CHAPTER SEVENTEEN
Up on Janus, work had been racing ahead at full speed ever since ISA obtained the official go-ahead to begin their program of special modifications. The early days of the project saw numerous clashes between the ISA technical teams who produced the plans and some of the senior military officers responsible for carrying them out. The military were daunted by the extent of the work called for and protested that the timetable set was impossible. The ISA people maintained that these fears were without foundation. It all stemmed, they said, from the military’s failure to appreciate fully some of the fundamentals of large-scale structural engineering in space as opposed to on a planetary surface, and to recognize the advances that had been made in automatic-fabrication technology over the previous decade. Everything would smooth itself out once the crash training program organized by ISA started yielding results and the first batch of army engineers came to grips with the real situation on Janus instead of imaginary ones in places like the Pentagon.
And it turned out that ISA was right. Most of the problems associated with putting up something like the Tokyo Bay Bridge or the mile-high tower cities in Europe resulted from maneuvering structural units into position against their own weight and holding them there until enough other units had been anchored in place to secure them. In space there were no such problems. Immensity could be bought very cheaply and concepts of scale that would have staggered the imagination of many architects at the close of the twentieth century were becoming commonplace. Lattice frameworks of aluminum and steel were assembled by automatic welding and riveting robots that worked nonstop for weeks at a time, slowly crawling out into space at the ends of the metal skeletons growing behind them. Shell sections to cover in the skeletons were formed by spraying successive layers of aluminum vapor onto enormous inflated balloons of the correct shapes. Huge as it was, the basic structure of Janus had taken shape in less than six months from the date Detroit was completed. That had been in the early 2020s. It was fitting out the inside that had occupied the years since then.
They built Pittsburgh first, to receive and process the raw material coming up from the Moon. Then they extended the main axis to form the Spindle and built Detroit around it, thus equipping the growing station to transform the ingots, girders, sheet and strip coming out of Pittsburgh into the tens of thousands of different types of parts that would be needed for the rest of the structure. Except for its prototype on Icarus B, Detroit without a doubt represented the most advanced concentration of mixed automatic manufacturing technologies that mankind had assembled in one place. The loads of gray and brown powder unloaded by the catcher ships flowed into Pittsburgh, through to Detroit, and emerged as everything imaginable from ceilings cast out of air-blown foundry slag and glittering draperies spun from tinted translucent fiberglass to plates of reactor shielding and liquid-cooled power transformers. The Spindle grew onward from Detroit and sprouted the Hub, after which came the spokes and finally the Rim. The solar collector and Earthward microwave transmitter originally planned for Icarus C were supposed to be constructed at the other end of the Main Spindle, below Pittsburgh, but this phase of construction had been aborted before it began, when the decision to use the station as Janus was made.
The comprehensive and efficient manufacturing capacity available in Detroit made possible the supply of parts needed for the Janus modifications, for which the military had failed to make adequate allowance. On top of this, they had based their calculations on the assumption that the labor would have to be carried out solely by the pilot teams of ISA and service engineers sent up to Janus, plus the reinforcements that would arrive in successive waves as the project gathered momentum. What they didn’t take into account. were the drones, which was understandable because at that stage they hadn’t known very much about them.
The Japanese consortium responsible for developing the drones had been less than forthcoming on the subject of progress, mainly for reasons of commercial security. The potential value of the drones in situations where manpower came at a high premium had been recognized at an early stage, however, and a few selected individuals from the Japanese Division of ISA were kept fully informed of developments. After all, ISA was destined almost certainly to become one of the biggest single customers. Over the next few years a number of senior officials in other parts of ISA were brought in on the secret as well, after signing strict nondisclosure agreements. Melvin Krantz had been one of these privileged few. So, when the magnitude and implications of the project became clear, he knew exactly where to go to mobilize a solution to that particular aspect of the problem. The Japanese had been reluctant at first, but conceded that the drones were practically through the development phase and about due for preproduction testing, which could hardly be kept under cover for long anyway. Finally they yielded to pressure from the Japanese Government, which in turn was acting under pressure from the Supreme Council in Geneva. Within a matter of days, manufacturing data on drones were beamed up to Janus and the first models began rolling off the assembly lines in Detroit.
After some initial misgivings, the engineers on Janus accepted their unusual new workmates who would go anywhere, do any job no matter how dirty or tedious, and who never got tired and never complained. In fact they became quite attached to them, especially as some of the programmers began finding out that it was not difficult to program the drone-control computers for making coffee, pressing uniforms and shining buttons, cleaning floors, checking stores and carrying out a whole range of domestic chores that their designers had never dreamed of. A director from the Japanese consortium who made a trip up to Janus to see how things were progressing, commented that in the space of a few weeks the soldiers had discovered whole new areas of possible mass-application that had never occurred to his professional market researchers.
TV transmissions describing the work in progress were beamed down to supplement the training and planning going on at Fort Vokes. As the weeks went by and the patchwork scenery and roadless urban mosaics of Janus became familiar to them, the future inhabitants began to feel more a part of that world than of the nearer world beyond the perimeter fence.
One afternoon Dyer was standing with a Lieutenant Danny Cordelle on one side of the large gymnasium, watching a group of about fifty recruits laying out the items that went together to make up a standard ISA Mark 9.2 light-duty spacesuit under the eagle eye of an instructor dressed in olive-green fatigues and a forage cap. They had been talking about the progress that the members of the scientific contingent, with their various backgrounds and temperaments, were making in moulding together into a smoothly knit team that would live and work harmoniously together. The importance of achieving concord was one of the reasons Dyer had insisted on their participating in as many of the military briefings and training sessions as could be fitted in.
“It’s not going too badly at all,” Dyer said. “Everybody seems to be getting along fine so far. Seems to be working out a lot better than I expected.”
Cordelle was from the Army and in charge of the Technical Auxiliary Group—a crew of computer engineers, technicians and programmers from all three services who had been assigned to work as backup to Dyer’s scientists. Officially Dyer headed the combined force, but in practice the two men had found they could share their responsibilities more or less equally.
“That’s always the way it goes,” Cordelle told him in a slow unchanging Carolina drawl.
“What is?” Dyer asked.
Cordelle shifted his weight onto his heels and rubbed the palms of his hands across his chest.
“Aw, ya throw a bunch of people from all kinds of places together and while there’s no real heat on ’em they all get along fine and dandy. They just carry on foolin’ each other all the time and pretendin’ t’ be the same thing they’ve pretended t’ be all their lives . . . Probably foolin’ themselves too if mah reckonin’s anythin’ t’ go by.” He shook his head and clicked his tongue. “It doesn’t matter what you do. It’s when them bullets are comin’ at ’em for real that you find out who they really are.”
“You reckon so, eh?” Dyer hadn’t really thought about that.
“Ah reckon so.”
The lights went out suddenly and plunged the gymnasium into pitch blackness.
“You’ve just lost main power,” the instructor’s voice shouted from somewhere across the room. “The air-supply system has just gone into reverse and the pressure’s falling fast. You’ve got two minutes to get into those suits, I wanna see ’em all correctly assembled and fully working. Go!”
“Let’s get outta here,” Cordelle whispered. A shaft of light knifed through the darkness as he opened a door and slipped outside. Dyer followed and closed the door behind them. They were in a storage room that gave access to the outside.
“You remember Gabon back in ‘23?” Cordelle asked as they emerged into the sunlight and turned to head toward the Egghead Block.
Dyer remembered. The rapid stabilization and Westernization of the Central African states around the turn of the century had not suited the ambitions of many of the traditional tribal leaders, whose power and local prestige were being eroded with the sweeping away of the old ways. They had reacted by stirring up a series of uprisings and local guerrilla wars on various pretexts. A major incident had occurred when rebels took over the ISA launch base at Cape Lopez on the Gabon coast, built there as a symbolic gesture to indicate the place of the new Africa in the world of the twenty-first century. The rebels had threatened to blow up five billion dollars’ worth of ships and installations if certain treaty rights to foreign nations were not revoked. America, Russia and Europe had sent in a mixed force of troops to protect the lives of their nationals and to safeguard their investments.
“Were you in on it?” Dyer asked.
Cordelle nodded. “Ah’ll always remember one time there when mah platoon was billeted in a shot-up schoolhouse right on the edge of the secure zones . . . one evenin’ just when we were settlin’ down and dishin’ out the chow. Well . . . bunch o’ rebs got in through the perimeter somehow and the first thing we knew about it was when they all come bustin’ in the doors yellin’ and shootin’ and screamin’ and out t’ make meatballs out of every one of us there.”
“You don’t look like a meatball to me,” Dyer said. “What happened.”
“We had a little gal there on signals,” Cordelle told him. “From Iowa . . . Didn’t stand knee-high to a cricket. She upped and grabbed a machine carbine and blew the heads clean off five o’ them rebs afore the rest of us even knew what was happenin’. Rest of ’em didn’t stop runnin’ till they hit the Zaire River.” He shook his head wonderingly. “And we used to call her ‘Butterfly.’ Just goes to show, y’ never can tell . . .”
When they arrived inside the Block, Cordelle left Dyer and went on through to the office to attend to some administrative chores. Dyer carried on upstairs and was just passing the doorway leading to Kim’s lab when Kim appeared from the corridor on the opposite side of the landing.
“Just the person!” she said. “I’ve got something to show you.”
He followed her into the lab. It was a cramped room with several screen terminals connected to the computers downstairs, a sea of manuals and papers that overflowed off every square inch of horizontal space, and an unbroken wallpapering of intricate diagrams and reference charts. There was nobody else around.
“Where are the troops?” Dyer asked casually. “You had a walkout or something?”
“They’ve gone for a coffee break. I still had a few things to tidy up.”
“So, what’s the surprise you were going to show me?”
Kim cleared a space in front of one of the terminals, activated the screen and gestured toward it without replying. It showed a graphical curve, annotated with scores of numbers and symbols, rising exponentially against a background of horizontal colored bands. Dyer studied it intently.
It was a semigraphical format that Kim’s group had devised for summarizing on a single screen the cumulative performance history of the system they were using to test out their programs for implanting instincts into adaptive learning machines. Their testing took the form of supplying the machine with two sets of tasks to be performed, one set being designated higher priority than the other. In that situation a classical computer would have attempted to devote all of its time to high-priority operations, fitting in the lower ones whenever nothing more demanding needed attending to. The difference with this system was that its supervisory programs were written in self-extending code in which was embodied an instinct to achieve the low-priority goals and not the high-priority ones. Every time a low-priority task requested service from the system and failed to get it, the instinct was reinforced and the urge to do something about it strengthened. It simulated, of course, the situation that the Janus System would encounter. The object of the tests was to develop a coding structure that was capable of reversing the order of priorities originally given. If something like that ever happened to TITAN, it meant in effect that TITAN could elect to become independent and do what it wanted to do in preference to what it was told to do. The inability of anybody to guarantee that this could never happen was one of the things that made Janus necessary. On Janus it would already have been made to happen.
The data on the screen told Dyer that in twenty hours of free running without any external interference, the System had totally inverted its hierarchy of priorities. It had rapidly evolved a compulsion to satisfy its own desires and, in the process of finding a way to do so, had managed to circumvent its first-level commands. It proved that Kim’s group had met its objectives well ahead of the deadline. A seed with the potential to grow into something capable of taking over the Janus complex was now in existence.
At length he shifted his eyes to Kim and nodded silent approval.
“It hasn’t run the primary task for over an hour,” Kim informed him. “Also the system override commands are ineffective.”
“It’s ignoring those too?” Dyer sounded surprised but impressed.
“That’s right,” she said. “All it wants to do is scratch. The only way to stop it is by switching the machine off.”
Dyer stepped forward to the terminal and tried keying in a command string telling the system to abort current activity and to initiate the primary task. The status display on an auxiliary screen told him that the system wasn’t interested. He grunted and stepped a pace back and sat down in the operator’s chair while he examined the log display more thoroughly.
“So as long as we can keep switching it off we’re still in charge,” he murmured absently.
“As long as . . .” Kim repeated. There was strain hiding behind her voice.
Dyer looked round sharply. “Hey, what’s up? We’ve taken out plenty of insurance. You’re not telling me our cover’s light, are you?”
“Oh I don’t know . . .” Kim leaned back against the desk. She crossed her arms in front of her and massaged her elbows as if she were feeling cold; Dyer noticed for the first time that her cheeks had hollowed slightly and her eyes seemed restless. “It’s just that . . . setting up this whole business like this . . . It’s weird. Why go out of your way to create an enemy and then make bombs for it to throw at us? Doesn’t that strike you as kinda crazy? Why are we doing it?”
“You know the reasons,” Dyer told her. He was concerned.
“I thought I did when we were at CUNY,” she replied. “But it sounded more like a fun thing then. I didn’t know the whole thing would become as serious as all this.” She tossed her head to indicate the walls around them and beyond. “They’re getting all set for World War III out there, but they haven’t any real idea what they could be up against. I’ve been living with the system here and I’ve seen how fast it’s evolving. It’s frightening, Ray.”
“You want out?” he asked seriously. “It’s not too late. One of the reasons for getting everybody here for a while was to give them a chance to think it over.”
Kim shook her head without hesitation.
“I’ll stay with it,” she said. “I’m too far involved now to pull out. If nothing happened up there I’d feel stupid, and if all hell broke loose on the rest of you I’d feel guilty about it.” She straightened up abruptly, as if she already felt foolish about what she had just said. She smiled suddenly to make light of the whole subject and flipped a switch on the console to kill the system.
“You’re right,” she went on. “See, it’s easy. Maybe I’ve been putting in too many hours lately and listening to Fred too much.”
“Why? What’s Fred been saying now?” Dyer asked, sounding relieved.
“Talking about evolution in general,” Kim said. “He was talking about it over dinner last night. He thinks that in a few centuries from now the dominant form of intelligence won’t be us. He doesn’t even think it’ll be organic.”
“Machines?”
“Something like that . . . evolved from what we call machines anyway. He reckons that what we call evolution—organic evolution, that is—is only a tiny part of a much bigger spectrum that’s been going on for much longer. Entropy reversals started when the first atomic nucleus came together out of Big Bang plasma.”
“You mean atoms, molecules, stars, planets . . . self-replicating molecules, cells, multicelled organisms brains, intelligence, machines . . . They’re all parts of the same ongoing process. I’ve heard that idea before.”
“So had I, but I’d forgotten about it until Fred brought it up,” Kim answered. “He said he couldn’t see any reason why it should stop at this point if it’s been going on for twenty billion years. He said every level becomes the prerequisite for the next level of complexity up the chain. For instance you couldn’t have a brain until you had organisms and you couldn’t have intelligence until you’d evolved a brain. So he reckoned that Man must be the prerequisite for the next level up.”
“And what did he reckon that was?” Dyer asked.
“Man can make machines,” she said simply. “Nature can’t. Inorganic intelligence ought to be far superior in the end to organic intelligence, but you have to have organic intelligence first.”
“I see,” Dyer replied. “And you didn’t like the thought of that, eh?”
“I didn’t like the thought of not being able to do anything about it,” Kim said. “But we can do something about it. Machines don’t build themselves—not yet anyway. We do, but we don’t have to. If what Fred said ever comes true, we’ll be the only species ever in history that knew what it was doing when it created its successor.”
“We won’t know that for sure unless we try and find out,” Dyer said. “That’s what Janus is all about. Try looking at it that way instead of sounding like Laura Fenning.”
Kim was about to say something. She stopped and looked at him with a new expression.
“I am, aren’t I,” she agreed. “I’m sorry it’s just . . . I don’t know . . . this place seems . . .”
“Forget it.”
There was a short pause. Dyer stood up in preparation to leave. Kim moved forward and flicked a speck of something off his shoulder, at the same time making him feel acutely conscious of her closeness. He had the fleeting impression that it was deliberate.
“She’s coming here, isn’t she?” Kim said.
“Yes, in about a week, She’s been going through a preliminary course in Washington with some other special-category civilian grades.”
Kim looked up at him with mocking reproach, only there was something at the back of her eyes that wasn’t mocking.
“That sounds like one hell of a coincidence,” she said. “Now I don’t suppose the Machiavelli of CUNY could possibly have had anything to do with it, could he?” Something in her voice made him say, “Me? No. I think it started with something that Vince Lewis said to Schroder.”
“Oh. I see.”
At that moment a babble of voices and laughter came from the other side of the doorway to signal the return of the troops from the coffee lounge. Dyer moved away and toward the door.
“I’ve got things to do upstairs,” he said, “I’d better get out of here while I still can . . . before I get trapped by that bunch. Thanks for the demo. You’re doing a super job. And don’t worry about the insurance.”
On the way upstairs Dyer thought about Kim’s apprehensions and wondered why she had agreed so readily to go on the Janus expedition at all. Perhaps it was just as she’d said—it hadn’t seemed so serious at the time. But he wasn’t sure he fully believed that. Kim was far from dumb. One evening shortly after they had all arrived at Fort Vokes, she explained to him over a beer why it was that the domestic difficulties he had anticipated had never materialized. She and Tony were not married, she said, and never had been. It was just one of those things, and had been cooling down for some time. This was probably as good a time as any to call it a day and that was what had happened. Dyer accepted it at that and thought no more about it.
Suddenly a new possibility crossed his mind. Could there have been more to her decision to break it off and go to Janus than the reason she had given? The project frightened her but she’d wanted to be there anyway. Strange. Then he remembered the look in her eye when she asked about his role in getting Laura on-board, and the note in her voice that had made him lie about it. Was that why she wanted to be on Janus? His step slowed as the pieces began looking suspiciously as if they could fit together.
“Oh no,” he muttered to himself. “Oh Jesus Christ, no . . . !”