Benjamin felt the need to change the subject. 'But you also said something about Vietnam?'
Wolfe looked up at him. 'That was a long time ago,' he said, looking suddenly serious, the typical smirk entirely absent, his eyes not focused on Benjamin.
'I was assigned to SIG-INT, signals intelligence. We were stationed in Saigon. At first, our job was to listen to the Vietcong's radio transmissions.'
'And that's where you learned cryptography?'
Wolfe nodded, but provided no further details.
'And afterwards?' Benjamin prodded.
'There aren't many things a man can do with that kind of experience. Not that I wanted to do, anyway. I kicked around for a while with various security firms, you know, for banks, corporations with offices abroad, that sort of thing.'
'But how…' Benjamin accepted the bottle from Wolfe, waved its neck around the room.
'How did I come to work for the Foundation?' He laughed. 'What does it mean to be secure, Benjamin?'
'I'm not sure what you-'
'That's the first question a security analyst must ask. What does the client wish to 'secure'?' He sighed heavily. 'It's always about fault lines. And I learned that, paradoxically, the best place to look for those fault lines is typically in whatever the client considers their greatest strength.'
Benjamin noticed that the lean lines of Wolfe's face had become tense, hard; that the humor that usually lurked in his eyes and at the corners of his mouth was gone.
'For instance, there are those who consider their love of freedom to be their greatest strength. I worked, here at the Foundation and other places, with such people. I know what they fear, what makes them insecure. And more often than not, it's the threat of that selfsame freedom. Only in others.'
Suddenly there was a little beep from the computer. Immediately Wolfe rose and went to the table. 'It appears our cake is done.'
Benjamin joined him at the table. The progress bar on the computer screen had disappeared, and in its place was a new window.
In the window were two lines, like graphs-the one on the left blue, the one on the right red-rising from opposite sides of the window, like two halves of a bell-shaped curve. But where the top of the curve would be, there was a gap. Underneath the curve were a series of complicated mathematical formulae, leading up to a final one in red:
'Now it's my turn,' said Benjamin. 'What on earth is that?'
'It's a formula for a concept in game theory,' said Wolfe, still studying the graphs. 'Something called a Nash equilibrium.' Wolfe began tracing the lines with a finger, as if touching the screen would somehow communicate to him something intimate about them.
'And what's that?' prodded Benjamin.
'Oh.' Wolfe looked up at him. 'It's when, in a game, there's no advantage to a player changing his strategy as long as the other player keeps doing whatever he's doing. It's a complicated way of saying they're at a stalemate. Do you know a game called The Prisoner's Dilemma?'
'I've heard of it. Something about two prisoners trying to decide whether to rat on one another?'
Wolfe nodded. 'They're told that the first one to betray his fellow will receive a lighter sentence, and the one betrayed will receive a harsher sentence. But if neither one betrays the other, they both receive moderate sentences. In such a game, the best strategy is for both players to remain silent. While neither really wins, it guarantees neither completely loses. It's the best aggregate result possible.'
'But what if they don't know what the other one will do?'
'Exactly. That's the random element. That's why the police-and other security workers,' here he smiled, 'always keep the two suspects separated, and try to convince each one that the other is turning snitch. If they collude, they might agree to stonewall it out and stymie the interrogators. In that case, all three parties will have reached a Nash equilibrium.'
Benjamin looked at the computer screen. 'Well, I have two questions. The first is, what does the eighty percent sign mean?'
'And the other one?' asked Wolfe.
'What has that game got to do with Dr. Fletcher's research on nuclear war?'
'Let me answer the second question first.' Wolfe pulled the chair over to the table. He took the bottle of scotch, studied it as if contemplating another drink, then set it with some finality on the table.
'You heard Herman Kahn mentioned at dinner.' Benjamin nodded. 'Cavendish was right, Kahn practically did invent theorizing about nuclear war. He worked out dozens of scenarios: What if the Soviets strike first? What if we do? What if they use only half their missiles in such a strike? Every conceivable variation of mass death. He was the one who argued that, however horrific the idea might be, it was better to come out of such a war with twenty million dead than a hundred and fifty million dead.'
'You're right,' said Benjamin. 'It does sound horrific.'
'He called it 'thinking the unthinkable.' ' Wolfe pointed at the computer. 'Kahn didn't really bring this level of sophistication to bear on such thinking. That came with computer wizards like Fletcher. And, in his day, Arthur Terrill.' He saw Benjamin's look of surprise. 'Yes, in his early days, Arthur was quite the whiz kid of Armageddon.'
'But he said he knew nothing about Jeremy's research.'
'Apparently Arthur's also become adept at an administrator's most useful skill: lying.'
'But how do you understand this stuff?'
'I told you I recognized the name Anton Sikorsky from Fletcher's list?' Benjamin nodded. 'Well, Anton and I worked together, some years ago, for the Foundation. On this 'stuff.' Not at Fletcher's level. His whole career had been devoted to examining Kahn's scenarios, the assumptions behind them, and subjecting them to rigorous statistical analysis. To establishing the probabilities of such unthinkable events.'
Benjamin turned, walked to the bed, but didn't sit down. Suddenly he turned.
'You mean his program calculates the probability of the Cold War?'
'Yes.'
Benjamin laughed. 'But isn't that like predicting rain after a flood?'
'Not quite.' Wolfe knitted his eyebrows. 'From what I've been able to understand of Fletcher's work, he simply didn't believe it.'
'Believe it?' asked Benjamin, confused. 'How can you not believe in the Cold War?'
'It's more about questioning the fundamental logic of the MAD doctrine-Mutual Assured Destruction-that supposedly kept the Cold War from turning hot. He doesn't come right out and say so, but he seems to imply that, at least statistically, it simply doesn't make sense. Or put in Fletcher's terms, that it has a high probability of such a doctrine being unstable. And if I'm reading this right, the TEACUP program is calculating that probability at 80 percent.'
Before Benjamin could say anything, Wolfe moved the pointer on the screen and slowly dragged the window with the graphs aside, so that the list of files underneath it was now visible. 'And I would assume this is the list of his data points from which to calculate that probability.'
Benjamin stepped closer, leaned down over Wolfe's shoulder.
There were three columns of file names. On the right, Benjamin recognized a number of books about King Philip's War. In the middle was a single file name: 'Gadenhower Data.' And on the left was a list of titles he didn't recognize, but they were clearly all about nuclear war: The Effects of Global Thermonuclear War, Thinking About the Unthinkable, The Fallacies of Cold War Deterrence… and a long list of journal articles and white papers.
Benjamin saw that each of the titles had a little X in front of its name… or so he thought until Wolfe suddenly placed his hand on Benjamin's arm.
'Look there,' said Wolfe, pointing to the bottom of the list.
'Bainbridge Data,' said the bottommost file name on the left. And there was no little X before its name. 'And there,' said Wolfe, pointing to the last entry in the column on the right.
Stzenariy 55, it read.
'Well, here goes,' said Wolfe. He clicked on the 'Bainbridge Data' file, and it opened… to reveal nothing.
