Anyone who remembers the 1964 Gulf of Tonkin incident will understand that at night radar can produce false targets. After attacks on one night, the next night a U.S. destroyer reported detecting attacking North Vietnamese torpedo boats that many believe were not there. This was not some intentional error; it was because of problems inherent in radar. During World War II a U.S. task force fought the "Battle of the Blips," firing hundreds of rounds at what turned out to be anomalous echoes of islands in the Aleutians. Seeing a nearby ship blow up would tend to make any radar targets seem quite real. Radar pulses bounce off the sea and can be trapped by layers in the atmosphere. A crew unaccustomed to operating very close to shore might easily make mistakes. A corvette, moreover, is unlikely to have enough radar operators on board to develop the sort of expertise needed to determine what is happening under unusual circumstances. The radar target the second corvette attacked can even have been reflection off the nearby island.
If the corvette was conducting a special mission, she may not have been using her radars, or was using them intermittently. One of the advantages of modern navigation using GPS is that, at least in theory, a ship need not use radar to determine her location in relation to known land masses. She can operate much more passively, though of course without radar she cannot detect other ships or craft. If two corvettes were operating together and the other corvette dealt with a radar target, it is possible that the Cheonan was operating in a silent mode, depending on her companion to create the surface picture for both ships. The corvette is in no way a stealthy ship, but by not using her radar she also would not be broadcasting herself to North Korean listeners and might better be able to pick up their radar and radio signals.
Sonar Not Much Help
The corvette did have sonar, but it was a standard antisubmarine type, which was probably ill-adapted to detecting mines in the water column - and most unlikely to detect a mine on the seabed. South Korean references to efforts by the North Koreans to achieve stealth by operating very quietly at low speed suggest that standard operating procedure is to listen rather than to ping, which would make mine detection entirely impossible.
Operations unusually close to shore could expose the ship to mines that had drifted south. It is also possible the ship was on a routine mission to probe North Korean shore defenses, and that a local officer reacted by detonating a controlled mine. The conditions, which make it difficult to operate in a littoral area, would also probably complicate the perceptions of anyone trying to sense movement close offshore. If the attack on the ship really was a local initiative, it would not have been associated with any preparation by the North Korean government, and the North Koreans are most unlikely to admit anything now.
Defectors from the north have claimed that since 1999 the North Korean navy has formed special attack (suicide) squads equipped with unusually quiet low-speed craft carrying torpedoes and acoustic mines. North Korea also has midget submarines armed with two torpedoes each.
The argument against North Korean involvement (or at least intentional involvement) is that, whenever it takes offensive action, the North goes into a pre-emptive defensive posture, for example limiting officials' travel. By way of contrast, at this time North Korean behavior was entirely normal, the defensive posture and threats of counterattack coming only after the sinking.
The ship sank in a relatively shallow area, which we might now call littoral, and the sinking may illuminate the problems of operating in other littoral areas. Above all, littoral waters, particularly the closer they get to a beach, are confusing. Things drift along the sea bottom, sometimes over considerable distances. Mines may remain active years after being planted, long after their theoretical lives are over. Swept waters may not really be clear. There have been enough examples of non-clearance to suggest that the Korean sinking is not so unique. For example, for years the Russians have been trying to explain why a World War II German ground mine destroyed the battleship Novorossiysk at Sevastopol in 1955, in an anchorage supposedly cleared years before. (Fanciful explanations include an attack by Italian commandos, who resented Soviet use of the ex-Italian battleship).
Littoral areas are confusing because lines of sensing - sight, radar, and sound - are inherently short, and local conditions and terrain complicate normal sensor operation. Those who operate habitually in such areas may not be too confused, but a blue-water navy venturing into them lacks that necessary experience. For example, in the 1980s the effects of anomalous radar propagation surprised the U.S. Navy when it began intensive operations in the Persian Gulf.
Unreliable sensing in turn makes surprise attack more feasible. If the reports of North Korean naval suicide squads are accurate, they reinforce this point: it takes less to mount an attack close to shore, and it is easier to initiate underwater attacks there. If it succeeds, an underwater attack is more likely to sink a ship than an above-water one. Antiship missiles tend to wreck a ship over a more or less fixed length considerably shorter than a ship. If the ship is large enough, and if her components are properly dispersed, she will probably survive such an attack as a fighting unit. It is unlikely that even a corvette like the Cheonan would have succumbed to a single Silkworm, for example.
The explosion, which sank the corvette, was probably from a warhead much smaller than that of an antiship missile. It snapped her hull in two - which suggests a ground rather than a moored mine - and a large part of the ship sank instantly. This lesson is very easy to forget. Because missiles are visible and dramatic, with well-known "faces" and names, many of the world's mines are little known, their designations virtually unknown. Even torpedoes receive little publicity compared with missiles. This is not, incidentally, a new problem.
In considering World War I's naval weapons, one thinks of heavy guns rather than mines and torpedoes. It is easy to forget that guns - the World War I equivalents of modern antiship missiles - were not considered ship-killers; they were expected to smash ships into ineffectiveness by inflicting cumulative damage. If they worked, torpedoes and mines could kill in one shot. Some contemporary tacticians understood the difference, but it is easy to miss in retrospect, and many historians have missed the point. It would be a pity if we did, too.
So what can we do ? The best solution to the sensing problem is to multiply sensors and cross-reference them, ideally using different kinds of sensors as well. It seems possible to, in effect, wire a littoral area of interest to make it much less likely that false targets will seduce us or that we will miss real ones. Sensors on that scale can be distributed either by unmanned vehicles or by air strewing. The first technical problem is that the sensors have to know where they are; otherwise their combined data provide false information. The second is to provide enough communication capacity to merge their data. The U.S. Navy is working hard on both problems, and we hope it will succeed. The South Korean disaster suggests that more traditional approaches can be unacceptably expensive.