Then, as she got to know him better, she found herself wanting to talk about her work. And she found, with some surprise, that Arthur didn't want to hear about it. Arthur was threatened by her work; he had problems about achievement. He was nominally a stockbroker - an easy thing for a rich man's son to be - and he talked with authority about money, investments, interest rates, bond issues. But there was an aggressive quality in his manner, a defensiveness, as if he were substantiating himself.

And then she realized what she should have known from the beginning, that Arthur was chiefly interested in her because she was substantial. It was - in theory - more difficult to impress her, to sweep her off her feet, than it was to impress the little actresses who hung out at Bumbles and the Candy Store. And therefore more satisfying.

Finally her role had begun to bother her, and she no longer drew pleasure from being frivolous around him, and everything became vaguely depressing. She recognized all the signs: her work at the hospital became busier, and she had to break dates with him. When she did see him, she was bored by his flamboyance, his restless impulsiveness, his clothes, and his cars. She would look at him across the dinner table and try to find what she had once seen. She could not find even a trace of it. Last night she had broken it off. They both knew it was coming.

Why did it depress her?

"You stopped talking," Gerhard said.

"I don't know what to say... Now is the time for all good men to come to the aid of the patient. The quick brown fox jumped over the pithed frog. We are all headed for that final common pathway in the sky." She paused. "Is that enough?"

"A little more."

"Mary, Mary, quite contrary, how does your garden grow?

I'm sorry I don't remember the rest. How does the poem go?" She laughed.

"That's fine, we have the level now."

She looked up at the loudspeaker. "Will you be interfacing at the end of the series?"

"Probably," Gerhard said, "if it goes well. Rog is in a hurry to get him onto tranquilizers."

She nodded. This was the final stage in Benson's treatment, and it had to be done before tranquilizers could be administered. Benson had been kept on sedation with phenobarbital until midnight the night before. He would be clearheaded this morning, and ready for interfacing.

It was McPherson who had coined the term "interfacing." McPherson liked computer terminology. An interface was the boundary between two systems. Or between a computer and an effector mechanism. In Benson's case, it was almost a boundary between two computers - his brain and the little computer wired into his shoulder. The wires had been attached, but the switches hadn't been thrown yet. Once they were, the feedback loop of Benson-computer-Benson would be instituted.

McPherson saw this case as the first of many. He planned to go from epileptics to schizophrenics to mentally retarded patients to blind patients. The charts were all there on his office wall. And he planned to use more and more sophisticated computers in the link-up. Eventually, he would get to projects like Form Q, which seemed farfetched even to Ross.

But today the practical question was which of the forty electrodes would prevent an attack. Nobody knew that yet. It would be determined experimentally.

During the operation, the electrodes had been located precisely, within millimeters of the target area. That was good surgical placement, but considering the density of the brain it was grossly inadequate. A nerve cell in the brain was just a micron in diameter. There were a thousand nerve cells in the space of a millimeter.

From that standpoint, the electrodes had been crudely positioned. And this crudeness meant that many electrodes were required. One could assume that if you placed several electrodes in the correct general area, at least one of them would be in the precise position to abort an attack.

Trial-and-error stimulation would determine the proper electrode to use.

"Patient coming," Gerhard said over the loudspeaker. A moment later, Benson arrived in a wheelchair, wearing his blue-and-white striped bathrobe. He seemed alert as he waved to her stiffly - the shoulder bandages inhibited movement of his arm. "How are you feeling?" he said, and smiled.

Chapter 8

"I'm supposed to ask you."

"I'll ask the questions around here," he said. He was still smiling, but there was an edge to his voice. With some surprise, she realized that he was afraid. And then she wondered why that surprised her. Of course he would be afraid. Anyone would be. She wasn't exactly calm herself.

The nurse patted Benson on the shoulder, nodded to Dr.

Ross, and left the room. They were alone.

For a moment, neither spoke. Benson stared at her; she stared back. She wanted to give Gerhard time to focus the TV camera in the ceiling, and to prepare his stimulating equipment.

"What are we doing today?" Benson asked.

"We're going to stimulate your electrodes, sequentially, to see what happens."

He nodded. He seemed to take this calmly, but she had learned not to trust his calm. After a moment he said, "Will it hurt?"

"No."

"Okay," he said. "Go ahead."

Gerhard, sitting on a high stool in the adjacent room, surrounded in the darkness by glowing green dials of equipment, watched through the one-way glass as Ross and Benson began to talk.

Alongside him, Richards picked up the tape-recorder microphone and said quietly, "Stimulation series one, patient

Harold Benson, March 11, 1971."

Gerhard looked at the four TV screens in front of him. One showed the closed-circuit view of Benson that would be stored on video tape as the stimulation series proceeded. Another displayed a computer-generated view of the forty electrode points, lined up in two parallel rows within the brain substance. As each electrode was stimulated, the appropriate point glowed on the screen.

A third TV screen ran an oscilloscope tracing of the shock pulse as it was delivered. And a fourth showed a wiring diagram of the tiny computer in Benson's neck. It also glowed as stimulations traveled through the circuit pathways.

In the next room, Ross was saying, "You'll feel a variety of sensations, and some of them may be quite pleasant. We want you to tell us what you feel. All right?"

Benson nodded.

Richards said, "Electrode one, five millivolts, for five seconds." Gerhard pressed the buttons. The computer diagram showed a tracing of the circuit being closed, the current snaking its way through the intricate electronic maze of Benson's shoulder computer. They watched Benson through the one-way glass.

Benson said, "That's interesting."

"What's interesting?" Ross asked.

"That feeling."

"Can you describe it?"

"Well, it's like eating a ham sandwich."

"Do you like ham sandwiches?"

Benson shrugged. "Not particularly."

"Do you feel hungry?"

"Not particularly."

"Do you feel anything else?"

"No. Just the taste of a ham sandwich." He smiled. "On rye."

Gerhard, sitting at the control panel, nodded. The first electrode had stimulated a vague memory trace.

Richards: "Electrode two, five millivolts, five seconds."

Benson said, "I have to go to the bathroom."

Ross said, "It will pass."

Gerhard sat back from the control panel, sipped a cup of coffee, and watched the interview progress.

"Electrode three, five millivolts, five seconds."

This one produced absolutely no effect on Benson. Benson was quietly talking with Ross about bathrooms in restaurants, hotels, airports-

"Try it again," Gerhard said. "Up five."

"Repeat electrode three, ten millivolts, five seconds," Richards said. The TV screen flashed the circuit through electrode three. There was still no effect.

"Go on to four," Gerhard said. He wrote out a few notes:

#1 -? memory trace (ham sand.).

#2 - bladder fullness.

#3 - no subjective change.

#4 -

He drew the dash and waited. It was going to take a long time to go through all forty electrodes, but it was fascinating to watch. They produced such strikingly different effects, yet each electrode was very close to the next. It was the ultimate proof of the density of the brain, which had once been described as the most complex structure in the known universe. And it was certainly true: there were three times as many cells packed into a single human brain as there were human beings on the face of the earth. That density was hard to comprehend, sometimes. Early in his NPS career, Gerhard had requested a human brain to dissect. He had done it over a period of several days, with a dozen neuroanatomy texts opened up before him. He used the traditional tool for brain dissection, a blunt wooden stick, to scrape away the cheesy gray material. He had patiently, carefully scraped away - and in the end, he had nothing. The brain was not like the liver or the lungs. To the naked eye, it was uniform and boring, giving no indication of its true function. The brain was too subtle, too complex. Too dense.

"Electrode four," Richards said into the recorder. "Five millivolts, five seconds." The shock was delivered.

And Benson, in an oddly childlike voice, said, "Could I have some milk and cookies, please?"

"That's interesting," Gerhard said, watching the reaction.

Richards nodded. "How old would you say?"

"About five or six, at most."

Benson was talking about cookies, talking about his tricycle, to Ross. Slowly, over the next few minutes, he seemed to emerge like a time-traveler advancing through the years. Finally he became fully adult again, thinking back to his youth, instead of actually being there. "I always wanted the cookies, and she would never give them to me. She said they were bad for me and would give me cavities."

"We can go on," Gerhard said.

Richards said, "Electrode five, five millivolts, five seconds."

In the next room, Benson shifted uncomfortably in his wheelchair. Ross asked him if something was wrong. Benson said, "It feels funny."

"How do you mean?"

"I can't describe it. It's like sandpaper. Irritating. Gerhard nodded, and wrote in his notes, "#5 - potential attack electrode." This happened sometimes. Occasionally an electrode would be found to stimulate a seizure. Nobody knew why - and Gerhard personally thought that nobody ever would. The brain was, he believed, beyond comprehension.

His work with programs like George and Martha had led him to understand that relatively simple computer instructions could produce complex and unpredictable machine behavior. It was also true that the programmed machine could exceed the capabilities of the programmer; that was clearly demonstrated in 1963 when Arthur Samuel at IBM programmed a machine to play checkers - and the machine eventually became so good that it beat Samuel himself.

Yet all this was done with computers which had no more circuits than the brain of an ant. The human brain far exceeded that complexity, and the programming of the human brain extended over many decades. How could anyone seriously expect to understand it?

There was also a philosophical problem. Goedel's Theorem: that no system could explain itself, and no machine could understand its own workings. At most, Gerhard believed that a human brain might, after years of work, decipher a frog brain. But a human brain could never decipher itself in the same detail. For that you would need a superhuman brain.

Gerhard thought that someday a computer would be developed that could untangle the billions of cells and hundreds of billions of interconnections in the human brain. Then, at last, man would have the information that he wanted. But man wouldn't have done the work - another order of intelligence would have done it. And man would not know, of course, how the computer worked.

Morris entered the room with a cup of coffee. He sipped it, and glanced at Benson through the glass. "How's he holding up?"

"Okay," Gerhard said.

"Electrode six, five and five," Richards intoned.

In the next room, Benson failed to react. He sat talking with Ross about the operation, and his lingering headache. He was quite calm and apparently unaffected. They repeated the stimulation, still without change in Benson's behavior. Then they went on.

"Electrode seven, five and five," Richards said. He delivered the shock.

Benson sat up abruptly. "Oh," he said, "that was nice."

"What was?" Ross said.

"You can do that again if you want to."

"How does it feel?"

"Nice," Benson said. His whole appearance seemed to change subtly. "You know," he said after a moment, "you're really a wonderful person, Dr. Ross."

"Thank you," she said.

"Very attractive, too. I don't know if I ever told you before."

"How do you feel now?"

"I'm really very fond of you," Benson said. "I don't know if I told you that before."

"Nice," Gerhard said, watching through the glass. "Very nice."

Morris nodded. "A strong P-terminal. He's clearly turned on."

Gerhard made a note of it. Morris sipped his coffee. They waited until Benson settled down. Then, blandly, Richards said, "Electrode eight, five millivolts, five seconds."

The stimulation series continued.

2

At noon, McPherson showed up for interfacing. No one was surprised to see him. In a sense, this was the irrevocable step; everything preceding it was unimportant. They had implanted electrodes and a computer and a power pack, and they had hooked everything up. But nothing functioned until the interfacing switches were thrown. It was a little like building an automobile and then finally turning the ignition.

Gerhard showed him notes from the stimulation series. "At five millivolts on a pulse-form stimulus, we have three positive terminals and two negatives. The positives are seven, nine, and thirty-one. The negatives are five and thirty-two."

McPherson glanced at the notes, then looked through the one-way glass at Benson. "Are any of the positives true P's?"