BMD

I see we managed to intercept another test missile with a kinetic kill vehicle over the Pacific. It's quite a technical achievement and I'm impressed that the kinetic kill vehicle can guide itself to hard impact at such a high closing rate.

Missile defense isn't entirely new. We first proposed the Sentinel system in the 1960s, if I recall, which was supposed to provide a "thin" defense over the entire United States. Not enough to take the edge of a determined Soviet attack, but maybe enough to blunt an attack from a lesser nuclear threat (that is, the People's Republic of China). The system comprised two missiles - Spartan, a long-range missile designed for exo-atmospheric interceptions, armed with a thermonuclear warhead and designed to score X-ray kills against incoming warheads. The other missile was the Sprint, a short-range terminal defense missile armed with a neutron bomb and designed to score neutron kills against incoming warheads. The Sprint in particular was a fabulous piece of rocket technology.

The value of this "thin defense" was eventually adjudged to be not worth the money, so the project was rescoped into Safeguard, which would protect a small number of ICBM silos in the Northern Great Plains. It used the same basic missiles, Spartan and Sprint, but now instead of protecting cities, it protected ICBM silos. (This has value in nuclear game theory. Whether one wants to believe it or not, the "doctrine" of Mutual Assured Destruction was the underpinning of the nuclear standoff, and any attempt to "harden" the civilian population was seen as destabilizing. On the other hand, hardening nuclear forces was seen as stabilizing because it made the enemy even more uncertain that his first strike could seriously incapacitate the enemy).

Safeguard was operational for a few days, but in the end it was canceled. One big reason was that its stated mission - reinforcing MAD by protecting US ICBM silos - seemed a little redunant in view of the fact that even one US Navy ballistic missile submarine could devastate the Soviet Union. The Soviets could never besure of killing all our submarines in a first strike, so the real nuclear safeguard lay in our submarines, not in the small set of Safeguard-protected ICBM silos. (A US nuclear submarine at the time carried 16 Poseideon missiles, each armed with about ten nuclear weapons and a few decoys. That's about 160 warheads, enough to wipe the Communist aggressors off the map. And that's just one submarine.)

And there the matter rested, inactivity enforced by the ABM Treaty. But in the 1980s, the US Army carried out the "Homing Overlay" experiments, an attempt to develop a missile that could destroy an incoming enemy warhead by direct impact. Homing Overlay didn't require nukes to compensate for errors in guidance; technology was adjudged sufficiently mature to enable direct hits. Homing Overlay was ultimately successful in intercepting a test missile, and the modern US ballistic missile defense system appears to be really nothing more than a better-developed Homing Overlay system.

In missile defense, there are roughly four "kill zones", or more properly four kill-times. They are, in order of effectiveness:

1. Prelaunch. Blowing the enemy's missiles up on the ground. This is both easy and hard. It's easy because if you can get a spy close enough to the missile she can simply stab it with an icepick. It's hard because it's hard to get enough spies that close to enemy missiles. In practice you end up having to rely on standoff kill techniques, preferably fast ones, and if the enemy shows any inclination to base his missiles in silos, you're basically forced to use nukes to preempt the enemy attack. This requires precise intelligence and carries significant political costs - try explaining to the world that you nuked Kreplakistan off the face of the Earth because you had a hunch they were going to launch a missile. And what if you're wrong?

2. Boost phase. Blowing up the booster while it is still burning. This is perhaps the best time to intercept an enemy missile. Their booster is still burning, producing a very energetic exhaust plume that is painfully easy for infra-red systems to track. The missile isn't moving very fast (in the early part of the boost phase at least) and best of all, you don't have to actually disable the warhead, just the booster. Liquid-fuel boosters are quite easy to disable. A brick would suffice. Solid-fuel boosters are harder to disable because their steel combustion chamber walls provide an armor effect for free, but it can still be done. The chief problem with boost-phase interception is the time problem. You've got a few minutes at the most in which to detect the launch, characterize its trajectory, assign an interceptor, and kill it. Time-of-flight considerations mean that your interceptor missiles have to be either in or very near the hostile country. The US Navy is developing the SM-3 and SM-4 missiles which may have some utility in boost-phase interceptions, but as a general rule of thumb, boost-phase interception pretty much requires directed energy weapons (lasers, particle beams and the like).

3. Exo-atmospheric phase (often called the mid-course phase). Blowing up the warhead while it is still outside the atmosphere. The idea here is that shortly after the enemy launches its missiles, you launch your interceptors. Once out of the atmosphere, the kill vehicle tracks the enemy warhead by some means (infra-red, most likely) and steers itself in for the kill. The actual interception takes place at altitudes of over 100 miles, above the atmosphere. The advantages of this are that the interception takes place above the atmosphere and presumably over an ocean, so if you use thermonuclear weapons (as Spartan did) to enhance your kill ratio, you don't have to worry so much about radiation. Also, enemy warheads in the exo-atmospheric phase aren't really able to maneuver all that much. Once you have a good estimate of its trajectory, getting the interceptor to the target is largely a matter of mathematics. The main disadvantage is the kill ratio may not be all that great. The enemy's booster has burned out so you don't have that juicy exhaust plume to track. The enemy warhead is moving about as fast as it ever will, so your closing rate may well be on the order of five kilometers per second. The fairly late intercept also gives the enemy time to deploy various countermeasures, chiefly balloons and other decoys, so your tracking and classification problem is more difficult. The Safeguard system's Spartan missile was designed to operate in this phase. It used a thermonuclear weapon whose powerful pulse of X-ray radiation could degrade, deactive or seriously mess up an enemy warhead even if the Spartan missed by a fairly wide margin. More modern systems delete the thermonuclear weapon and aim to destroy the enemy warhead by physically smacking into it at extremely high speed - again, a brick would suffice if you could somehow guide the brick to its target.

4. Terminal defense. Intercepting the warhead just over its target. This is the least feasible technique, used only in desperation. It amounts to intercepting the enemy warhead directly over its target, presumably a city. Reaction time is critical as the enemy warhead is coming down extremely quickly. The enemy warhead can also use the atmosphere to maneuver, flying a primitive but reasonably effective spiral evasion pattern that, at high enough speeds, makes interception very difficult. You'll need to augment your interceptor with nuclear weapons to improve your chance of a kill. The Sprint missile used a neutron bomb (also known as an "ERRB" or "Enhanced-Radiation Reduced-Blast" warhead, basically a thermonuclear weapon whose X-ray mirrors don't absorb neutrons). The pulse of neutrons from the ERRB warhead had a pretty good chance of either deactivating the enemy nuke or causing it to fizzle in a low-order detonation, especially if it used plutonium instead of uranium. Good, right? Well, not really. According to the designer of the neutron bomb, detonating a neutron bomb ten miles above a city would produce so much neutron radiation at ground level it would kill trees and grass, not to mention people. So maybe you're better off not intercepting the warhead at all and hoping that it's a dud (which isn't necessarily a vain hope).

It thus strikes me that kinetic exo-atmospheric kill is the only feasible interception technique. Boost-phase interception requires reaction times faster than missiles can provide, and terminal phase interception causes so much damage it's hard to see how a successful interception is any better than an unsuccessful one.

So where do I stand on all of this politically?

I was opposed to deployment of anything like the Reagan Star Wars program, for the following reasons. First, I thought it was too expensive and would bankrupt the country. Second, I thought it was inherently destabilizing and would make nuclear war more likely rather than less likely. (Assume you're the Soviet commander in chief and your staff tells you that tomorrow the Americans are going to turn on a ballistic missile defense system that can intercept 99.9% of your warheads. What do you do? Submit to American supremacy, or nuke them before they turn the system on?) And third, I couldn't imagine how the system could ever be subjected to a full-scale test. Its first combat employment would be its first test, and I just didn't think that was good.

I was in favor of Star Wars research. My rule of thumb is that more research is always better than less research. It's difficult to say what will come out of any given line of research, and I was prepared to fund Star Wars research at a modest level even though I was quite sure that a functional Star Wars system would never come out of the end of the research pipeline. But other stuff might, and I couldn't say what.

I am in favor of US Army and US Navy attempst to develop theater-level ballistic missile defenses, in particular the US Navy's Aegis effort. Protection of fielded forces from enemy ballistic missiles seems like a worthy goal to me, and forward-deployed warships armed with SM-3 or SM-4 missiles seems like a reasonable basis for such a system.

I am also in favor of the current GMD system deployed in Alaksa and California, but only so long as its objectives are clearly stated. It should not be sold as a nationwide defense system, because it can't deliver on that claim. It can't blunt a Chinese attack, far less a Russian one, and attempts to morph it into such a system turn it into just another variant of Star Wars. But as long as the system's objective is to defeat small attacks from countries like North Korea and Iran, I am in favor of it. Intercepting ten relatively crude North Korean ICBMs is something I believe the system can affordably do.

It's a matter of percentages. Let's say that the GMD system can intercept 50% of incoming warheads. Say the North Koreans launch ten missiles. We pick off five. The result is a horrendous catastrophe in that we lose five cities, but better to lose five cities than ten (though one could assume that the North Koreans would double-target five cities, so we'd lose two or three). Now let's assume a Chinese attack. I'm not saying that a Chinese attack is likely, but just for the sake of discussion let's make that supposition. They could probably deliver 400 warheads in a first strike. We intercept 200. That leaves 200, and 200 nukes on the United States would still be the end of American civilization, so why bother? Better in this case to work to make sure the war never happens in the first place than to try to survive 200 nukes.

My point is that as long as the objectives of ballistic missile defense remain modest - blunting a North Korean attack, say - I'm in favor. But the moment the system begins to be sold as a Star Wars-like system capable of blunting a full scale attack from other quarters, I'm opposed for the same reasons I was opposed to the original Star Wars program - cost, destabilizing influence, and untestability.

The best nuclear war, of course, is the nuclear war that is never fought. The best arms race is the arms race that doesn't cost any money. The best missile defense system is one that can protect us against rogue states who have no vested interest in the status quo, but one that is not so large that it makes countries who do have a vested interest in the status quo nervous. The system needs to be big enough to blunt a North Korean attack, but not nearly big enough to give Chinese and Russian nuclear strategists the jitters - or American taxpayers the vapors.

And that's where I stand.