War Is Virtual Hell

The First Company of the 12th Armored Cavalry Regiment prepared for virtual battle.

At the Combined Arms and Tactical Training Center (CATTC) in Fort Knox, Ky., the troops prepared to enter SIMNET - a virtual war delivered via network links. With the almost Disney-like mimicry typical of SIMNET operations, the warriors were briefed in an actual field command-post, with folding camp-stools, fly swatters, and stenciled jerry cans. The young tankers wore green- and-brown forest camouflage fatigues, black combat boots, and forage caps.

Their command-post canvas tent was pitched inside the giant CATTC barn, right in the midst of silent rows of plastic tank simulators.

The Americans listened to a British officer on NATO exchange, Maj. Rogers, a two-year veteran of Fort Knox's simulator network. The major wore British olive- green, with rolled sleeves and gold-crowned epaulets and a Union Jack at the shoulder. He swiftly explained the tactical situation with deft scribbles on the plastic overlay covering a large topographical map.

Today's engagement would take place in a digital replica of California's Mojave Desert, the bleak, much-mangled terrain that is the heavy-armor stomping grounds of the US Army's National Training Center. Thanks to the Defense Advanced Research Projects Agency (DARPA), the Defense Mapping Agency, and the Army's Topographic Engineering Center, the US military's vast Mojave acreage had been replicated virtually. The virtual Mojave is now available for daily use even in distant Fort Knox (and in an increasing number of other simulation centers around the planet).

The NTC's Mojave was a very harsh terrain, a hell of a place to lose a cow or to throw a tank track, and today it was worse yet, because it was swarming with the Opposing Forces.

The Threat were on their way in overwhelming numbers. Their assault force was four times larger than the beleaguered Americans, and they were blitzkrieging headlong in Soviet T-72 heavy tanks and mechanized transports.

The unlucky One-Twelve Cav were to take their initial stand in the ruggedly digitized Mojave hills on a baseline code-named Purple. Their orders were to fight in their sector, delaying the advance as best they could, while retreating in good order to Baseline Amber, where the survivors (if there were any) were to take another stand.

The attacking enemy would advance from west to east. That much was already known. But the exact enemy tactics were obscured by the fog of war.

The US company commander, Capt. Van Aken, studied the terrain and deployed his meager forces with care. Alpha Platoon to guard the center. Bravo Platoon to the north. Charlie Platoon to the south. The command post to the rear, near Baseline Amber. And the scouts, in their swift but lightly armored Bradleys, to range ahead of Baseline Purple.

The One-Twelve Cav had their orders. They understood their strategy. They left the command post for the squat plastic ranks of simulators. The Jacuzzis of Death.

From the outside, a SIMNET M1 Abrams tank simulator is clearly not a tank. It looks like an oddly humped gray fiberglass shower-stall. The simulator is, in fact, built by Jacuzzi from the same materials as a whirlpool bath. Its interior, however, is designed to psycho-logically replicate the basic tank experience, and it does.

Inside, the simulator is the proper shape and size for a tank's crew chamber. It makes all the proper sounds: the loud engine whine, the ominous rumble of treads, the multi-ton coffee-grinder racket as the turret slews, the concussive thud of the main gun firing. It has the instruments of a tank, though many of those controls are nonfunctional and only painted-on. There are no actual 40- pound high-explosive shells inside simulators, though the loader, by design, must still go through the physical motions of cramming them into the cannon, with all the proper timing, proper footwork, and the proper clanks and thuds.

A real M1 Abrams battle tank is a nightmarish vehicle. It weighs 70 tons. It's 26 feet long and 12 feet wide. It carries a 120-millimeter cannon that fires rounds that travel a mile-per-second: high-explosive shells, or armor-piercing uranium slugs. The M1 tank can climb obstacles three feet high with no trouble, cross ditches eight feet wide with ease, and roar down roads at 42 mph. It is an extremely lethal and frightening machine that can kill anything it can see.

It is also a horrible place in which to die. The Abrams holds four men. Three of them (the tank commander, the gunner, and the loader) ride in the crew chamber which is about the size of a large bedroom closet. The tank commander sits on a swivel-seat with his knees at the upper back of the gunner, who is crammed into a tiny ergonomic nook. The loader heaves shells into the butt of the 120- millimeter cannon, which juts like a dinosaur's rump into the turret cavity. The fourth man, the driver, lies on his back in a padded niche much the size and shape of a coffin. He steers the tank with a pivoting pair of black rubber handles from a metal post over his belly. He is not inside the turret with the other men; instead, he is squirreled away into the bowels of the machine and communicates by headset. Like the commander and the gunner, the driver's view of the world comes through "vision blocks," three rectangular blocks each the size and shape of a rear-view mirror.

Almost every visible surface within the chamber is covered with readout screens, switches, sensors, gauges, and maintenance monitors. The area around the tank commander's tall black stool has a weirdly shaped black joystick, a targeting scope, and two flat screens with buttons bearing cryptic acronyms. These big square buttons are designed to be pressed by hands encased in chemical-warfare gauntlets. They're like a lethal parody of the child-sized buttons on a My First Sony.

Tanks are, of course, very well-armored vehicles, but there is very little on earth that can resist a 120-millimeter uranium slug traveling at a mile-per- second. Anything hit by this projectile instantly buckles and splatters. Modern tank-to-tank warfare is extremely lethal and the exchange of direct fire generally lasts only seconds.

Those seconds are precious, so time spent inside a simulator is not a picnic. Simulators are not toys. They are "fun" in some sense, but only about as much fun as an actual no-kidding tank. You can drive these simulators across cyberspace landscapes, coordinate their tactics, advance and retreat, aim their cannon, fire and be fired upon. You can smash into obstacles, bog down in mud, fall off cliff edges, and experience various kinds of simulated mechanical and engine trouble. You can panic, you can screw up, you can make a fool of yourself in front of your comrades and your commander. You can directly affect your real- life military career through what you do in simulators. And you can be killed inside simulators - virtually speaking.

The One-Twelve Cav deployed to their virtual tanks, opened the thick plastic doors on their hefty refrigerator-style hinges, took their posts at the black plastic seats, and were sealed inside. The drivers were also formally encased in their own separate plastic sarcophagi.

They started their virtual engines. They began exchanging virtual radio traffic. They examined their virtual navigation, and squinted at the desert-colored polygons in their vision-blocks. From the Ethernet lines dangling from metal frames overhead, SIMNET packets began to flow to and from the gloss-black Computer Image Generators, and the SIMNET recording angel, the big network machine they called "Radcliff," started to monitor the battle.

In another area of the simulator barn, the wily Threat commander brooded over his color Macintosh. Capt. Baker, a US Marine tactical instructor on loan to Fort Knox, was taking on the entire American force single-handedly. The Yankee opposition were sealed inside their simulators, gazing nervously at the pixelated desert and jockeying for position. But Baker could see the entire landscape at a glance. His on-screen map showed red roads, yellow badlands, the milling icons of the blue Friendlies, and the red lozenges of his own approaching Threat task force, rumbling forward west of Baseline Purple.

Capt. Baker followed Soviet tactical doctrine scrupulously. He gave his unmanned, computer-generated tanks and armored vehicles their instructions with deft points-drags-and-clicks of his Macintosh mouse.

His strategy was to spot or create a weakness in the Yankee defenses, pour as much of his armor through the chink as possible, then roll at blitzkrieg speed to a target deep behind enemy lines: "Objective Kiev."

Capt. Baker coolly sent three groups of digital scouts to certain death.

In the north, Bravo Platoon was the first to spot the approaching enemy scouts. Three Bravo tanks lurched suddenly from ambush and blasted the mechanized transports into smoking digital wreckage. The dying transports took a posthumous vengeance, though, by calling in an artillery bombardment on the Yankee position. Bravo Platoon saw red and yellow impacts spike their hillside landscape, and a vicious crump of high explosives burst from the Perceptronics audio simulators.

A second enemy probe tried the center of the American line. Alpha Platoon called in a hasty artillery strike of their own against the enemy reconnaissance. Unfortunately, the map coordinates were badly garbled in the growing excitement. Lethal "friendly fire" now whumped and blasted around Alpha Platoon's own scouts. One scout was killed by an enemy transport; the other shot dead by friendly tanks as it fled into the trigger-happy muzzles of its own backup units.

By now the radio traffic was going wild. Back at the SIMNET system operator's omniscient "Stealth Station," every howl and yelp was spooling onto a cheap K- Mart boombox for later analysis by trainers.

Under the stress of battle, the American chain of command was disintegrating, and the engagement was becoming a wild scrap.

But one Alpha tank survived. He had found a slope of ground in a sharp declivity, a sniper's paradise. Inside the simulator, the tank commander of Alpha Unit 24 began to lacerate the enemy column, rolling back behind the safety of the virtual ridge, reloading his cannon, then surging up again to swiftly nail another victim with his laser target reticule. It is a terrible thing to snipe with a laser-guided 120-mm cannon. Alpha 24 was methodically tearing the enemy column apart. Within some 30 seconds four enemy vehicles were reduced to burning hulks.

The robotic enemy column seemed stunned by Alpha 24's lethal jack-in-the-box tactic. They milled around in confusion, unable to get a clear shot. Then the American artillery kicked in, bracketing the column in lethal fire. With their position absolutely untenable, the column charged the sniping tank. Alpha 24 killed two more tanks before being outflanked and forced to retreat.

Bravo Platoon was standing firm in the north, but it had been outfoxed. No one was coming their way. Instead, two more enemy columns suddenly appeared in the far south, in Charlie Platoon's turf. Seen through the Threat commander's Macintosh map, the jittering red icons resembled angry ants.

Charlie Platoon as a whole was caught unawares. Despite their wire-guided TOW missiles, Charlie Platoon's Bradley Fighting Vehicles were no match for the Threat heavy armor. Charlie Platoon was swiftly overwhelmed, howling through the radio network for backup that was too slow, and too far away.

Charlie Platoon's survivors called in air-support as they struggled to reach the relative safety of Baseline Amber. In answer, two automatic Apache attack helicopters emerged from the blue nothingness of SIMNET's cyberspace sky. They fired air-to-surface missiles and swiftly roasted a pair of enemy tanks; but the other T-72 tanks potted both the choppers on the wing. The Apaches fell in crumpled digital heaps of flaming polygons.

As the engagement proceeded, dead men began to show up in the CATTC video classroom. Inside the simulators, their vision blocks had gone suddenly blank with the onset of virtual death. Here in CATTC's virtual Valhalla, however, a large Electrohome video display unit showed a comprehensive overhead map of the entire battlefield. Group by group, the dead tank crews filed into the classroom and gazed upon the battlefield from a heavenly perspective.

Slouching in their seats and perching their forage caps on their knees, they began to talk. They weren't talking about pixels, polygons, baud-rates, Ethernet lines, or network architecture. If they'd felt any gosh-wow respect for these high-tech aspects of their experience, those perceptions had clearly vanished early on. They were talking exclusively about fields of fire, and fall-back positions, and radio traffic and indirect artillery strikes. They weren't discussing "virtual reality" or anything akin to it. These soldiers were talking war.

"Get them, sir," a deceased tanker muttered vengefully as he watched Alpha 24's heroic stand in the fake Mojave Hills. Another tanker, from the Alpha scout unit, griped bitterly about his death by friendly fire: "fratricide." Dying at the hands of his own platoon had been especially cruel. It was clear that the real-life lesson of unit coordination had sunk in well - at least for this poor guy.

"It's only SIMNET," another tanker told him at last. "You're not bleeding."

They weren't bleeding. That was undeniable. On the contrary, they'd just been killed in combat, yet also had the amazing luxury to learn by this experience. The CATTC trainers called off the battle in time for lunch; the result was now a foregone conclusion. As Capt. Baker explained to his virtual enemies and real- life students, "There'll be hot borscht and vodka at Objective Kiev tonight." The dead soldiers, and the few pleased survivors, had shakes, fries and burgers from the local Burger King.

When they returned from their lunch, Maj. Rogers replayed the battle for them, hitting the high points with detailed graphics from the big machine called Radcliff. Any event can be scrutinized, from any angle of vision, at any moment in time that the trainers desired.

Virtual Reality as a Strategic Asset

SIMNET today is a clunky and rapidly aging mid-1980s technology; its giant, $100,000 image generators are so large that they bear red adhesive labels: "WARNING: RISK OF PERSONAL INJURY FROM RACK TIPPING FORWARD." SIMNET still thrives in everyday use at Fort Knox, Fort Rucker, Fort Denning and a number of other sites, sometimes linked together through long-distance lines, more often not. But better stuff is coming: faster, cheaper, more sophisticated, and far better-connected.

The people at the Institute for Defense Analyses know all this. The Institute is a large, brown, campus-like building set in a pleasant wooded lot outside the Beltway of Washington, D.C. Its tall brick walls are festooned with white security telecameras. White shuttle-vans with the IDA logo - an infinity-sign in a triangle with the IDA acronym - pull up periodically, disgorging small scholarly groups of tweed-jacketed military-academic spooks.

I visited the Institute last fall. Groups of Air Force bluesuiters ambled periodically into the "Stealth Room." "Stealth technology" cloaks observers in digital invisibility, so that they can travel to any point inside a simulated battle. A huge triptych of full-color computer screens showed the simulated activity of a certain weapons system I was forbidden to identify publicly. The tarpaulin-shrouded chambers within the Institute were draped with wrist-thick clusters of black cabling leading to Sun workstations, networked Macintoshes, and a variety of prototype simulators. Everything hummed.

Col. Jack A. Thorpe, USAF, Ph.D., spends a lot of time in the Institute. Col. Thorpe is the "Father of Distributed Simulation" and is America's foremost advocate of virtual reality as a strategic asset.

The Colonel wore a civilian pinstriped suit, an understated maroon tie, and polished black wingtips. Col. (or Doctor) Thorpe is a cognitive psychologist specializing in training techniques; he is tall and lean and bespectacled, with a straight nose, dark hair, and hollow temples, and he possesses the vigorous air of a man with a vision and clear ideas of how to get there. He is somewhere near his early forties.

Col. Thorpe's highly unusual expertise makes his position in the military hierarchy somewhat anomalous. He is a career Air Force officer who nevertheless pioneered virtual reality networks for the US Army. He is also the special assistant for simulations at DARPA. He clearly has a lot of pull at the Institute for Defense Analyses, his institutional home away from home, where DARPA sponsors the IDA Advanced Distributed Simulation Laboratory.

Col. Thorpe also has a number of friends among the computer-networking experts at the University of Central Florida in Orlando, and more colleagues yet at the Defense Mapping Agency, and yet more in the Topographic Engineering Center, and plenty of eager listeners from all over the defense-contracting industry.

And yet Col. Thorpe's primary role in today's USmilitary is as "Leader of Thrust Six" for the Director of Defense Research and Engineering.

Dr. Victor Reis, Col. Thorpe's immediate superior, is Director of Defense Research and Engineering. Dr. Reis has a seven-point plan for distributing $3 billion worth of defense research in fiscal year 1993. The plan involves fairly standard post-Cold War matters such as global surveillance, air superiority, precision strikes, and advanced land combat. But the sixth point in the Reis plan is "Synthetic Environments."

Col. Thorpe is the premier Defense Department evangelist for synthetic environments. His interest in these matters goes back to the late '70s, when he was in the Air Force Office of Scientific Research. In those days, full-scale Air Force jet simulators cost $40 million each. The simulators - odd devices that perch on hydraulic stilts and pitch and toss their wannabe-ace occupants like broncos - clearly worked well in flight-training, but they were clumsy and they cost far too much, and worst of all, they were not connected.

Col. Thorpe is a connectivity visionary first and foremost.

His reasoning is simple but profound. An army is not an armed mob of heroic individualists. An army is a connected, coordinated, disciplined killing force, working systematically in close cooperation to a desired end. In any stand-up fight, an army will destroy a mob, even an armed and heroic mob, with very little trouble.

There are two basic problems with isolated simulators. They don't connect to other soldiers, and they don't connect to an enemy. They might train individual pilots how to fly very well, but they can't train squadrons how to fight. They can teach the skill of handling an aircraft, but they can't teach combat with your own comrades at hand, against an intelligent enemy who can see you and react to what you do. Similarly, a single tank simulator might train a single crew to some brilliant pitch of mechanical efficiency, but it can't build platoons, companies, battalions, or regiments of armor that can work together, confront enemies, and conquer the battlefield. Armies win wars, not lone heroes. In real wars, Rambos die quick.

On a higher level of organization, the same logic of coordination and networking applies across the individual armed forces. Single branches of the American military establishment can no longer play the lone-wolf game. Interservice rivalry (though still very real) is officially out of fashion in the post-Cold War world of rapid deployment. Maximum speed, maximum impact, and minimum American casualties all demand that the services be fully coordinated, that all assets be brought into play in a smooth and utterly crushing synchrony. Navy ships support land offensives, Air Force strikes support mud-slogging Marines. And space-based satellite intelligence, satellite communications, and satellite navigation support everybody.

That is the core of modern American strategic military doctrine, and that is what Col. Thorpe's new project, the Distributed Simulation Internet, is meant to accomplish for the military in the realm of cyberspace.

DARPA is an old hand at computer networking. The original ARPANET of 1969 grew up to become today's globe-spanning civilian-based Internet. SIMNET was another DARPA war-child, conceived in 1983 and first online in May 1986. DARPA invented SIMNET just as it invented the Internet, but DARPA spun SIMNET off to the US Army for day-to-day operations.

DARPA, by its nature, sponsors the cutting edge; the bleeding edge. The Distributed Simulation Internet, projected for the turn of the century, is to be a creature of another order entirely from SIMNET. Ten thousand linked simulators! Entire literal armies online. Global, real-time, broadband, fiber- optic, satellite-assisted, military simulation networking. Complete coordination, using one common network protocol, across all the armed services. Tank crews will see virtual air support flitting by. Jet jockeys will watch Marines defend perimeters on the pixelated landscape far below. Navy destroyers will steam offshore readying virtual cruise missiles... and the omniscient eye of trainers will watch it all.

And not just connected, not just simulated. Seamless. "Seamless simulation" is probably the weirdest conceptual notion in the arsenal of military virtuality. The seams between reality and virtuality will be repeatedly and deliberately blurred. Ontology be damned - this is war!

Col. Thorpe emphasizes this concept heavily. And seamless simulation is not a blue-sky notion. It's clearly within reach.

Most of the means of human perception in modern vehicles of war are already electronically mediated. In Desert Storm, both air pilots and tank crews spent much of their time in combat watching infrared targeting scopes. Much the same goes for Patriot missile crews, Aegis cruisers, AWACS radar personnel, and so on. War has become a phenomenon that America witnesses through screens.

And it is a simple matter to wire those screens to present any image desired. Real tanks can engage simulator crews on real terrain which is also simultaneously virtual. Fake threats can show up on real radar screens, and real threats on fake screens. While the crews in real machines can no longer tell live from Memorex, the simulators themselves will move closer to the "scratch and sniff" level of realism.

Granted, simulators still won't fire real shells. "They know how to load shells," Col. Thorpe points out. "That's not what we're trying to teach them." What he's trying to teach them, in a word, is networking. The wired Army, the wired Navy, the wired Air Force and wired Marines. Wired satellites. Wired simulators. All coordinated. All teaching tactical teamwork.

A wired Armed Forces will be composed entirely of veterans - highly trained veterans of military cyberspace. An army of high-tech masters who may never have fired a real shot in real anger, but have nevertheless rampaged across entire virtual continents, crushing all resistance with fluid teamwork and utterly focused, karate-like strikes. This is the concept of virtual reality as a strategic asset. It's the reasoning behind SIMNET, the "Mother of All Computer Games." It's modern Nintendo training for modern Nintendo war.

The War We Won

The walls inside the Institute for Defense Analyses are hung with Kuwaiti topography. In some entirely virtual, yet final and terrible sense, the USmilitary now owns Kuwait. The Pentagon has a virtual Kuwait on a hard disk - SAKI, the Saudi Arabia-Kuwait-Iraq database. It has the country mapped meter by meter, pixel by pixel, in 3-D, with weather optional. You can climb into one of Col. Thorpe's tank simulators and you can drive across that cyberspace doppleganger voodoo Kuwait exchanging gunfire with the polygonal ghosts of Iraqi T-72 tanks.

There was a war in Kuwait recently. They don't call it "Desert Shield-Desert Storm" at IDA or DARPA. They certainly don't call it the "Persian Gulf War" - that would only irritate the Arab coalition allies who insist on calling that tormented body of water the "Arabian Gulf." No - they like to call this event "the war in Southwest Asia."

The US military hasn't forgotten Southeast Asia. To hear them talk, you would think that they had discussed very little else for the 16 long years between Saigon and Kuwait City. In Southeast Asia the Pentagon sent Americans into tunnels below the earth to fight peasant guerrillas hand-to-hand with knives and pistols. They sent soldiers sweeping through rice paddies in hopes of attracting gunfire from some Viet Cong group large enough to be spotted from helicopters. As the situation became more hopeless, they sent in more American flesh to be ambushed and pierced with punji sticks. The United States lost a major war in Southeast Asia.

However, the US recently won a major war in Southwest Asia. With some handy but basically political and cosmetic help from its Coalition allies, the US destroyed the fourth-largest land army on the planet in four days at a cost of only 148 American dead. Geopolitically, this war may have been less significant than Vietnam (with almost everybody in the civilized world versus a clear megalomaniac, victory of some sort was probably not much in doubt.) Strategically and tactically however, Desert Storm was one of the most lopsided and significant military victories since Agincourt. And the American military is quite aware of this.

"Southwest Asia" may have vanished into the blipverse of cable television for much of the American populace, but the US military has a very long institutional memory. They will not forget Southwest Asia, and all the tasty things that Southwest Asia implies, for a long time to come.

Col. Thorpe and his colleagues at DARPA, IDA, and the Army Office of Military History have created a special Southwest Asian memento of their very own - with the able help of their standard cyberspace civilian contractors: Bolt Beranek & Newman and Illusion Engineering. The memento is called "The Reconstruction of the Battle of 73 Easting."

This battle took place at a map line called 73 Easting in the desert of southern Iraq. On 26 February 1991, the Eagle, Ghost, and Iron Troops of the US 2nd Armored Cavalry Regiment attacked the Tawakalna Division of the Iraqi Republican Guard. These were untested UStank troops, without any previous combat experience, blundering forward in a sandstorm to confront entrenched Soviet-made heavy tanks manned by elite veterans of an eight-year war. Thanks to the sandstorm, the Americans had no air support either; this was a straight-on tank- versus-tank scrap in the desert, right out of the Rommel and Patton strategic notebook.

The Americans annihilated the Iraqis in 22 minutes.

The Battle of 73 Easting has become the single most accurately recorded combat engagement in human history. Army historians and simulation modelers thoroughly interviewed the American participants, and paced the battlefield meter by meter. They came up with a fully interactive, network-capable digital replica of the events at 73 Easting, right down to the last TOW missile and .50-caliber pockmark. Military historians and armchair strategists can now fly over the virtual battlefield in the "stealth vehicle," the so-called "SIMNET flying carpet," viewing the 3-D virtual landscape from any angle during any moment of the battle. They can even change the parameters - give the Iraqis infrared targeting scopes, for instance, which they lacked at the time, and which made them sitting ducks for high-tech American M1s charging out of blowing sand. The whole triumphal blitzkrieg can be pondered over repeatedly (gloated over even), in perfect scratch-free digital fidelity. It's the spirit of Southwest Asia in a digital nutshell. In terms of American military morale, it's like a '90s CD remix of some '60s oldie, rescued from warping vinyl and remade closer to the heart's desire.

Col. Thorpe and his colleagues first demo'd "73 Easting" in late 1991 at the Interservice/Industry Training Systems and Education Conference (I/ITSEC) #13, the premier convention for the military training, simulation, and VR industry. The virtual battle was the hit of the show, and it went on to tour the Senate Armed Services Committee, where it much impressed Sam Nunn and John Glenn.

"The Reconstruction of the Battle of 73 Easting" is an enormously interesting interactive multimedia creation. It is fast and exhilarating and full of weird beauty. But even its sleek, polygonal, bloodless virtuality is a terrifying thing to witness and to comprehend. It is intense and horrific violence at headlong speed, a savage event of grotesque explosive precision and terrible mechanized impacts. The flesh of real young men was there inside those flam- ing tank-shaped polygons, and that flesh was burning.

That is what one knows - but it's not what one sees. What one really sees in "73 Easting" is something new and very strange: a complete and utter triumph of chilling, analytic, cybernetic rationality over chaotic, real-life, human desperation.

Battles have always been unspeakable events, unknowable and mystical. Besides the names of the dead, what we get from past historical battles are confused anecdotes, maybe a snapshot or two, impressions pulled from a deadly maelstrom that by its very nature could not be documented accurately. But DARPA's "Battle of 73 Easting" shows that day is past indeed. The omniscient eye of computer surveillance can now dwell on the extremes of battle like a CAT scan detailing a tumor in a human skull. This is virtual reality as a new way of knowledge: a new and terrible kind of transcendent military power.

A Virtual Military/Industrial Complex?

What is it that Col. Thorpe and his colleagues really want? Well, of course, they want the unquestioned global military pre-eminence of the American superpower. Of course, they want to fulfill their patriotic duty in the service of the United States and its national interests. They want to win honor and glory in the defense of the American republic. Those are givens. Col. Thorpe and his colleagues already work to those ends every day.

What they really want is their own industrial base.

They want the deliberate extension of the American military-industrial complex into the virtual world. They want a wired, digitized, military-post-industrial complex, reformed and recreated to suit their own terms and their own institutional interests.

They want a pool of contractors and a hefty cadre of trained civilian talent that they can draw from at need. They want professional Simulation Battle Masters. Simulation system operators. Simulation site managers. Logisticians. Software maintenance people. Digital cartographers. CAD-CAM designers. Graphic designers.

And it wouldn't break their hearts if the American entertainment industry picked up on their interactive simulation network technology, or if some smart civilian started adapting these open-architecture, virtual-reality network protocols that the military just developed. The cable TV industry, say. Or telephone companies running Distributed Simulation on fiber-to-the-curb. Or maybe some far-sighted commercial computer-networking service. It's what the military likes to call the "purple dragon" angle. Distributed Simulation technology doesn't have to stop at tanks and aircraft, you see. Why not simulate something swell and nifty for civilian Joe and Jane Sixpack and the kids? Why not purple dragons?

We're talking serious bucks here. It's not the most serious money in a superpower's massive military budget, granted - at least not yet, it isn't - but it's very damned serious money by the standards of your average Silicon Valley multimedia start-up. The defense simulation market is about $2.5 billion a year. That's Hollywood-serious and then some. Over the next 10 years the Pentagon plans to drop about $370 billion on electronics R&D. Some of that money will fall to simulation. Maybe a lot of it, if the field really takes off.

There are some very heavy operators in the simulation market - and they were all at the 14th I/ITSEC in San Antonio, Texas last November.

The gig was sponsored by the National Security Industrial Association - a group that basically is the military-industrial complex. I/ITSEC was graced by the corporate presence of General Electric, General Dynamics, McDonnell Douglas, Rockwell, Hughes, Martin Marietta, and Bolt Beranek & Newman. And yes, they were also favored by IBM, Lockheed, Motorola, Silicon Graphics, Loral, Grumman, and Evans & Sutherland. And plenty more: a whole cloud of hangers-on, suppliers, dealers, niche marketeers, and brand-new startups.

All these nice-suited people were in handsome display booths in a very large carpeted hall within hollering distance of the Alamo. The place was alive with screens, top-heavy with humming megabytage. General Dynamics ran their new tank simulator live, right on the display floor. Bolt Beranek & Newman ran a hot new image generator that made mid-1980s SIMNET graphics look like Hanna-Barbera.

They were running demos at every side, and handing out promotional videos, and glossy display brochures, and every species of carnivorous mega-corporate public relations. They boasted of clinching major sales in foreign markets, and of their glowing write-ups in specialized industry journals such as Military Simulation & Training ($73/year, Britain) and Defense Electronics ($39/year, Englewood, Colo.) and National Defense (American Defense Preparedness Assn., $35/year, Arlington, Va.). Strange magazines, these. Very strange.

The attendees attended the keynote speeches, and the banquet speeches, and the luncheon speeches. And they attended the presentations, and the paper sessions, and the six tracks of formal programming. And they industriously leafed through their blockbuster, 950-page I/ITSEC #14 Proceedings. This enormous red-and-white volume, officially "approved for public release" by the Department of Defense, was crammed-to-bursting with scholarly articles such as "Computer-Supported Embedded Training Systems for the Strike/Fighter Aircraft of Tomorrow," and "Hypermedia: a Solution for Selected Training and Prototyping Applications."

And even "Virtual Training Devices: Illusion or Reality?" Not much debate there. Simulators are, of course, both illusion and reality. They're not entirely real, but they function just fine. And they pay like gangbusters.

These people weren't there for their health. They were there for a simple, basic reason. Call it cyberpork. Cyberpork put the slash in "Interservice/Industry." It put that handy hyphen into "military-industrial." Industry wasn't lonely at I/ITSEC. Their patrons were there in spades. Military brass - heavy brass, shiny brass. TRADOC, the Training and Doctrine Command. STRICOM, the Simulation Training and Instrumentation Command. Air Force Training Command. Naval Training Systems Center. Naval Air Systems Command. People in crisp uniforms and polished shoes, from weapons divisions, and materiel commands, and program offices, and from forts and bases and academies and institutes, all across the US.

Suppose that you were an ambitious and visionary leader of the post-Cold War '90s military establishment, like, say, Col. Jack Thorpe. Or perhaps Col. Ed Fitzsimmons of the Defense Modeling and Simulation Office, or Lt. Col. James Shiflett from the Information Science and Technology Office, or Col. William Hubbard from Army Battle Labs. What are you supposed to do with all these people at I/ITSEC? On the face of it, your situation doesn't look all that promising. The 40-year Cold War military-industrial gravy train has clearly gone off the rails. There's gonna be - there's bound to be - some "downsizing" and "restructuring" and "conversion" and "transition," and all those other euphemisms for extreme and wrenching economic pain to your own suppliers, and your own people, and your own colleagues. Not to mention the potential threat to your own career.

Your answer, of course - you being the kind of guy you are - is to seize this magnificent opportunity. Wire everyone up! Global, real-time, broadband, networked vendors and suppliers! They're hurting now. They're worried. They'll go for anything that looks like survival, that looks like a hot new market. Seize the day. No more of this time-wasting, money-squandering, inter-vendor rivalry with their incompatible standards. One standard now. The Distributed Simulation Internet Standard.

The Distributed Simulation Internet doesn't even exist yet. It may never exist. That's not a problem. What it does have is its own protocol. The DSI Protocol will link simulation machines from manufacturers across the field and across the planet.

This virtuality standard emerged from Orlando, Fla., in the early '90s, from the potent nexus of Orlando's Institute for Simulation & Training, Orlando's University of Central Florida, Orlando's US Army STRICOM, Orlando's Naval Training Systems Center, and the Orlando-based, 400-strong Standards for the Interoperability of Defense Simulations working groups. (One mustn't rule out the possible cultural influence of Disneyworld, either.)

They demo'd the new standard on a network link-up at I/ITSEC #14, live. They went for the opportunity. They had to rip up some of the Ethernet wiring that they'd laid before the show, because it had so many crimp-failures from the tramping legions of wingtip-shod vendor feet. It got hairy for a while there. But they got the demo to run.

Of course a system crashed. Somebody's system always crashes at any multimedia demo. It's like a force of nature. In the case of the DIS Interoperability Demo, it was the Mac Quadra 900 running the slide show. The sucker iced when its screensaver kicked in, and the sweaty-palmed techies from IDA had to re-boot live. They winged it, and got the slides up. It looked okay. Most people didn't notice.

The protocol worked just fine. They had a big digitized section of the terrain from Fort Hunter-Liggett in California, running live on-screen, cunningly combined with an actual long-distance link to an actual wired tank in actual Fort Hunter-Liggett.

"Seamless simulation," live onstage.

The demo was far from real virtual war. There was some ritual gunfire here or there, but this wasn't real combat training. This was a fashion show in seam- free camouflage haute-couture.

Everybody took a formal runway-model turn, up on the big virtual stage. With live narration at the mike: "The bogeys are generated by Bolt Beranek & Newman." General Dynamics Land Systems Division modeled the virtual M1A2 Battle Tank. From their own show-booth, General Electric thoughtfully supplied an Abrams tank and an F-16. Hughes proudly displayed a robot spy-drone. McDonnell Douglas had a surface-to-air missile, and Lockheed demo'd a virtual Patriot battery. Twenty- four companies - twenty-five, if you count the guys who supplied the video projectors. All of them packed snugly in the DARPA virtual corral.

They had the brass lined-up right at the front, in a row of folding chairs. A rear admiral here, a couple of lieutenant generals there; a full brace of Cold War veterans, braid and chest ribbons and hats. The brass watched the three monster screens with squint-eyed, show-me skepticism.

And the brass weren't blown-away, either. The network looked pretty good, and it ran without crashing, but they weren't stunned or amazed. The brass didn't leave San Antonio raving that they'd just seen the future and it worked. They clearly didn't know quite what to make of what they had just seen. One got the impression that they figured this virtual-network stuff might turn into something useful someday. Cute gimmick. Clever. Worth a look, I guess. Learn something new every day. Glad we came down here to I/ITSEC. Lemme know when we can use this to invade Normandy.

The brass were on public exhibit themselves, actually. Whether they knew it or not, they were legitimizers, stalking horses, Trojan Horses. Generals and admirals from a very long-lasting but swiftly vanished era. Compared to their tech-crazed subordinates - the Southwest Asian, baby-boomer, carnivorous cyber- colonels, majors, and captains who are now actually running the digitized New World Order American military - the Cold War guys looked like a line of stuffed ducks.

Today Kuwait, Tommorow the World

There was some interesting stuff backstage at I/ITSEC. There was a big rope- handled canvas bag full of the tools of the virtual trade: hex crimpers, nut drivers, metric wrenches, soldering wire, cable strippers. There were big ugly powerful rock'n'roll amps stenciled PROPERTY OF US GOVT INSTITUTE FOR DEFENSE ANALYSES, and big color display monitors shimmed up on cardboard, and there were powerstrips and orange extension cords and some loose Mac floppies. And there was a handscrawled brag on a backstage chalkboard, written by the techies from Orlando: "DIS Interoperability Demonstration. Today's feature: DIS. Tomorrow: the holodeck!"

The natural question arises: Is this some kind of wacky egghead DARPA media hype, or is this a genuine military technology? Can governments really exercise national military power - kick ass, kill people - merely by using some big amps and some color monitors and some keyboards, and a bunch of other namby-pamby sci-fi "holodeck" stuff?

The answer is yes.

Yes, this technology is lethal. Yes, it is a real strategic asset. Military virtual reality is not a toy or a joke. There is a lot of vaporware in "virtual reality," but this technology definitely will help people kill each other. Virtual reality happens to be very fashionable at the moment, with some ritzy pop-cultural overtones, but that is accidental. Whether or not VR becomes a major new medium of commercial entertainment, or some vital new mode of artistic expression, it still will be of enormous use to the military. Thriving civilian VR will probably make military VR expand even faster; giving the virtual battlefield better and glossier set designs.

There was a demo at I/ITSEC called "Project 2851." This is a new standard for digital terrains, a standard for all American armed forces. It will let them share terrain databases on any number of different machines.

But there is another aspect to Project 2851. Project 2851 is about the virtual reproduction and archiving of the entire planet. Simulator technology has reached a point today in which satellite photographs can be transformed automatically into 3-D virtual landscapes. These landscapes can be stored in databases, then used as highly accurate training grounds for tanks, aircraft, helicopters, SEALS, Delta Force commandos.

What does this mean? It means that soon there will be no such thing as "unknown territory" for the United States military. In the future - soon, very soon - the United States military will know the entire planet just like the back of its hand. It will know other countries better than those countries know themselves.

During the Battle of 73 Easting, an American tank regiment came roaring out of an Iraqi desert that the Iraqis themselves could not navigate. The Iraqis couldn't enter their own desert, because they would have died there. But the Americans had satellite navigation units, so the Americans knew where they were on our planet's surface right down to the yard.

The Stealth pilots who blew downtown Baghdad into hell-and-gone had already flown those urban landscapes before they ever put their butts in the cockpit seat. They knew every ridge, every skyline, every road - they'd already seen them on console screens.

During Desert Storm, some Iraqi soldiers actually surrendered to unmanned flying drones. These aircraft are disembodied eyes, disembodied screens, network peripherals basically, with a man behind them somewhere many miles away. And that man has another screen in front of him, and a keyboard at hand, and a wire from that keyboard that can snake through a network and open a Vent of Hell.

This is what it all means. Say you are in an army attempting to resist the United States. You have big tanks around you, and ferocious artillery, and a gun in your hands. And you are on the march.

Then high-explosive metal begins to rain upon you from a clear sky. Everything around you that emits heat, everything around you with an engine in it, begins to spontaneously and violently explode. You do not see the eyes that see you. You cannot know where the explosives are coming from: sky-colored Stealths invisible to radar, offshore naval batteries miles away, whip-fast and whip- smart subsonic cruise missiles, or rapid-fire rocket batteries on low-flying attack helicopters just below your horizon. It doesn't matter which of these weapons is destroying your army - you don't know, and you won't be told, either. You will just watch your army explode.

Eventually, it will dawn on you that the only reason you, yourself, are still alive, still standing there unpierced and unlacerated, is because you are being deliberately spared. That is when you will decide to surrender. And you will surrender. After you give up, you might come within actual physical sight of an American soldier.

Eventually you will be allowed to go home. To your home town. Where the ligaments of your nation's infrastructure have been severed with terrible precision. You will have no bridges, no telephones, no power plants, no street lights, no traffic lights, no working runways, no computer networks, and no defense ministry, of course. You have aroused the wrath of the United States. You will be taking ferries in the dark for a long time.

This is not the future that I'm describing. Basically, this is the present - this is what actually happened to the world's fourth largest army, in Southwest Asia. Will the US Government continue to expand the course that led us in that direction? After all, we've won the Cold War and our domestic economy's hurting rather badly. Will the new Clinton Administration follow the DARPA lead? Continue pouring money into the gold-plated rathole of ultra-high-tech military- technological advance?

You might judge the likelihood of that by Bill Clinton's statements on the campaign trail. "While we will need a smaller military in the post-Cold War world, we must retain our superior technology, high-quality personnel, and strong industrial base." That's what he told National Defense magazine, anyhow.

Clinton and Gore may have little reason for fondness for the Army that brought us Vietnam, but they've got plenty in common with their generational contemporaries, the cybercolonels. They are calling for a "civilian DARPA," but you can bet good money that they won't lose their fondness for the military one. Defense Simulation Internet? The White House is now in the hands of rabid fiber- optic enthusiasts.

The virtual iron is hot. Want to see a real vision of the virtual future? It's a future in which large sections of the American military-industrial complex have migrated entirely into cyberspace. This is the real DARPA Virtual Reality Vision Thing, the plans they allude to with quiet determination just after the big multimedia displays. "Simulate before you build." They want to make that a basic military principle.

Not just simulated weapons. Entire simulated defense plants. Factories that exist only in digital form, designed and prepared to build weapons that don't even exist yet either, and have never existed, and may become obsolete and be replaced by better ones, before a nail is ever hammered. Nevertheless, these nonexistent weapons will have entire battalions of real people who are expert in their use, people who helped design them and improve them hands-on, in the fields of virtual war.

"Simulate before you build" is a daring ax-stroke at the very tap-root of the Cold War-era military-industrial complex. It is a potential coup that could deliver the whole multi-billion-dollar shebang - lock, stock, and barrel - into the hands of the virtuality elite. If it shrinks the military by 50 percent or so, so what? Instead of the 1 percent or so of the Pentagon budget that they currently control, the simulation cybercolonels will own everything, the whole untidy, hopelessly bureaucratic, crying-for-improvement mess. No military object will see physical existence until it is proven, under their own institutional aegis, on the battlefields of cyberspace. They'll be able to shove the ungainly Cold War camel through the cold glass eye of the cyberspace needle. And God only knows what kind of sleek, morphing beast will emerge from the other side.

Does this sound farfetched? Why? If something as delicate and precise as virtual surgery is possible (and it is), then why not virtual military manufacturing? Sure might solve a lot of pollution problems. And military storage problems. All kinds of problems, when you come to think about it.

Let's have a speculative look at the 21st-century USA. Amber waves of grain and all that. Peaceful place; scarcely resembles a military superpower at all. Hardly any missile silos, hardly any tanks, hardly any concertina wire. Until the Americans need it. Then the whole massive, lethal superpower infrastructure comes unfolding out of 21st-century cyberspace like some impossible fluid origami trick. The Reserve guys from the bowling leagues suddenly reveal themselves to be digitally assisted Top Gun veterans from a hundred weekend cyberspace campaigns. And they go to some godforsaken place that doesn't possess Virtual Reality As A Strategic Asset, and they bracket that army in their rangefinder screens, and then they cut it off, and then they kill it. Blood and burning flesh splashes the far side of the glass. But it can't get through the screen.

Maybe you can believe that idea and all that it implies - "simulate before you build." Or maybe you might wax a little more cynical. Maybe what we're presented here, under the slick rhetoric of the Paperless Office, is yet another staggering stack of old-fashioned Pentagon paperwork - a brand new way to make megabuck hammers and toilet seats to an entire new set of ridiculous, endless bureaucratic specs. Only this time, after all the studies and form-filling, you end up with absolutely no tangible product at all!

Maybe it's just a bizarre Silicon Valley power-play. Every other major American industry has got a sucker deep in the military-industrial juice. Maybe it's time for the virtual reality, CAD-CAM, multimedia crowd to hunker down with the older industries and have some long, life-giving sips from the taxpayer's bloodstream. Maybe the whole scheme is just updated hype - for that same old fat-cat, imperialistic, hypertrophied, overfed, gold-plated military bureaucracy... .

Could be. It could go either way, maybe both ways at once - make your own decision. One thing's for sure though. The US military today is the most potent and lethal gold-plated military bureaucracy of all time.

You can't fault DARPA for lack of vision. Vision they've definitely got. There's one matter, though, which they don't discuss much. That's the possibility of a virtuality arms race.

If military virtuality really works, everyone's gonna want it.

Now imagine two armies, two strategically assisted, cyberspace-trained, post- industrial, panoptic ninja armies, going head-to-head. What on earth would that look like? A "conventional" war, a "non-nuclear" war, but a true War in the Age of Intelligent Machines, analyzed by nanoseconds to the last square micron.

Who would survive? And what would be left of them?