Cryptonomicon

hate hackers. But something different happened with the Greeks. The Greeks liked their geeks. That's how we get Athena.'

'I'll buy that-but where does the war-goddess thing come in?'

'Let's face it, Randy, we've all known guys like Ares. The pattern of human behavior that caused the internal mental representation known as Ares to appear in the minds of the ancient Greeks is very much with us today, in the form of terrorists, serial killers, riots, pogroms, and aggressive tinhorn dictators who turn out to be military incompetents. And yet for all their stupidity and incompetence, people like that can conquer and control large chunks of the world if they are not resisted.'

'You must meet my friend Avi.'

'Who is going to fight them off, Randy?'

'I'm afraid you're going to say weare.'

'Sometimes it might be other Ares-worshippers, as when Iran and Iraq went to war and no one cared who won. But if Ares-worshippers aren't going to end up running the whole world, someone needs to do violence to them. This isn't very nice, but it's a fact: civilization requires an Aegis. And the only way to fight the bastards off in the end is through intelligence. Cunning. Metis.'

'Tactical cunning, like Odysseus and the Trojan Horse, or-'

'Both that, and technological cunning. From time to time there is a battle that is out-and-out won by a new technology-like longbows at Crecy. For most of history those battles happen only every few centuries-you have the chariot, the compound bow, gunpowder, ironclad ships, and so on. But something happens around, say, the time that the Monitor,which the Northerners believe to be the only ironclad warship on earth, just happens to run into the Merrimack,of which the Southerners believe exactly the same thing, and they pound the hell out of each other for hours and hours. That's as good a point as any to identify as the moment when a spectacular rise in military technology takes off-it's the elbow in the exponential curve. Now it takes the world's essentially conservative military establishments a few decades to really comprehend what has happened, but by the time we're in the thick of the Second World War, it's accepted by everyone who doesn't have his head completely up his ass that the war's going to be won by whichever side has the best technology. So on the German side alone we've got rockets, jet aircraft, nerve gas, wire-guided missiles. And on the Allied side we've got three vast efforts that put basically every top-level hacker, nerd, and geek to work: the codebreaking thing, which as you know gave rise to the digital computer; the Manhattan Project, which gave us nuclear weapons; and the Radiation Lab, which gave us the modern electronics industry. Do you know why we won the Second World War, Randy?'

'I think you just told me.'

'Because we built better stuff than the Germans?'

'Isn't that what you said?'

'But why did we build better stuff, Randy?'

'I guess I'm not competent to answer, Enoch, I haven't studied that period well enough.'

'Well the short answer is that we won because the Germans worshipped Ares and we worshipped Athena.'

'And am I supposed to gather that you, or your organization, had something to do with all that?'

'Oh, come now, Randy! Let's not allow this to degenerate into conspiracy theories.'

'Sorry. I'm tired.'

'So am I. Goodnight.'

And then Enoch goes to sleep. Just like that. Randy doesn't.

To the Cryptonomicon!

* * *

Randy is mounting a known-ciphertext attack: the hardest kind. He has the ciphertext (the Arethusa intercepts) and nothing else. He doesn't even know the algorithm that was used to encrypt them. In modern cryptanalysis, this is unusual; normally the algorithms are public knowledge. That is because algorithms that have been openly discussed and attacked within the academic community tend to be much stronger than ones that have been kept secret. People who rely on keeping their algorithms secret are ruined as soon as that secret gets out. But Arethusa dates from World War II, when people were much less canny about such things.

This would be a hell of a lot easier if Randy knew some of the plaintext that is encrypted within these messages. Of course, if he knew all of the plaintext, he wouldn't even need to decrypt them; breaking Arethusa in that case would be an academic exercise.

There is a compromise between the two extremes of, on the one hand, not knowing any of the plaintext at all, and, on the other, knowing all of it. In the Cryptonomiconthat falls under the heading of cribs.A crib is an educated guess as to what words or phrases might be present in the message. For example if you were decrypting German messages from World War II, you might guess that the plaintext included the phrase 'HElL HITLER' or 'SIEG HElL.' You might pick out a sequence of ten characters at random and say, 'Let's assume that this represented HEIL HITLER. If that is the case, then what would it imply about the remainder of the message?'

Randy's not expecting to find any HEILHITLERs in the Arethusa messages, but there might be other predictable words. He's been making a list of cribs in his head: MANILA, certainly. WATERHOUSE, perhaps. And now he's thinking GOLD and BULLION. So, in the case of MANILA he could pick out any six-character string from the intercepts and say, 'What if these characters are the encrypted form of MANILA?' and then work from there. If he were working with an intercept only six characters long, then there would be only one such six-character segment to choose from. A seven-character-long message would give him two possibilities: it could be the first six or the last six characters. The upshot is that for a message intercept that is ncharacters long, the number of six-character-long segments is equal to (n— 5). In the case of a 105- character-long intercept, he will have 100 different possible locations for the word MANILA. Actually, a hundred and one: because it's of course possible-even likely-that MANILA is not in there at all. But each of these 100 guesses has its own set of ramifications vis-a-vis all of the other characters in the message. What those ramifications are, exactly, depends on what assumptions Randy is making about the underlying algorithm.

As far as that goes: the more he thinks about it, the more he believes he has some good stuff to go on- thanks to Enoch, who (in retrospect) has been feeding him some useful clues when not spamming him through the bars with theogonical analysis. Enoch mentioned that when the NSA started attacking what later turned out to be the fake Arethusa intercepts, they were going on the assumption that they were somehow related to another cryptosystem dubbed Azure. And sure enough, Randy learns from the Cryptonomiconthat Azure was an oddball system used by both the Nipponese and the Germans that employed a mathematical algorithm to generate a different one-time pad every day. This is awfully vague, but it helps Randy rule out a lot. He knows for example that Arethusa isn't a rotor system like Enigma. And he knows that if he can find two messages that were sent on the same day, they will probably use the same one-time pad.

What kind of mathematical algorithm was used? The contents of Grandpa's trunk provide clues. He remembers the photograph of Grandpa with Turing and von Hacklheber at Princeton, where all three of them were evidently fooling around with zeta functions. And in the trunk were several monographs on the same subject. And the Cryptonomiconstates that zeta functions are even today being used in cryptography, as sequence generators-which is to say, machines for spitting out series of pseudo-random numbers, which is exactly what a one-time pad is. Everything points to that Azure and Arethusa are siblings and that both are just implementations of zeta functions.

The big thing standing in his way right now is that he doesn't have any textbooks on zeta functions sitting around his jail cell. The contents of Grandpa's trunk would be an excellent resource-but they are currently stored in a room in Chester's house. But on the other hand, Chester's rich, and he wants to help.

Randy calls for a guard and demands to see Attorney Alejandro. Enoch Root goes very still for a few moments, and then shunts directly back into the loping, untroubled sleep of a man who is exactly where he wants to be.