Beauty of Gray

Thursday, October 24, 2002

Snipers, terrorism, and the Battle of the Atlantic

With the recent arrests of the two suspects, this article is bordering on irrelevancy, but this is my website and I like the idea so I’m going to write it anyway. If you need relevant, there has been widespread worrying that, even if the snipers weren’t terrorists, that terrorists and other copycats could pick up on the idea, making it a real ongoing national problem, rather than a limited local one. (I’m not convinced this is plausible, but that’s an article for another time.)

The sniper attacks, and terrorism in general, are in many ways similar to the German submarine offensive in the Atlantic during WWII. As with the sniper, the submarines would deploy, waiting in hiding for a defenseless merchant vessel to come along, which they would then attack and attempt to sink. As with the sniper, often the first warning of a submarine presence was the sighting of a launched torpedo or its impact. As with the sniper, the submarines were able to hide in a wide area, only coming in contact with allied forces when they made their attacks. And as with the sniper, once they had made their assault, the submarine’s goal was to escape undetected, while the allied forces would respond attempting to catch the U-boat and sink it.

These attacks, and the allied counter-attacks and defensive strategies, are usually referred to as the Battle of the Atlantic, and the Allied success in this area was a vital requirement for the war in North Africa and Europe. For the US to project power into Europe, to get troops and equipment to the theater, and to supply their own and allied forces, required control of the shipping lanes, or at least the reduction of attrition rates during shipping to a sustainable level.

The entire story of the Battle of the Atlantic, all the tactics and counter tactics, is quite interesting. For the mathematically inclined readers (don’t be scared—you only need to know probability and maybe Calculus I) I’d recommend this book, a declassified version of an after action report written by the US scientists who helped head up the anti-submarine warfare effort. There’s lots of fascinating information in there, as well as an idea of the ways that mathematical analysis can aid in military decision making.

Side note: Most people are aware of the Manhattan Project, and many know about the MIT Radiation Laboratory which helped refine and develop radar technologies during the war. But much less well known is the third major area of effort for US scientists during the war, the field of military operations research (OR.) In its broadest form, it was the application of mathematical and operational analyses to the specific problems faced by the military. The British pioneered the field in the years before the war, trying to figure out the best way to use the newly developed radar systems. The US applied these concepts in many fields, but particularly in anti-submarine warfare, forming ASWORG, the Antisubmarine Warfare Operations Research Group, which went through several incarnations but is still around today as the Center for Naval Analyses. (Look here and here for brief histories of military operations research.) In fact, this effort was so successful that it really spawned the entire new discipline of OR, and its sister field of systems analysis.

So sniper attacks have a real family resemblance to submarine attacks. Is there anything that ASW responses in WWII can tell us about effective countermeasures against snipers (and terrorism in general)? Well, sort of, if you’re willing to stretch it a little. There were three main tactics the allies employed to fight the submarine menace.

1. Stopping the supply—they bombed ports and German industrial centers trying to slow or prevent the manufacture of submarines or destroy them before they deployed.
2. Interdiction—for submarines entering the Atlantic from Germany’s Baltic ports, there were choke points that were small enough to be effectively patrolled, catching some of the submarines before they made it to the Atlantic.
3. Convoying, about which more below.

For sniper attacks, the first two tactics don’t really have that much to offer, although they are the methods of choice for combating terrorists. (Elucidation of this historical analogy is left to the reader.) Depending on the location, some police effort towards interdiction and early detection could bear dividends, but many areas have too many possible shooting sites for this to be very effective. Which leaves convoying.

Convoying was a simple yet extremely effective basic tactic for warding off submarine threats. Instead of having ships leave port in ones and twos, the Allies waited until they had a sizable number of ships, and then sent them out together. The most important thing that convoying allowed was the maximizing of defensive resources. The Allies only had a limited number of destroyers which could escort the convoys all the way across the Atlantic. There was a window in the central Atlantic that no land-based aircraft could reach, and in that window the convoys had only their destroyer escorts to protect them. (It’s wasn’t until later in the war, when the Battle of the Atlantic had already largely been won, that the Allies had enough small carriers—modified cargo ships that could carry a half-dozen aircraft for convoy protection—that integral air cover was available in the Atlantic.)

Escorts did not serve in a purely defensive role, however. To the contrary, their main purpose was not so much to prevent a submarine attack—the convoys were too large and the escorts too few to accomplish that task—as to make such an attack as dangerous as possible. (Again, the later innovation of escort carriers changed the balance here, as planes can search a large enough area to have a good chance of catching a submarine before it attacked.)

The goal was to have such a rapid and effective response by escorting warships that they could sink the attacking submarine. They used some technological advances to achieve this—namely sonar to locate submarines and the “hedgehog” depth charge launcher to sink a submarine if you did find it. (The hedgehog was a multi-charge launcher that would shoot a salvo of depth charges in a pattern. This spread of charges made a kill much more likely than an individual shot. One of the problems discussed in the OR book mentioned earlier was what pattern to use to ensure the maximum probability of kill against a target whose exact location was unknown.)

So, how does this translate to the sniper situation? Obviously, a complete convoy system for everyday life is impossible, and there’s not enough police presence to “escort” all such convoys even if it weren’t. However, the basic principle—that there’s safety in numbers—still holds. One reason the DC sniper was able to remain at large for so long was that he chose mostly isolated locations for his attacks. More people in the vicinity makes it more likely that there will be a good witness to the shooting, and also makes a response quicker. If someone is right there to call 9-11, the police can respond more quickly to try and catch them.

As with the submarine case, the most critical parameter is the response time. The initial shot establishes a rough position, but the longer you wait after the shot, the great the area of uncertainty is. If you wait too long, there’s no way to catch the sub (or sniper); you need to get a response force on the scene while the enemy is still in the area. The submarine had the advantage of being able to hide underwater; the sniper has the advantage of being able to “hide in plain sight,” mixed in with the rest of the traffic.

The ultimate goal, as with submarines, the best counter-strategy is to make the attack as dangerous as possible. Make escape more difficult, which means being aware so you can be a good witness and help to catch them. If you can swing the odds, then sniper will only be able to get away with one or two shootings, rather than 13. (The ultimate extension of this idea would be to have enough armed civilians that any group of people could, on its own, take immediate action against a sniper.)

Link posted by Doug Turnbull at 1:28 PM

Monday, October 21, 2002
A little wild speculation

Everyone else is putting forth their theories on the DC sniper, so why not me? Like an LA TV station jumping to the latest car chase, I’ll pile on the story of the hour, to increase ratings during the all important sweeps week.

So, what do the events of the weekend—another shooting near Richmond, and a mysterious message left there—tell us? First, from the locations of the shootings, it seems to me more and more likely that the sniper is a resident of Northern Virginia, rather than suburban Maryland. While the first shootings took place in Montgomery county, the two far away shootings, designed either to muddy the waters for investigators or strike outside the police dragnet, both took place in Virginia—first in Fredericksburg and second in Ashland. A Maryland based sniper, looking to stretch the police, would be more likely to strike up in Maryland or near Baltimore, it seems to me. Striking another major urban area would also increase the total terror sown, so to the extent frightening people gets this guy off, the fact that he hasn’t attacked near Baltimore could be telling. Whatever the reason, the “center of mass” of the shootings is definitely down in Virginia now, rather than up in Maryland.

Further evidence of the Northern Virginia connection comes from the fact that the two most recent shootings, which were both at night, took place in Virginia. A shooter might be less likely to stray really far from his home for night-time shootings, since the additional hour or more of travel time would be significant. A last, more tenuous link to Northern Virginia is the location of Seven Corners. That is an extremely confusing intersection—as the name says, there are 7 roads in and out of the area. If you haven’t driven through the intersection 3 or 4 times, your chances of actually ending up on the road you want to be on, going the way you want, are not that high. And more broadly, as others have noted, the road system in Northern Virginia is not that easy to follow for a non-native. Unless you’ve lived here for a while, it’s very easy to get lost, something the killer would presumably like to avoid. My wild guess is that the killer was up in Montgomery County for his job, and went on the first shooting spree, which seemed a little more haphazard and less planned out than the more recent attacks. Now, more worried about getting caught, he’s planning things better and sticking to places he knows.

Also worth noting is that the while the Ashland shooting breaks the weekend pattern, it doesn’t refute the retail worker (or other weekend worker) theory. Since it took place late at night, the shooter would have had plenty of time to get there after a full day of work.

Now, on to the mysterious message with a phone number that Chief Moose alluded to in his latest press conference. There are three possibilities for who left the message.

1. It was left by the shooter or an accomplice
2. It was left by a witness who has some info
3. It was left by an unrelated person who wants to jerk the police around

In reverse order, the third option seems unlikely, for a number of reasons. First, the police presumably rapidly closed off the area, so access would be limited. So someone would have had to be right there and immediately made a spur of the moment decision to write a fake note for the police. It also couldn’t have been premeditated, since no-one in Ashland would have suspected the sniper would strike there. Add in the prosecution of the false witness at the Home Depot, and I think this option can be ruled out.

Option 2 also seems unlikely. If a witness wanted to give info to the police, this seems like a very strange way to do it. There’s the 24 hour tip line if you want to be anonymous, and if not, why not just walk up to one of the 100 police officers on the scene that night rather than leaving a note?

That leaves either the killer or an accomplice. If the killer left the note, he presumably wants to taunt the police. The phone number seems an odd touch, though. Why would the killer actually want to directly talk to the police? His actions and any notes would seem to be sufficient taunting. But it is a possibility.

That leaves the last, and most intriguing possibility, that an accomplice of the killer left the note. This fits what little we know. The accomplice could either fear for his own life if he made any more explicit moves towards the police on the scene or at a roadblock. He might very well want to communicate anonymously, too, in order to arrange some sort of a plea bargain. He might want a deal in place before he turned himself in, rather than just taking the plunge and trusting the court system to behave kindly towards him.

It’s very possible that an accomplice could, at this point, be getting cold feet and figure it’s only a matter of time until they get caught. So better to arrange the capture to your own advantage, rather than dying in a shootout or getting sent to the chair. I wouldn’t be at all surprised if the killer was actually caught in one of the roadblocks around the Home Depot, but made it through, which could put some fear into either he or his accomplice or both. (Which would also explain the long delay until the latest shooting, and the remote location.) Unless the police were doing a complete, thorough search of every single car they let pass, it’s hard to see what good a roadblock could do. As long as the killer didn’t leave his rifle in plain view in the backseat, he’d be fine. Because if there’s one thing we can conclude with some confidence, it’s that the killer is not now driving a white van or truck, if he ever was.

Link posted by Doug Turnbull at 7:17 AM

A Renaissance blog: Politics, sports, literature, history, and whatever else strikes my fancy.


Sneaking Suspicions
Crooked Timber
Unqualified Offerings
Asymmetrical Information
Cranky Professor
The Poorman
newsrack blog

Powered by Blogger