefer to believe otherwise. At least he doesn't put those thoughts into the mind of his non-Americans. I believe that is accurate. American culture is sexually overcharged, but not the rest of the world. I suspect this phenomenon is more recent than the War, but Stephenson obviously does not agree. Grisham and Clancy mostly spare us, but Crichton novels tend to wallow in the gutter too, just not as pervasively as Stephenson. Even Pink Panther novels do it. I could not in good conscience write like that. I suspect the modern American public would find my writing less than credible. That's the biggest reason why I'm not a novelist.

If you could somehow bleep over the vulgarities, Cryptonomicon has some interesting insights. Alan Turing, the fellow who broke the German "enigma" code during the war, and who lent his name to the "Turing Test" for determining if computers can think, was a known homosexual. Actually, it was not known until after the war, then he took the "honorable" British way out when it became known. Stephenson has Turing and fictional German mathematician Rudolf von Hacklheber as lovers at Princeton when Waterhouse was there. He then weaves these three characters through the story line. Anyway, Waterhouse offers this curious insight into the nature of homosexuality:

From an evolution standpoint, what was the point of having people around who were not inclined to have offspring? There must be some good, and fairly subtle, reason for it. [p.11]
He then goes on to suppose the anachronistic explanation modern Darwinists give, that not having kids somehow supports the survival of the whole society. This is of course nonsensical. Whatever benefit homosexual behavior could confer on the rest of society, the first such person to exhibit this behavior in the evolutionary scenario has no offspring, so his DNA is never passed on. Every genetic homosexual thus must be a new mutation, because it never gets passed on to any offspring, even if it confers survival. Stephenson got half of that. No Darwinist gets the other half.

Other insights, like this honest assessment of unix:

Like every other high-powered academic computing network, this one was based on an industrial-strength operating system called UNIX, which had a learning curve like the Matterhorn, and lacked the cuddly and stylish features of the personal computers then coming into vogue. [p.59]
And this one on geeks and society:
Randy was forever telling people, without rancor, that they were full of [it]. That was the only way to get anything done in hacking. No one took it personally. Charlene's crowd most definitely did take it personally. [p.80]
To that same crowd, at a party populated by friends of soon-to-be-ex girlfriend Charlene, Randy adds, "Just because it's an old idea, doesn't mean it's wrong." This is an unusual insight among the Athenians [see Acts 17:21] who populate most of modern culture. Later, a different character praises the Greek goddess Athena. Curious.

More insightful is the ambiguous focus on gold through the last half of the book. Everybody wants gold. During the war the Japanese and the Germans are hiding it, then after the war everybody else is trying to find what they hid. The other half of the ambiguity is put in the mouth of Goto Dengo, the Japanese military man:

"The leaders of Nippon were stupid. They took all of the gold out of Tokyo and buried in holes in the ground in the Philippines! Because they thought that The General [MacArthur] would march into Tokyo and steal it [as the Japanese previously did in China]. But The General didn't care about the gold. He understood that the real gold is here --" he points to his head "-- in the intelligence of the people, and here --" he holds out his hands "-- in the work that they do. Getting rid of the gold was the best thing that ever happened to Nippon. It made us rich. Receiving the gold was the worst thing that happened to the Philippines. It made them poor." [p.858]
I can't tell from the story line whether Stephenson believes that or not, but it's incredibly insightful. Gold is only as valuable as people want to believe of it. Although the world currencies bounce around a lot these days, they get along just fine without gold.

There is also a delightful jibe at British pronunciation. Waterhouse gets himself sent off to England to work with Turing on wartime cryptography. In Stephenson's stream of consciousness style, Waterhouse is confused by all the people there exclaming "Woe to Hice!" and wondering who this Hice fellow is who is in such trouble. Later he figures out that it is the British pronounciation of his own name, Waterhouse. Bye and bye the mispronunciation is dropped, exactly as the perception of a real American would become acclimated to the British way of speaking. I once went to a Shakespeare play in London. For the first quarter hour, I could not make out a word they were saying, but after a while my ear got tuned and it became quite enjoyable. This week I saw an old movie made in London, and one of the actors pronounced "house" as "hice". They really do speak that way.

A large part of the first half of Cryptonomicon is devoted to the nature and technology of encryption. Later in the story Stephenson weaves in a manual encryption technique using a deck of cards, which an appendix written by cryptography expert Bruce Schneier explains in detail. Stephenson has obviously taken the time to understand the technology; I know this because cryptography was an interest of mine when I was in grad school (I implemented a verson of RSA as a project). Turing cracked the German enigma crypto system during the war, and Stephenson expends a lot of his story line explaining how and why the Allies must not let the Germans know they did. Of course we now knew their battle plans and submarine orders, but when the Germans began to realize the code had been cracked, then needed to replace it with a new, harder code, which left the Allies in the dark again. To forestall that result -- according to Stephenson: this is fiction, but it makes sense -- the Allies expended substantial military activities on faking ways to "discover" what they already knew from intercepted and decrypted messages. Waterhouse spends one winter on the fictional island of Qwghlm pretending to listen in on a "huffduff" radio location device which was actually completely inoperable, but the antenna rotated convincingly, so German spies could believe that U-boats were being located by triangulation rather than by intercepting and decoding their orders. Another main character, Bobby Shaftoe, is constantly shuttled around the world on missions designed to fail in strange ways, such as running a merchant ship aground so that the code book can be "compromised", thereby justifying a complete replacement whose real reason is that an intercept said the Germans had cracked it. The fake compromise thus concealed the fact that the German code was no longer secure.

It's an interesting problem. If you have secrets that need encryption, there are many subtle ways for the attacker to figure them out, either by breaking the code, or if that's too difficult, by observing behavior that goes along with the supposed secrets. In the 1990s thread of the story, Waterhouse's grandson Randy is hiding some business secrets from a competitor, who uses "Van Eck Phreaking" to extract images from Randy's laptop computer by analyzing electronic emmissions. Randy foils the assault by keeping his secret numbers off the screen, and -- this is not a spoiler, the reader is told what Randy is thinking -- putting up instead fake numbers that the attacker will think are real, thereby sending him off on a wild goose chase. The card deck crypto system can appear to be about bridge hands [p.785]. At another point Waterhouse speaks "code" [p.572] which is his way of lying through his teeth while imagining himself virtuous. I guess this is not all that different from my use of quotes, except that at least I put the markers there; Waterhouse prides himself in being the only one who could possibly know the truth -- except maybe Alan Turing, if he were there.

I mentioned above that Waterhouse was a "hero" in the classic sense, with super-human abilities. I personally do not believe that some people are smarter than others. We all have the same number of brain cells, but we choose different data to fill them up with. Some people fill their brains with sports stats or romance novels; others with math and science. Guess which one will be better at math and science? I happen to be better than average at programming computers -- and much worse at figuring out how to make people feel affirmed when they are flat wrong. Here Waterhouse is sitting in church thinking about sex and his eventual wife singing in the choir, while at the same time running thought-experiments on how to tune the church organ and suddenly how to build a computer that will eventually crack cryptosystems tougher than enigma, all the while actually playing a Bach fugue on that church organ. [p.576] Bach is some of the most complex keyboard music, and multi-stop, multi-manual organs are the hardest to play; there is no way a real person can think about all that at the same time. But this is fiction.

One of the main characters is Enoch Root, a strange religious nut woven in and out of both time frames. Sometimes he seems to be a Catholic priest, other times he is more of a NewAger, and then he comes off more like a modern secularist. This part is not very credible, and betrays Stephenson's own inexperience with the nature of religion. One of the things I keep noticing in fiction, is that when the author is describing what he knows, it is very good; when he is making something up about a field where he personally is not expert, it comes off badly. Stephenson is no exception. He has Waterhouse thinking at one point that "the churches are merely one branch of the [sex] Control Conspiracy." [p.571] This perception is not later modified as are "wrong" ideas, but only adopted as a means to gratification, so I infer it reflects Stephenson's own thinking about religion.

Stephenson does make one interesting correlation between religion and social function. He finds religious people better able to cope with social dislocation. He paradoxically relates that to UNIX admins, ostensibly because of their better access to documentation [p.585], although I have not found that to be the case in my quest for Linux functionality.

Stephenson does not have a good grasp of linguistic principles. He quotes a few foreign language phrases, but he utterly fails to grasp the linguistic differences between languages -- for example supposing the letters "e" and "t" are most frequent in any language. That is actually another variable that cryptanalysis must take into account. His made-up Qwghlm language just doesn't make sense linguistically. But hey, it's fiction. Similarly, he has Randy taking time to translate some old C code into modern C++ [p.815]. Any programmer worth his salt knows that C++ includes all of C, no translation needed. Nor particularly desirable, since OOPS offers no advantage in computationally intensive work like cryptanalysis.

Since cryptology is about communicating messages, Stephenson does get some parts of linguistics right. The mysterious character Enoch Root starts off his acquaintance with Randy through email, then later by phone, and even later in person. He offers this comment on the fact that Randy recognized him immediately:

Doesn't it strike you as remarkable that you can look at a stream of characters on the screen of your computer -- e-mail from someone you've never seen -- and later 'recognize' the same person on the phone? ... Some complain that e-mail is impersonal -- that your contact with me, during the e-mail phase of our relationship, was mediated by wires and screens and cables. Some would say that's not as good as conversing face-to-face. And yet our seeing of things is always mediated by corneas, retinas, optic nerves,... So is looking at words on a screen so very much inferior? I think not; at least then you are conscious of the distortions. Whereas, when you see someone with your eyes, you forget about the distortions and imagine that you are experiencing them purely and immediately. [p.800]
I'm not sure I would go that far, but some people get offended with my preference for written communication in times of hostility. Those same people tend to have completely wrong ideas about me gathered from personal contact -- and are often surprised at what I am able to discern just from email. Stephenson is obviously onto something.
 

Tom Pittman
2008 October 21