David's Sling (10 page)

Hammond speaks. "This is the main screen upon which the Institute carries out its largest and most important decision duels. Its not used too often for that purpose. What we re looking at now is one of our demonstration duels—a duel held over a decade ago to determine the merits of strategic defense."

PAN. Bill looks back at the display in fascination. The colorfiil splashes that streak through the wall resolve, as he approaches, into lines of text. With a few exceptions, the entire screen holds only words, arrows, and rectangles. The rectangles enclose and divide the text displays.

They stop at a console perched high above the audience: clearly it is the display master controller. Hammond continues. "As you can see, the overall dueling area is divided into three sections." He taps a track ball on the console, and a pointing arrow zips across the screen. It circles the left half of the screen, then the right, and finally runs up and down the center band of gray. "Each duel pits a pair of alternatives against one another. Often, the alternatives are negations—one position in favor of some action, and one position against that action. The left part of the screen belongs to the proponent for the action, and the right half belongs to the opponent. These two people are known as the
slant moderators
. They have slanted viewpoints, of course, and they actually act as moderators—anyone can suggest ideas to them for presentation. Of course, no one calls them slant moderators. The nickname for slant moderator is
decision duelist
."

The pointer continues to roam across the center band. "The center is the 'third alternatives' area, where ideas outside of for-or-against may be presented by either of the duelists, or by anyone in the audience. In the duel we have here, the third alternatives section remained closed—no one came up with any striking ways to finesse the question."

The arrow shifts upward. Above the colored swirl of text boxes stands a single line of text, a single phrase. It dominates the screen, with thick letters black as asphalt. The lettering seems so solid Bill wonders whether it is part of the display, or whether it has been etched into the surface in bas relief.

This one phrase running across the top overlaps all three sections of the screen. Bill presumes that top line describes the theme of the duel, the title of the topic under discussion. Reading it now, he sees it does not. Instead, this dark, ominous line—so striking and hypnotic, as if sucking the light from the air—reads:

LET ACCURACY TRIUMPH OVER VICTORY

Despite the hypnotic pull, Bill tears his eyes away from the words. They disturb him.

Hammond speaks. "It's easy to get wrapped up in one's own point of view on a topic. After much study of the matter, we've concluded that you can't overemphasize the need for objective search for
rightness
, as opposed to
victory
." He smiles. "As it happens, we keep records on the duels written by certified slant moderators. We do
not
keep our records based on who wins the duel. We keep them based on whether the decision that comes out of the duel is
correct
. Thus, a duelist can have a perfect record even if he 'loses' every time he moderates."

A dark-haired woman wearing a smart business suit asks, "How do you know who's right? Sometimes it takes years to find out, and sometimes you can never find out. ''

Hammond concurs. "You're quite right. We can't trace the correctness of all the duels, so not every duel yields a record. However, we don't lose things just because it takes a long time to determine the outcome. The memories of the Zetetic data bases are very long indeed. In fact, the Zetetic data bases have started to free those of us who work here from our own short-term concerns. Even if we forget events and decisions as quickly as the news media forget, the knowledge remains available for recall with only a slight effort. We have automated the tracking of all the predictions of duelists, stock-brokers, crystal ball readers, and economists. A couple of years of lucky hits do not impress us." He snorts. "We also keep records of the promises made by politicians. I'm sure no one would be surprised by the results of that comparison."

Hammond leads them halfway down to the audience area, where a pair of work stations sit in friendly proximity. The left work station display shows a section of the left half of the main display—the proponents half. Some early duelist has scratched PRO into the edge of the desk top. The right station shows a part of the right half of the display—the opponent's half; another duelist long ago labeled it CON. Bill runs his finger over the rough cuts in the desk top, amazed by the presence of such graffiti.

The cloth-covered arms of the work station chairs show frays and dark stains. Bill wonders how many hundreds of anxious decision duelists have sat in these chairs, rubbing nervous hands over those arms.

"We also keep score on the development of third alternatives that are better than either of the listed alternatives. In general, duels that settle on third alternatives find the best answers of all."

The lecturer drones on, but his voice blends into the scene as Bill watches the cursor on the main display flicker from point to point.

Beneath the blinking warning are the titles: YES, A USEFUL STRATEGIC DEFENSE CAN BE BUILT. And NO, A USEFUL STRATEGIC DEFENSE CANNOT BE BUILT. Beneath the titles are the assumptions, written in cautionary amber. About a dozen assumptions show for the PRO side, and another dozen on the CON side. On the PRO side is the comment, "We assume we are discussing alternatives for getting high rates of kill against ICBMs. We are not discussing engineering absurdities such as 'perfect' defenses."

Next to this assumption lies a picture of a button. When the arrow touches the button, a further discussion of this assumption—why it is necessary—expands into view.

On the CON side, one three-part assumption stands out. "A strategic defense system must pass three feasibility tests: It must be technically feasible. It must be economically feasible. It must be politically feasible."

Beneath the yellow assumption boxes are the opening arguments. The first of these are the cute slogans with much cleverness but little content, much favored by the media. The duelists put them up even if they disagree with them, just to get them out of the way. Appropriately, here the coloration of the text seems playful--purples and oranges and reds splashing about as though written with a child's crayons.

A purple background marks off a quote on the PRO side:

BUILD WEAPONS THAT KILL WEAPONS, NOT WEAPONS THAT KILL PEOPLE.

The CON duelist colored it purple because the statement has no meaningful content, only emotional appeal: whether a weapon kills people directly or not is irrelevant; what matters is whether the weapon increases or decreases the odds that more or fewer people will die.

On the CON side, the PRO duelist had marked a statement in red: PREVENT THE MILITARIZATION OF SPACE. When Hammond's arrow touches this field, an explanation window blossoms to explain the fetal flaw in this reasoning. Space was already militarized: if a war started, 10,000 nuclear warheads could fall through space in the first half hour. Moreover, as with the purple comment on the pro side, the important issue was not whether a weapon was space-based—the question was its effect on human life. The PRO duelist had placed another purple, satirical statement on level with the PREVENT MILITARIZATION slogan, connecting them with a thin line indicating they were two different ways of saying the same thing. The satirical alternative to PREVENT THE MILITARIZATON OF SPACE was MAKE THE WORLD SAFE FOR NUCLEAR WARHEADS.

Beneath the opening comments the words shrank in size, becoming more densely packed as the two sides parried back and forth with increasing verbosity. Both sides agreed to the format described in the CON assumptions: first came discussion of technical feasibility, then economic, then political.

The technical feasibility debate ran to great lengths. PRO constructed alternative after alternative, only to see each one knocked down by CON. But PRO responded to the CON objections, refining the alternatives to overcome the objections one by one. Bill realizes that he sees the evolution of a high-level design for strategic defense outlined before him.

Two-thirds of the way down the screen, the discussion ends, the PRO side triumphant. They have constructed over a hundred different approaches. CON has marked up all but two with bright red fetal flaws, but those two approaches seem capable of defense against words, and maybe also missiles.

Then the debate on the economics begins. This is a short discussion, surprisingly. Both sides agree upon a single criterion for economic viability: Can the system knock a missile down for less money than it costs to put a missile up? If you can shoot them down for more than they cost, then the defense is cheap; if you cannot, then the whole thing is easily defeated by building more missiles. A small amber button glows next to this agreement, which expands to explain the underlying assumption that the defender is not so much wealthier than the attacker that he can afford an extravagant imbalance. A millionaire can spend thousands of dollars protecting himself from a ten-dollar pistol, for example, and easily afford it.

Here both sides invoke large windows filled with spreadsheet calculations. Again, the PRO side shows one possible way of keeping the costs within the economic limit, while CON shows the other approach would fail. Both sides note the inaccuracies in these forecasts, and the size of the ranges. But only one successful approach is needed. The strategic defense system has passed the economic test.

Hammond explains that the political feasibility test is where the CON duelist had focused his attention all along. It is here that the brilliant thrust took place, the insight that makes this a classic in decision dueling. For though there are several approaches that are technically feasible, and one of those is indeed economically feasible, there are thousands of approaches that would fail. With brutal clarity the CON duelist demonstrates, with one military program after another, that the American military development system cannot select a solution that is better than mediocre. With the wings of the C-5, with the computers of TACFIRE, with the armor of the Bradley, the CON duelist demonstrates mediocre solutions that cost factors of two and three times as much as good solutions should.

The PRO duelist concedes: given the American system of military research and development, strategic defense is a hopeless proposition.

"And as everyone here knows, this early decision duel predicted the future quite accurately." Hammond sighs. "This also demonstrates another fundamental consideration of the decision duel—one that engineers all too often forget: the critical importance of finding an approach that can succeed, not only technically and economically, but also politically. This engineering blind spot mirrors the politician's tendency to forget technical viability. Politicians live in a universe where reality seems to be controlled by the perceptions of other politicians. In the heat of finding an approach that he can get other politicians to agree to, he forgets that there are laws he has no control over."

They walk from the room. The warning in asphalt-black from the top of the dueling display continues to etch itself in Bill Hardie's mind.

CUT. After several more demonstrations of Zetetic networks and techniques, he shuts off his flatcam. There is nothing in the Sampler to help him humiliate the Zetetic Institute.

A man could easily starve, wandering the halls of the Pentagon in search of an exit. The faceless, endless corridors contain few distinguishing landmarks for the novice explorer. And the corridors are truly endless—once aligned on a ring, a person could veer gently at each pentagon-corner and return eventually to the place whence he had started. Of course, whether he recognized his starting point or not was another matter.

Sitting at his mahogany desk in the heart of the great building, Charles Somerset reflected on a story he had once heard about wild turkeys. A wild turkey, when confronted with a fence, would simply spread its wings and leap the fence. But when confronted with a thick tree trunk lying on its side, the turkey would run around the log, that being an easier scheme. So the clever farmer strung a low wire, the height of a tree trunk, in a large circle. Instead of leaping, the turkey would run around the edge of the wire to find its end. By the time it returned to its starting place, the turkey had quite forgotten it had been there before. It would run, around and around, till it collapsed in exhaustion.

Charles didn't know whether the system worked with turkeys, but it certainly did with people in the Pentagon. Dazed, dizzy, and defeated by the corridors, exhausted Pentagon commanders could easily have their wings clipped by the smart operator. For some projects, the clipping took a lot of time and gentle nudging, but it yielded results in the end.

For these reasons, Charles loved the Pentagon. The dingy corridors did not dismay him. The hollow echo of air conditioning, the sometimes painful squeal of old battered chairs, the pounding rhythm of remodeling never quite completed, the echoes of conversations that seemed to linger in the hallways long after the speakers had departed: all contributed to the sensation of fighting under hostile conditions. Such conditions made victories over the maze of circles that much sweeter. Charles seldom noticed that the endless circles had trapped even him. Only on days like today did he feel encircled himself.

Today, he felt like a wagon train struggling against a circle of Indian warriors. He had assembled a fine flock of generals, colonels, and majors for the FIREFORS projects, not to mention the gaggles of civil servants and defense contractors. But they had left a few stray turkeys beyond the fence. Strays did not present an abnormality, but when they started acting like an Indian war party, he had to do something about it. Billions of dollars in FIREFORS projects could be canceled if people started concluding that the Sling Project, with a few paltry millions of dollars, could provide more capabilities at ludicrously less cost. Rumors had started already; an intense new school of treaty-loving budget-butchers waited for an opening to storm the FIREFORS train. It was exhausting to think about.