No Defense is Perfect
It's great to be back at TPMCafe. And it's a pleasure to have the chance to discuss and debate my new book, On Nuclear Terrorism, which came out just last week.
To understand how to confront nuclear terrorism, we need to get inside as many nuclear plots as we can. There's only one catch: no real-world terrorist attempt at a nuclear attack has ever gotten very far. Without reality to anchor us, we tend to conjure fantastic terrorist schemes, obsess over worst-case scenarios, and demand perfect defenses. That distorts our thinking about how to defend against nuclear terrorism. My book is a guide to kicking that habit.
In 1984, the IRA came within a whisker of killing Margaret Thatcher in her hotel room at Brighton's Grand Hotel. The day after, it issued a statement: "Today we were unlucky, but remember; we only have to be lucky once. You will have to be lucky always." That maxim has become popular amongst our political leaders in recent years.
If you look inside nuclear terrorism, though, the exact opposite becomes true. Move from stage to stage of a nuclear plot and you will discover a host of ways that a terrorist group might fail. That turns the old saw on its head. We only have to be lucky once for a plot to fail. To succeed, a terrorist group must be lucky always.
Each nuclear plot three basic pieces. A group needs to acquire either a nuclear weapon or the nuclear explosive material (highly enriched uranium or plutonium) required to make a bomb. It has to either activate its weapon, or, if it has only gotten its hands on nuclear materials, it needs to build one. And it needs to move the weapon to its intended target before detonating it. Each of these basic tasks, in turn, has many moving parts. Some might be technical, like designing a weapon or hiding one from detectors; others, like recruiting scientists or sailing a bomb into a port, would be demand non-nuclear skills. Put this all together and you've got a feasible but complicated terrorist task.
On the flip side, you have a host of opportunities for confronting terrorist plots.
This gets more interesting when you mix in what we know about how terrorist leaders make decisions and about how terrorist groups operate. The most ambitious groups, terrorism scholars tell us, tend to be the most afraid of failure. This means that defenses need not be anywhere near perfect to work.
Imagine an approach to border security that only has a one-in-four chance of stopping a terrorist plot. Pretty useless, right? Not necessarily. Those odds of failing seem grossly inadequate from where we sit. But put yourself in the shoes of a terrorist leader contemplating a nuclear strike, add up all the other ways you might fail, and you might easily come to a different conclusion.
Terrorist psychology gives us another assist. We tend to imagine terrorists as endlessly nimble and innovative. But many are not. Afraid of failure, they are conservative in their operations, changing as little as possible from plot to plot. Even the 9/11 attackers stuck to their tried and true tactics for moving money and recruiting people.
We should keep this in mind as we design defenses. Here is how the process usually goes: someone proposes a defense; someone else points out a way that a terrorist group can get around it; everyone goes back to square one. (I exaggerate a bit, but it's close.) The problem is that there's a big difference between what a group could do and what one actually would do.
Take al Qaeda. A defense that focused on variations on that group's typical tactics might invest effort in intercepting nuclear materials at official entry points to the United States (al Qaeda has hardly ever used illicit crossings) and on making it hard for relative amateurs to build nuclear bombs (the group is historically wary of recruiting outside experts to assist in its plots.) It is easy to imagine ways to evade these defenses, but that might not be as important as we intuitively think.
None of the approaches I've just described will deliver perfect defense. But anyone who promises perfection is fooling you. And since some chance of a nuclear attack will always remain, the second goal of any strategy should be to minimize the consequences of any plot that succeeds. This is probably the one point in my book that provokes the greatest discomfort.
Most obviously, it means that we must prepare to save lives after any attack. But it also means that forcing a terrorist group to detonate a bomb at the border, rather than in the center of a city, has real value. A defense that makes it hard to recruit skilled engineers might do something similar, if a terrorist group ended up settling for a less effective weapon. Either of these might cut the number of people killed by an attack from hundreds of thousands to a few thousand instead. Yes, that is still an unacceptable outcome. But saving a hundred thousand lives is not something to be waved away.
From these foundations, the book spins out a host of observations and recommendations, and I trust that my fellow book clubbers will dredge a few up and have a go at them before the week is out. Indeed I'd be disappointed if they didn't. My goal in writing this book was not to dictate a solution – it was to get people to think differently about the problem. That's essential if we are to have a smarter, broader, and more realistic debate about how to deal with one of the most pressing security challenges facing the world.
Two huge differences exist that separate imagined nuclear terrorism from "normal" terrorism. One is the extreme difficulty of acquiring the materials, the expertise, the assembly capacity, and the delivery. The other is the consequences.
We tend to think of the question in the same terms as terrorism that employs knives, guns, or chemical explosives. Can we stop for a second and note that those items are pretty much everywhere?
In the US, McVeigh did not need any permit to buy fuel oil and fertilizer. In Iraq, explosives are ubiquitous. And there is a truly enormous number of assault rifles circulating in the world. So forget that now---consider uranium.
It can't be used out of the ground---it is not found in useful nuggets like gold. So one has to steal from the supply operation at some point. And since it's not really useful, even for the nasty but relatively trivial dirty bomb, one has to get it after enrichment. Commercial enrichment won't do, so one must steal military uranium. Not really much of that just lying around, actually.
And if terrorists found a source for highly enriched U, they would need fifty pounds or so. (Someone was just caught with a pound of low-enriched.) A bomb would be easy to make if they got this far, but would weigh as much as a modest artillery piece, needing a truck to deliver.
Plutonium makes for small bombs, but a Pu bomb is trickiest piece of machinery imaginable. And plutonium also is not found in the ground, but must be stolen from a reprocessing site or weapons plant.
In other words, the choke point is supply. Only the actual sale or theft of a working weapon is a major worry. Let's now consider that. Working nuclear weapons are the highest achievement of industrial art, a huge expense, and the crown jewels of national defense. How is one going to buy one? Who would sell the national treasure? Who would let someone sell the family atomics? Not time to panic over that, surely. No country, whether Pakistan, or Libya, would ever give control over a weapon to anyone. But let's set that aside.
Imagine you now have a weapon that is in place and ready to trigger. Here's the question---how likely is it that anyone with the intelligence, resources, persistence, and allies to get to that stage will be insane enough to light it off? Even when nuclear states ran defense drills, their staff, (or in one case. Gorbachev), were too often unwilling to act.
There seems to be a significant concentrating-of-the-mind when people consider unleashing destruction on that scale. I think when we consider that most people are not very eager to release that djinn, and that it's difficult to actually deliver such a bomb, and that it's even harder to make one, and that it's really hard to get the materials, we're probably doing the right things now, and can stop spending any time worrying about it regarding policy.
December 3, 2007 7:16 AM | Reply | Permalink
I always enjoy reading the essays from physicist Gordon Prather and he's got a good one here: Deck the Malls with Dirty Nukes.
ecotourism
WeGoEco.com
December 3, 2007 7:35 AM | Reply | Permalink
Thanks. I got the bare-sphere number wrong for U-235, which is 55 Kg, not pounds. Point being one needs a lot of a very precious metal, and the resulting bomb is big and heavy.
December 3, 2007 7:53 AM | Reply | Permalink
Cool. Sounds like an interesting book. At least there is some sanity out there. The question will be whether you can get the public to listen. Nuclear terrorism has been and will be the favored method of fear-mongering by right wing demagogues. May be this book is simply for policy makers, but it's better than nothing.
December 3, 2007 8:02 AM | Reply | Permalink
I think we should be concerned about terrorists building a giant "laser" on the moon, turning it into a "death star."
thosethingswesay.blogspot.com
December 3, 2007 8:12 AM | Reply | Permalink
Sounds wonderful. I'd only add a point to an already pointed sentence: "Without reality to anchor us, we tend to conjure fantastic terrorist schemes, obsess over worst-case scenarios, and demand perfect defenses." Those fantasies, obsessions, demands, and lack of anchor to reality are actively foisted upon us by the right, which depends on fear to retain influence with the electorate. And of course the lack of relation to reality is not often called into question by the media.
John
http://www.haberarts.com/
December 3, 2007 9:39 AM | Reply | Permalink
That's more the figure for a very, very crude gun-type uranium bomb. Now, plutonium lends itself better to miniaturization, as long as there isn't too much Pu-240 in the material; Pu-239 is the usual fissionable isotope.
Uranium will work in the simpler gun, but plutonium must be an implosion design -- uranium can as well. The challenges in getting the amount of material down and the yield up are complex.
One basic principle is that you need to get as much mechanical energy, from the implosion high explosives, as possible. There are lots of techniques here, but they tend to need to be tested. It isn't strictly necessary to have any nuclear yield, although you might have a tiny and undetectable one, to get the compression to be more efficient -- but the problem is that you need what is called a hydrodynamic or hydronuclear test facility, which has X-ray cameras fast enough, and rugged/protected enough, to take multiple time-lapse pictures as the high explosives go off and compress. So, the instrumentation for tuning, as well as the fissionables and other special materials, are not exactly off the shelf.
--
Howard
*equal opportunity offense to both extremes*
"Those who cannot remember the past are condemned to repeat it" [George Santayana]
December 3, 2007 9:48 AM | Reply | Permalink
Are radiologic weapons in the scope of this discussion? While they aren't likely to produce massive effects, they are good fear weapons and much easier to build than anything that goes boom (in a fission or fusion manner).
There are steps that can be taken. Cesium-137 is widely used as a gamma source, but it's actually inferior to other isotopes that are safer. The poisoning incidents in the UK have made people question if polonium-210 is too available, but that's something of a scare issue -- it's an alpha emitter and you pretty well have to drink or inhale it.
Here's a link to an incident of cesium-137 contamination due to negligence, which might give some baseline. A worst case would be a finely ground powder in an air conditioning system, although it's not going to float very well. Making a fine enough powder that won't cling to itself is a technical challenge, and, without protective equipment, would probably be lethal to the people working on it. Grisly as it may sound, one strong potential early warning of a radiologic or even nuclear improvised attack being planned is a corpse, or victim, of radiation sickness.
--
Howard
*equal opportunity offense to both extremes*
"Those who cannot remember the past are condemned to repeat it" [George Santayana]
December 3, 2007 9:52 AM | Reply | Permalink
Since there's another thread in process (although no guest post as yet), here's a link to my more bomb-oriented post there.
--
Howard
*equal opportunity offense to both extremes*
"Those who cannot remember the past are condemned to repeat it" [George Santayana]
December 3, 2007 9:55 AM | Reply | Permalink
I seem to recall that when Rocky Flats was decommissioned, they found a lot of Plutonium in the vent system. Does anyone know the details? In particular, how many people got radiation (or, in the case of Pu, chemical) poisoning?
That might give us a clue as to the potential for a building HVAC system used to spread radiation.
December 3, 2007 2:47 PM | Reply | Permalink
Building a nuke is complicated. We know that. A dirty bomb is much easier to build. Just wrap radioactive material, medical X-ray and nuclear medicine waste will do, around some dynamite or ammonium nitrate and kerosene and blow it up. It won't level a city like Hiroshima was leveled, but everything within a radius of the blast will be contaminated to some degree. Maybe a President Giuliani would declare the area safe so that business could continue, meanwhile people develop cancers.
Not mentioned was nukes arriving via shipping container. Heavy things are expected in containers. Only a small fraction of them are inspected. Bush has decided that this isn't important in the "War on Terror" by cutting funds for port security.
I don't mean to be a fearmonger. We still face a greater risk driving our cars. And there's always the John Ritter phenomenon. Out of the blue- you're dead.
December 4, 2007 10:18 PM | Reply | Permalink
Correct that container is a worry, especially in that it need not have to come into port to be a threat---an offshore bang would be pretty effective.
The good news is that, as discussed above, all true nuclear explosives are difficult in one way or another. And the dirty-bomb scenario implies a dectectable radioisotope, so it is in principle discoverable. Uranium by itself is a lousy dirty bomb.
There is a decent amount of professional literature on the subject and I think the consensus is that it is less of a challenge than it seems, to both defend against and deal with afterward.
December 5, 2007 6:49 AM | Reply | Permalink
It does give one caise for concern when I think of a friend of mine who made a nuclear bomb collecting the radium from a bunch of discarded Timex watches with two cherry-bombs as a detonator.
December 5, 2007 12:58 PM | Reply | Permalink
Not.
December 5, 2007 1:02 PM | Reply | Permalink