IN A BIZARRE AND PATHETIC RITUAL, thousands of North Koreans are herded into public squares and told to stare into the starry skies. Suddenly they are told that their country's glorious space satellite, Kwangmyong-song-i, is passing overhead. They all cheer. Like the citizens, according to a typical Pyongyang radio broadcast, "the world is watching our artificial satellite, the national strength of self-reliant Korea."

But in this case, North Korea's dictator/emperor, Kim Chong II, is wearing no clothes whatsoever. And no one in his famine-ravaged country dares point it out--no satellite ever made it up there.

The charade of a successful space achievement began with the satellite launch attempt on 31 August 1998 [Fig. 1]. Not that the event went unnoticed. The North Korean satellite rode on a rocket that was all too real. In fact, the make-believe and disinformation surrounding its demise are reflected in the confused international debate on what Third World states such as North Korea intend to do with the impressive new rockets they are building.

For many Third World nations, the powerful appeal of rockets lies in their combination of three impressive functions. First is the obvious military value of dropping war- heads on enemies hundreds of kilometers away, or threatening to do so. Second, military rockets can be easily adapted toward space applications, both for launching satellites and for threatening attacks on other nations' satellites. Third, having rockets is widely considered by both political leaders and the general public to be the admission ticket to world-power status.

The tyros of the missile club understand the triple threat of ballistic missiles. No wonder their club is growing.

The threat becomes real

The heartlands of other states have been under threat for years now. Their protection, unlike the vast moats of the United States, mostly has been changes in first-use policy, or new plans for hair-trigger military responses. The only active defense systems besides Russia's antiquated nuclear-armed system circling Moscow are still in planning, and by and large are courtesy of the United States Ballistic Missile Defense Organization. The projects include ground-and sea-based missile interceptors and an airborne laser (the latter predominantly for the North Korean threat) as well as two joint U.S.-Israeli projects: one smaller-scale ground-based missile interceptor, soon to be fielded, and a missile interceptor launched from patrolling unmanned air vehicles, a project still being defined.

As for the missile threat to the U.S. nation, the North Korean "satellite" launch, using the Taepo Dong 1 missile, caught the U.S. spies flatfooted. But now the agency, like others in the U.S. defense establishment, is publicly telling the Congress to up the stakes in this dangerous game. One startling card recently dealt to Congress was President Clinton's January 1999 request for an additional $6.6 billion to funding for development of national missile defense (that is, interceptors for missiles aimed at the continental United States), for a total of $10.5 billion through fiscal year 2005.

No decision to deploy NMD will be made before 2000, by which time the debate over the status of the Anti-Ballistic Missile Treaty and Start II (the Strategic Arms Limitation Act) may have reached feverish levels. Testifying on 2 February before Congress in support of the request, George Tenet, CIA director, could "hardly overstate [his] concern about North Korea," which he said will soon be able to deliver weapons payloads on the United States. (The country next in line to develop intercontinental ballistic missiles, according to Tenet, is Iran.)

Speaking at the Center for Strategic and International Studies in Washington, D.C., last 8 December, Robert Walpole, the national intelligence officer for strategic and nuclear programs at the CIA, explained why the launch of the North Korean missile in August--and other missile launchings last year in Iran and Pakistan--proves that the missile threat to the United States is "real, serious, and growing."

Referring to military missiles with ranges of over 1000 km, Walpole asserted that "theater-range missiles already in hostile hands pose an immediate and increasing threat to U.S. interests, military forces, and allies."

Walpole's remarks and the new CIA public statements can be considered a direct result of one of the most widely circulated reports on missile proliferation ever released. Chartered by the United States Congress, a private commission headed by Donald Rumsfeld, secretary of defense under President Ronald Reagan, released the report in mid-1998, stressing not only the extent of missile development in Third World countries, but also how wrong U.S. intelligence had been in estimating the speed with which missile technology was circling the globe.

Before the Rumsfeld Report, U.S. defense government analysts had made a number of limiting assumptions, like supposing that each nation's missile program would be indigenous -- run by home-grown scientists and technology. But as one missile analyst caustically pointed out to Spectrum, "There hasn't been an indigenous missile program on earth since about 1935," when Robert Goddard built the first liquid-fueled rockets.

Since then, technology has easily been siphoned off from country to country, either through espionage, emigration of key personnel, publication of technical data, or other common channels [Fig. 2]. Simply put, it is not all that hard anymore to build or buy a missile [see table on pp. 30-31].

The off-balance of terror David Wright, a missile technology analyst for the Defense and Arms Control Study Program at the Massachusetts Institute of Technology (MIT), in Cambridge, has concerns about the scope of the report. "The Rumsfeld Report's mandate was too narrow," he told Spectrum. "When one considers the range of threats and all the credible delivery means, the ballistic missile threat was hyped all out of proportion. There were far more likely delivery means not examined."

Theodore Postel, also an MIT missile technology analyst, agreed: "People have exaggerated the extent of the damage ballistic missiles can do. One might compare the dangers to those we already face, such as a military jet."

The argument that there are better alternatives to missiles, and that it wouldn't "make sense" to build them, has been made before. The topic is sometimes referred to as "the donkey option"--that is, weapons of mass destruction are now small enough for the humblest of cross-border delivery methods.

In contrast, the Rumsfeld Report tried to dispatch that idea. "A successfully launched ballistic missile has a high probability of delivering its payload to its target compared to other means of delivery," it claimed.

Essentially the same idea was put forth in a January 1998 report of the U.S. Senate Committee on Governmental Affairs, "The Proliferation Primer." This report stated that some two dozen countries have or are working to develop ballistic missiles and that the trend toward longer ranges shows that these platforms are "the delivery vehicle of choice."

North Korea's 'cash crop'

The crop took root decades ago. At first, the Pyongyang regime cooperated in a Chinese missile program, the DF-61, which was aimed at developing a liquid-fueled missile with a I-ton payload and a range of 600 km.

When that project was canceled by the Chinese government, North Korea then reverse-engineered the 300-km Scud B ballistic missile, a venerable war rocket based on the Russian R-ll (SS-1 in the terminology of the North Atlantic Treaty Organization), first flown in 1953. The Scud was not particularly accurate, with a circular error probability (CEP) of approximate- ly 1 km -- that is, at least half of the warheads would fall within that range of the aim point. But when many of them were fired toward large-area targets -- a city, for example -- they could be effective [Fig. 3].

North Korea reportedly flight-tested this missile in 1984 and improved its range further, probably by upgrading engine performance. Within two years, it was manufacturing its own Scud Bs and was reported to be exporting them soon afterwards, mainly to Iran. With cash earned in these sales, the Government then developed a Scud C with a range of 500 km, probably by lengthening the missile and increasing its propellant load by about 25 percent. This improvement may have been aided by a study of the wreckage of long-range Iraqi Scuds fired by Iraq into Iran in the mid-1980s.

In the past decade, North Korea may have sold as many as 400 Scuds, mainly to Iran and Syria, and the country is now churning the missiles out at a rate of 100 per year. Perhaps an even more alarming fact is that the Russians have reportedly been cutting into North Korea's market share, particularly with Iran, "in areas ranging from training to testing to components," according to CIA director George Tenet, during his 2 February congressional testimony. There's nothing like competition to build an efficient market.

Scud stacking

In some cases, such as a nuclear warhead, the weight of the required warhead sets the missile's upper range. In other cases, such as the Iraqi Scud-variant called the al-Hussein, the range to the desired target sets the upper limit on the amount of high explosive that can be carried [a typical tradeoff between range and warhead weight is shown in Fig. 4].

Regardless of their payloads, North Korea's missiles are inch-ing toward intercontinental mode. In May 1993, the nation made the first -- and so far only -- test flight of a more powerful missile called the No Dong (named for the city where Western intelligence first detected it), estimated to have a maximum range of 1000-1300 km. Support equipment soon was spotted in operational military units, but Western observers are uncertain about whether actual missiles have been deployed -- which is probably the North Koreans' intention.

Foreign customers, particularly Libya, Iran, and Syria, are reported to have helped to finance the development of the No Dong in exchange for the promise of future deliveries.

So the No Dong may already be armed and dangerous. In any case, North Korea has not stopped, but has gone on to develop longer-range, two-stage missiles. Its August 1998 pseudo-satellite launch was powered by the Taepo Dong 1 missile, which it is anticipated will have a range of 2000 km [Fig. 5].

The most detailed unclassified analysis of the North Korean missile was published in 1994 by MIT's Wright in conjunction with Timur Kadyshev, at the Russian Academy of Science's Center for Program Studies. The authors concluded that the No Dong range was near the maximum possible for a single-stage missile derived from the Scud. Even though single-stage U.S. and Soviet missiles of the late 1950s (so-called intermediate-range ballistic missiles, or IRBMs) reached a distance of 2500 km, they were substantially more powerful than the No Dong.

A more productive approach for improving range compared to pumping up the engine power involves mounting a smaller rocket atop the No Dong, as a second stage. Wright and Kadyshev calculated that a No Dong with a modified Scud as a second stage could travel 1800 km with a 1-ton payload-- and further, of course, with a smaller warhead.

To judge from the photograph North Korea released of the supposed satellite launch, this appears to be exactly what the North Koreans did. The missile flew due east, dropping its first stage into the Sea of Japan. The path took the second stage right over Japan -- and deep into a world controversy that has yet to die down.

At first, U.S, and Japanese official statements expressed alarm at North Korea's long-expected but dreaded advancement to multi-stage long-range missiles, as well as to the unwarned-of overflight of Japan's northern Honshu island. But after four days of silence, North Korean officials shocked the world with the claim that the launching had actually carried a satellite into orbit and denounced foreign reports of a military missile test.

Baffled Western observers went back to the data, searching for signs of a small third-stage missile carrying a satellite, and re-checking low-earth orbits for any uncataloged objects that might correspond to the North Korean claims. After a week, spokesmen for the U.S. Space Command, in Colorado Springs, Cole., announced that they had been unable to locate anything in orbit that seemed remotely related to the claim.

At the same time, amateur radio listeners around the world had tried in vain to pick up the radio beacon alleged to be aboard the satellite. As for tracking data obtained during the launching itself, the Pentagon admitted that "there is considerable disagreement throughout our own intelligence community as to how to interpret those tapes."

If the satellite story is entirely fictional, it is a well-crafted hoax (indeed, the photograph has received its fair share of scrutiny as a hoax). Many of the details, such as the eastward launch and an orbit announced to be highly elliptical, have the feel of a genuine program that happened to go bad. Similarly, the high acceleration of the launch -- reportedly reaching orbit in under 5 minutes -- is comparable to early U.S. intercontinental ballistic missile (ICBM) launches.

But in October 1998, MIT's David Wright published calcula- tions that cast doubts on the plausibility of the North Korean claim. He compared the required performance of a hypothetical solid-fuel third stage to the known performance of a wide variety of other upper stages. His conclusion: "North Korea would not have been able to put a satellite into the highly-eccentric orbit it claimed, and very likely could not have even achieved a circular low-earth orbit." The validity of his underlying assumptions about velocity requirements, acceleration profile, and payload, however, remains uncertain.

Assuming there really was a failed satellite launch, new attempts can be expected both for the propaganda and the commercial advertisement value. Some North Korean scientists have even talked about plans to launch a geosynchronous communications satellite for domestic and overseas television broadcasts. A far simpler, yet much more spectacular, mission well within North Korean capabilities would be a suborbital manned space flight to an altitude of 200-300 km.

Observers also believe North Korea is working on even bigger military missiles, specifically the Taepo Dong 2, which would be a true intercontinental launcher. It would have a range of 4000-6000 km, and lightweight versions (warheads of half a ton or less) might reach 10 000 km. Existing military rockets would be used in wartime to attack U.S. forces in Japan and even on Okinawa and Guam.

But North Korea's ultimate goal, according to North Korean defector Colonel Choi Ju-hwai, "is to reach the mainland of the United States." At the current rate of progress, Western intelligence sources claim, the nation could reach that level of capability in five years or less.

Without any usable information about the No Dong missile from its builders, Western analysts have tried to design their own equivalent using Scud-based technology. In their 1994 unclassified analysis of North Korean missiles, Wright and Kadyshev found that a missile harnessing four Scud engines-- which the North Koreans have well in hand--and a larger set of propellant tanks would indeed be able to send a 1-ton payload about 1000 km. They examined a number of ways this missile could be improved even further.

The specs of the new model

All four turbopump-fed engines together deliver a thrust of about 520 000 newtons with a specific impulse of 240 seconds. The missile is steered by motorized graphite vanes in the engines' exhaust stream, based on commands generated by three gyroscopes and body mounted accelerometers.

From launch to burnout takes about 70 seconds, followed by about 10 minutes of ballistic flight until impact. The warhead probably separates soon after engine shutdown in order to limit aerodynamic dispersions on reentry.

The Wright-Kadyshev study also indicated that if the missile body were manufactured from aluminum instead of steel, the reduction in structural weight would in effect increase the range from 1000 to 1300 km--exactly the figure suggested by Western military intelligence sources. Whether that has actually already been done or is merely a worst-case forecast is unclear.

Estimating accuracy, on the other hand, is trickier, and depends on the precision with which the missile can be pointed during powered flight and with which the engines can be shut off. To an even larger extent, it depends on how much buffeting and uncontrolled aerodynamic lift (and crosswinds) are encountered during reentry. Wright and Kadyshev made a "best guess" of 2-4 km CEP for the 1000-km No Dong.

Reliability analyses are also extremely difficult. According to one report, Iran sent back some North Korean Scud C missiles, and withheld substantial payments, because of defects in the delivered missiles.

Whether or not there are reliable and operational No Dongs in the North Korean arsenal today, this particular genus of genie is clearly out of the bottle. In 1998, two other countries flight-tested their own missiles, also seemingly based on the No Dong design: Pakistan's Chauri, in the spring of 1998 and Iran's Shahab-3, that summer. Both of them had advertised ranges in excess of 1000 km.

The new balance of terror surely does not approach the Cold War's superpower strategic principle of Mutually Assured Destruction. But the military and (and hence) political land- scape has inalterably changed, and new strategic principles have yet to be formulated.

The ones that got away

For example, despite the highest level of U.S. military attention to East Germany in the 1980s, only after the collapse of the Communist regime in 1990 did the United States find out that the country had obtained two dozen Soviet SS-23 ballistic missiles (500-km range). These missiles, and similar equipment delivered to Bulgaria and Czechoslovakia, completely eluded U.S. detection.

In another intelligence "oversight," in 1987 the United States was caught by surprise when Saudi Arabia acquired about 50 CSS-Z missiles (with a 3000-km range) from China. The Saudis had secretly negotiated and planned the purchase for more than a year, in order to counter the large missile forces in Iraq and Iran. The deployment was discovered only when imported Chinese missiles supposedly destined for Iraq were unloaded in Saudi ports and then trucked south, not north. The Saudis have never fired any of these missilesi their chief motivation to do so--the Iraqi mis- sile threat--has receded since the Gulf War, but the missiles remain somewhere in Saudi Arabia.

Although the United States has occasionally successfully deterred some North Korean missile shipments, in 1982 a North Korean freighter loaded with Scud C missiles managed to elude U.S. naval pursuit and reached an Iranian port, from which the missiles completed their delivery overland to Syria

Knowing what, how, and why

One intriguing approach to predicting Third World missile progress is what is called intelligence anticipation. The idea is to set up teams with constrained information and resources, with which they have to try to emulate what another nation might be able to do. William Graham, formerly science advisor to the Reagan administration and in 1997-98 a member of the Rumsfeld Commission, dubbed this procedure TRYINT, or try-it-out intelligence.

Two TRYINT programs came from the U.S. Air Force. Using only open source literature and commercially available products, small teams of educated but inexperienced engineers were turned loose to build countermeasures to either U.S. planned theatre missile defense systems or a homegrown cruise missile

According to Air Force congressional testimony, "both programs have been very successful." Clearly many official government estimates had grossly underestimated the capabilities of missile programs developed by unfriendly countries.

These examples should serve as cautionary tales that U.S. knowledge of the state of Third World missile systems is imperfect and subject to sudden upset. This theme is particularly disquieting when seen in the context of: other differences between U.S. missiles and those of other states. For example, neither the United States nor Russia would dream of deploying a military missile after only a single test night. But from the point of view of the North Koreans, the requirements are entirely different. So they may well have done so.

This assessment was echoed in last July's Rumsfeld Report. "Newer ballistic missile development programs no longer follow the patterns initially set by the U.S. and the Soviet Union," the report stated. "These programs require neither high standards of missile accuracy, reliability and safety nor large numbers of missiles."

Speaking to a Senate hearing a year ago, North Korean defector Colonel Choi explained the difference. "Unlike U.S. missiles which require capability for surgical strikes," he testified, "the North Korean missiles are not designed For such precision. What they are targeting is a general region rather than specific facilities."

As MIT's Postel put it, "Ballistic missiles are weapons of terror, and they will be used by terrorist states, or even terrorist groups, if they can get them. They are fundamentally psychelogical in nature."

"These missiles get people's attention," he continued. But, he added, maybe too much so. "I think the U.S. shouldn't be yelling as much at every missile we see. The more noise we make about them, the more incentive there is for someone to build them to get the attention.

Deterrence clearly is a prime motivating factor. Rather than planning to use these missiles in an ideology based, out-of-the blue attack on the United States, it seems more likely that nations such as North Korea, Iran, and others plan to balance and neutralize U.S. strategic weapons and dissuade the United States from applying force within regions of interest to the nations themselves.

Choosing warhead size

In the World War II-era V-2 rocket and the 1950s' Scuds, when the task was to drop high explosives on wide-area targets, a payload of 1 to 2 tons was needed. For the early nucleararmed missiles, multi-ton warheads had to be thrown for thousands of kilometers, often with great accuracy.

But the modern concept of weapons of mass destruction encompasses not only nuclear weapons (which have been great ly reduced in mass), but also chemical and biological weapons. The latter classes of warheads can be much smaller, while requiring far less accuracy than brute-force explosives, conventional or nuclear.

It is widely believed that effective missile-launched chemical and biological dispensers can be constructed to weigh 10 kg or less. Large numbers of these bomblets can be stacked into a nose cone for delivery across a wide area in the target zone. Depending on required range, a greater or lesser number of the bomblets could be loaded.

Late in 1997, a U.S. Senate committee investigating missile proliferation was briefed by General Bernard Schriever, a living legend of U.S. rocketry. In the '50s, he had headed up development of the Atlas missile, the nation's first ICBM. Schriever told Congress that three main technical challenges faced any group trying to build a long-range missile: range, guidance, and systems integration.

Meeting each challenge is many times easier today than it was 40 years ago. Extending a missile's range, noted the veteran rocket man, "is among the easiest and most straightforward things to do." Adding side mounted engines or upper stages is one possibility. And as warheads become lighter, range is automatically increased. Required accuracy is made easier through commercial inertial guidance systems, or those based on the Global Positioning System.

On a larger scale, missile systems integration and testing was a monumental challenge 40 years ago, primarily because it had never been done before. Now the methods are taught in engineering school. Much of the expensive integrated testing for the Atlas program would today be done through computer analysis and simulations, with software made public by NASA and other aerospace information centers

On the large and small scale, the availability of information can allow today's missile builders to piggyback on the shoulders of their predecessors. In the big picture, they already know that such missiles are feasible (40 years ago, no one knew if anyone could build one). At the nitty-gritty level, the scope of technical papers available over the Internet or to the tens of thousands of non-U.S. graduate students (including many from the hostile nations) answers almost all imaginable technical questions, and then some, about building missiles.

Despite U.S. hopes based on funding for international cooperation in space, thousands of experienced Russian rocket technicians have lost their jobs in the past decade. The price for such services testifies to the buyeis market: Iran is reportedly offering US $200 per month, only a little more than their wages within Russia--if they had a job at all.

Reports of individual Russian missile experts in Iran and elsewhere have circulated for several years. Still, most nations get missile advice from each other on a more formal basis, with technical teams swapping solutions.

The Iraqi experience

In 1986, the Iranians began hitting Baghdad with their own Scud B missiles. Iran's capital Tehran was much farther from the front lines and Iraq's 300-km-range Scuds were incapable of retaliating over that distance.

In a crash program, Iraqi technicians (aided by contractors from North Korea, Egypt, East Germany, and, it should be noted, West Germany) lengthened the Scud fuselage and lightened its warhead. Three Scud Bs provided tankage and structure from which they could assemble two new missiles. The result had dou- ble the range of the Scud, and was dubbed the al-Husseyn.

By early 1988, the new missile was dropping warheads on Tehran. In two months, more than 200 al-Husseyns were fired, killing or wounding thousands of Iranian civilians and forcing about half the city's population to evacuate. The operational use of this new missile apparently was a complete surprise to U.S. intelligence agencies, which had received no warning that it was being developed.

Iraq then advanced to the al-Abbas missile with a range of 900 km, and also began development of a 3000-km range missile. On 7 December 1989, in another launching that caught the United States unawares, Iraq fired a three-stage missile called the al-Abid, which Bagdad, like its future imitator North Korea, claimed was a space satellite launching.

The first stage appears to have been four or five Scuds strapped together and the second stage consisted of two side-by-side Scuds. As a 2000-km two-stage military missile, it was dubbed the Tammuz. An unidentified third stage was added for the satel- lite launch--which was a total failure.

Assessing the assessments

First, estimates based on assumptions of so-called indigenous programs are dangerously naive. For more than half a century, missile technology--and missile experts-- has sped from country to country with dazzling ease. For every set of itinerant scientists stopped at some border, or every truckload of surplus missile gyros fished out of a river, it can be assumed that others succeeded in reaching their destinations.

Secondly, a nation's intelligence assessments based on its own historical precedents--an intelligence failure called mirror-imaging will also lead to surprises While some missile design solutions can easily be discovered by any reasonably inventive group, there are also wild-card approaches that nations may choose to adopt, in the light of unique requirements or opportunities.

Third is the possibility of ignoring the commonly used warning: "To every difficult problem, there is a solution that is obvious, easy, and wrong." One need only bear in mind the intense debate over even the nature of the missile threat, let alone over the huge investment in ballistic missile defense.

Lastly, the purely military value of such missiles is clearly not the motivating force behind them. However inefficient they may be in the calculus of casualties for "kills per buck," they have vastly enhanced value as "frights per buck."

From the days of the V2 of 1944 to the Iraqi missiles of the Gulf War, such systems have inflected massive damage that could be called virtual attrition: compelling forces to divert vast amounts of men and materiel to counter a weapon whose prime value is in terrifying home populations.

Such systems can most probably be expected to inflict similar asymmetric costs--in military sorties, in operational con- centration, and in countervailing hardware developed--for the foreseeable future. The technology is out there, the clever engineers are out there, and the nations willing to pay the price are out there--and as a result, so are the missiles.
------

To Probe further

Also worth checking out are the individual articles and their detailed references listed in a special report entitled "Ballistic missile defense: it's back." That material was published in IEEE Spectrum, September 1997, pp. 26-69.