The Samovar

Some sense on terrorism
February 11, 2009, 1:46 pm
Filed under: Civil Liberties, Politics, Terrorism

Reg article reporting on Nigel Inkster, former Assistant Chief of MI6:

There are limits to what we can sensibly aspire to…

Efforts to establish a global repository of counterterrorist information are unlikely ever to succeed. We need to be wary of rebuilding our world to deal with just one problem, one which might not be by any means the most serious we face.

We need to keep terrorism in some kind of context, for example, every year in the UK, more people die in road accidents than have been killed by terrorists in all of recorded history.

We should keep our nerve and our faith in our own values. Our own behaviour – especially with respect to the rule of law – is very important.

Comments Off on Some sense on terrorism

Fake terrorism as a commercial strategy?

The small news here in Paris is that five explosive packages (pleasingly, pains d’explosifs in French, which would literally be explosive breads) were found in the big Printemps department store on Boulevard Hausmann (link in English, more up to date link in French). The curious thing though, is that there were no fuses on the explosives. In other words, they couldn’t have detonated! And this isn’t the only curious thing about the case. The French news agency AFP received a letter from a group that was entirely unknown called the Front révolutionnaire afghan (FRA) which the antiterrorist police say hardly fits the style of these things at all. There are no religious references, and instead they use anti-capitalist language. Also, the type of explosive found (dynamite) isn’t typically used by Islamic terrorists.

So there are various theories about what is going on here. One possibility is that this is a warning, the message is that we have explosives and we’re not afraid to use them, so listen to our demands (removal of French troops from Afghanistan). The fact that the actions are atypical suggests maybe that this is not a group with ties to established terrorist organisations, but the work of a small group without ties or even a single individual. And this is probably correct.

However. There is another possibility. The question is: who has the most to gain from this? Who gains from one of the biggest department stores in France being effectively shut down for most of a day in the run up to Christmas? Surely the rivals to Printemps on boulevard Haussman, the Galeries Lafayette, who would most likely have got all the business that Printemps would have got during that day? I don’t know how much these shops would make during a day, but I’d imagine it’s really quite a lot, making it a risky but highly effective strategy.

Now I’m not suggesting that Lafayette did this, indeed it’s very unlikely. It would be incredibly risky for them, and anyway the note delivered to AFP says that the would-be terrorists are targeting the “grands magasins” (referring to both Printemps and Lafayette), and so this could potentially hurt both their profits. Still, it’s an interesting question: are fake terrorist incidents being created by companies to hurt their rivals and increase their own business? It seems like a strategy that would be too good to resist.

Comments Off on Fake terrorism as a commercial strategy?

Giddens: more rhetoric
July 21, 2007, 2:14 pm
Filed under: Civil Liberties, Politics, Risk, Security, Surveillance Society, Terrorism

A while ago, Anthony Giddens wrote a piece on terrorism and security that I replied to rather light heartedly. Others wrote more serious replies – see my previous entry for links. Today he wrote a rather odd piece on CiF replying to comments on his last “dozen or so articles”. Obviously since he was replying to so much, his comments were little more than a reaffirmation of what he’d already said, but for what it’s worth, here’s my reply to what he said about terrorism and security.

Whatever some of the bloggers want, Brown won’t commit electoral suicide by lurching towards the traditional left. Moreover, he is correct not to do so.

It’s worth pointing out that terrorism and security is not a left/right political issue. Authoritarians and civil libertarians exist at all points on the left/right spectrum. This is just misdirection – a complete red herring.

For instance, he owes it to citizens to make sure that they are protected against the threat posed by global terrorism. As I said in my article on the subject – written well before the latest attacks…

I love it. As if the fact that some utterly hopeless incompetents entirely failed to carry out what would have been a rare terrorist attack with a fairly small number of casualties (a few days worth of traffic fatalities at most) supported his argument. As if one could draw conclusions about risk and probability from a singular event.

– the debate about security in relation to civil liberties hangs a great deal upon how serious one believes the threat actually is.

At first when reading this I thought it was odd that he understands that the case must be based on the actual threat given that his argument was based on hypothesis and supposition, but then I look a closer look at the words he used. “The debate,” he says, “depends upon on how serious one believes the threat actually is” – not on how serious it actually is, or on the basis of evidence, but on the basis of belief.

It has to be analysed in terms of risk, a subject of some complexity, which I have studied in detail for many years.

Yessss!! I love it when they use appeal to authority. It’s especially delicious when their own argument undermines that authority (“written well before the latest attacks”).

Most of the blogs on this issue were hostile to what I said, but I stand by it. Taking high-consequence risks seriously,

But they’re not high-consequence risks. The largest terrorist attack ever killed under 3000 people. That’s no joke, but as I point out again and again, it’s absolutely tiny in comparison to so many other risks we face.

and mobilising against them, are the conditions of reducing them to manageable proportions, whether they be those associated with global warming, avian flu, world financial meltdown or international terrorism. The more seriously we take each issue, the less chance there is of a destructive outcome; but then those who disagreed with the policy in the first place will always say: “You were scaring us unnecessarily – look, nothing significant has happened.”

It is entirely right that the issue of civil freedoms should continue to be intensely debated. The level of risk should be monitored in a continuing way. One contributor asks, what will happen to freedoms that have been in some part suspended when the threat of terrorism recedes?

I would also add that the threat – such as it is – isn’t going to recede for a very long time, so we should take note that changes to our society based on the threat of terrorism have to be considered semi-permanent.

It is a very necessary question. There must be regular reports made to parliament, which can be scrutinised in detail; an independent role for the judiciary in making judgments has to be sustained; public debate must continue. How far anti-terrorist policies might produce an Orwellian state is itself a matter of risk assessment;

Certainly, if one is going to give the government and police powers which could be abused there should be independent scrutiny to minimise the dangers. The point is that when the state itself is the potential threat, you can’t rely on it to make reports to itself and supervise itself. The effectiveness of an independent method of scrutiny depends very much on the precise details of who exactly is doing it, what their relation to those in power is, what powers they have to investigate and overrule state decisions, etc. Can the judiciary be relied on for this sort of role? I’m not sure one way or the other. Either way, a better way of minimising the risk of abuse of powers is to not grant those powers in the first place, and to put practical obstacles in their way so that future governments cannot give themselves greater powers. Not building a surveillance infrastructure would be a good start.

but such procedures, robustly applied, will keep such an eventuality as the remotest of possibilities.

The remotest of possibilities? How remote is this possibility compared to say, the threat of an effective nuclear, biological or chemical weapons based terrorist attack upon which he based the entirety of his original argument? Presumably he thinks it’s much more remote, but what is the basis of this claim? While there have been no examples of successful such attacks despite much will to use them, there have been plenty of examples of governments that have turned bad based on manipulating the fears of the governed.

That some contributors talk as though such a state is already here, while dismissing new-style terrorism as offering no significant threat, strikes me as absurd.

Comments Off on Giddens: more rhetoric

Function creep: surveillance in London
July 17, 2007, 1:42 pm
Filed under: Civil Liberties, ID Cards, Politics, Security, Surveillance Society, Terrorism

Function creep is a very useful concept for understanding government and surveillance. When a new technology is introduced to do one thing (one function), and is later used for an entirely different thing, that’s function creep. It often seems as though governments plan to bring in potentially unpopular technologies by exploiting function creep. It goes like this: the government wants to do X where X requires some new and expensive technology Y. Unfortunately for them, X is fairly unpopular and if everyone knows that they’re spending money on Y in order to do X then there’ll be a huge fuss about it in the papers. So what they do is invent a new and popular thing Z that also requires the technology Y. When they’re building Y they say it’s for Z, but all the time they have in the back of their mind that they’ll introduce X later on.

Function creep is one reason why civil liberties campaigners are so worried about ID cards. The government plans to introduce them as a non-compulsory thing which will only be used in ways that are useful to most people, or for purposes that are popular (like being nasty to immigrants, or catching terrorists). It won’t actually do those things effectively, but that doesn’t matter because that’s not what they’re really for. It’s really there to build a large database on everyone to make the job of the civil service and police that much easier, and it may also undergo function creep in the future to make it compulsory to have one, and maybe later than that to make it compulsory to always carry it, etc.

Today the BBC reports an interesting example of function creep in London.

Police are to be given live access to London’s congestion charge cameras – allowing them to track all vehicles entering and leaving the zone.

The reason given is terrorism:

Home Secretary Jacqui Smith blamed the “enduring vehicle-borne terrorist threat to London” for the change.

There is function creep going on at many levels here. The first is that an infrastructure of cameras built to help manage congestion in London is now going to be used for routine surveillance by the police. Would we have agreed to a network of cameras being built in order to spy on us all the time? Almost certainly not, but they can just apply function creep to a system that’s already there. In this case, it was almost certainly opportunistic rather than planned function creep.

There’s also a hint as to some planned function creep:

But they will only be able to use the data for national security purposes and not to fight ordinary crime, the Home Office stressed.

In other words: don’t complain about this on civil liberties grounds, we’re only going to use it on terrorists. For the moment.

This is suspect for two reasons. First of all, they might change their minds about it in the future. Alarm bells should be going off when they reassure us it won’t be used to fight ordinary crime, given that the actual dangers associated to ordinary crime are so much larger than the negligible threat of terrorism. Secondly, because they’re already using terrorism laws in ordinary police work:

Since 2001, some 436 people have been charged in relation to terrorism investigations. Almost 200 of these were under standard criminal offences such as conspiracy to murder.

And let’s not forget Walter Wolfgang, the Labour party member who was kicked out of the party conference and detained under anti-terrorism legislation for shouting the word “Nonsense!”.

To finish off with, the article also makes a passing reference to an earlier function creep:

Although charges are only in force at peak times, the system runs 24 hours a day, a TfL spokesman said.

In other words, the system was already being used as a de facto surveillance infrastructure – running when it had absolutely no need to in order to carry out its stated and original function.

Update (18 July 2007): And for anyone who thought I was being paranoid, only one day later plans to extend this scheme nationwide for use in fighting ordinary crime were leaked to the Guardian. SpyBlog has more in depth coverage.

The past few days
July 3, 2007, 12:17 am
Filed under: Civil Liberties, Politics, Risk, Terrorism

I’ve been away which is why I haven’t written anything about terrorism. I don’t like to blow my own trumpet (OK I do), but I’d just like to point out that on June 15th I wrote:

… This suggests that the number of willing would-be terrorists is actually fairly small, or that they are generally incompetent.

The recent technically incompetent and hopelessly executed attempt at a terrorist attack bears this theory out:

Earlier reports said bouncers from a nearby nightclub saw the car being driven erratically before it crashed into a bin. They claimed the driver then got out and ran off.

In case these attacks are used to justify further incursions on civil liberties, it’s worth pointing out that these people were wholly unknown to the police, that increased detention without trial or other similar powers would not have been useful in stopping them, etc. In other words, existing powers were more than sufficient.

Fortunately, it seems that most people are not giving this attack undue significance. Leninology notes that BBC News 24 were using the phrase “terror fatigue” to describe people’s lack of reaction. Well… at last!

Comments Off on The past few days

Overactive imagination: Giddens on risk
June 19, 2007, 1:44 pm
Filed under: Civil Liberties, Politics, Risk, Security, Terrorism

Anthony Giddens has a piece on CiF explaining that we need to crack down on freedom because of the danger of nuclear terrorism. It’s funny and kind of gratifying that this should appear just a couple of days after I posted a long piece arguing the exact opposite. Unusually, the CiF commenters do a fine job of demolishing his argument – to my amazement there didn’t seem to be a single one supporting him.

What’s nice about this piece though is that he explicitly makes all the mistaken arguments rather than concealing the foolishness of his argument.

First, it cannot be known in advance with certainty how great the risk really is. Second, the consequences are potentially cataclysmic, so we have to bend our efforts to preventing them, rather than picking up the damage afterwards. Third, how we respond to the risk – how seriously we take it – affects the very nature of that risk.

I wonder if he’s familiar with Pascal’s wager? This is the argument that says:

  1. There is a non-zero probability that God exists, call this probability p.
  2. If God exists and we believe in Him the rewards are infinite, \infty.
  3. If God exists and we don’t believe in Him, the punishment is infinite, \infty.
  4. If God doesn’t exist and we believe in Him, the reward is zero, and the cost of the belief is finite, call it C.
  5. If God doesn’t exist and we don’t believe in Him, the rewards and costs are zero.
  6. Consequently, on average, your expected reward-cost for believing in God is p\infty-(1-p)C=\infty (because \infty*something non-zero is \infty, and the cost C is finite).
  7. On average, your expected reward-cost for not-believing in God is -p*\infty=-\infty.
  8. Therefore, on the basis of cost-benefit analysis, we should opt to believe in God.

One problem with this wager is that it misses out on the question of which God to believe in. I can hypothesise an infinite number of different possible Gods, each of whom will infinitely punish you if you believe in any of the other ones. So which God do you opt for? Oh nyoe! We’re all doooomed.

And so, I invite readers to imagine other creative disasters we should be worrying about, and what we should be doing about them. I’ll start:

There is a danger that particle accelerators might produce a rogue particle that could destroy all matter in the universe. The chance is low, but the consequences would be infinitely disastrous. Hence, we should shut down all particle accelerators, cancel all teaching of physics, and what they hey, lock up or execute anyone with any knowledge of particle physics.

Update: Unity has a (long) post on this too, as does UK Liberty. Also, some of the CiF commenters came up with the same game as me. Suggestions so far include the threat of alien invasion and how we all should join the army and learn firearms for when it happens; and more pertinently (but less funny), the threat of a future government turning fascist, and how we should therefore not undermine civil liberties. Another person pointed out – if you really take the threat of nuclear terrorism destroying our civilisation seriously, we ought to be teaching people how to survive in a post-nuclear-holocaust society. Unless you think that new terror powers are almost certain to stop the terrorist nuclear bombs, you should consider this as just as high priority.

Terrorism: Hysteria, Control, Profit
June 15, 2007, 2:38 am
Filed under: Civil Liberties, Manifesto, Media, Politics, Risk, Security, Security Theater, Terrorism

We shouldn’t be afraid of terrorism.

This is something I have written about before, but never really set out my reasons in detail.

We shouldn’t be afraid

John Mueller summarises:

… the number of people worldwide who die as a result of international terrorism is generally only a few hundred a year, tiny compared to the numbers who die in most civil wars or from automobile accidents. In fact, until 2001 far fewer Americans were killed in any grouping of years by all forms of international terrorism than were killed by lightning. And except for 2001, virtually none of these terrorist deaths occurred within the United States itself. Indeed, outside of 2001, fewer people have died in America from international terrorism than have drowned in toilets.

Even with the September 11 attacks included in the count, however, the number of Americans killed by international terrorism since the late 1960s (which is when the U.S. State Department began its accounting) is about the same as the number killed over the same period by lightning—or by accident-causing deer or by severe allergic reactions to peanuts. In almost all years the total number of people worldwide who die at the hands of international terrorists is not much more than the number who drown in bathtubs in the United States.

Mueller plots this data (colours added by me to make it look better in the tiny picture):


There is a massive spike in 2001 as a consequence of the WTC attacks. Data after 2003 is slightly more difficult to come by because after 2001 the methods of collecting terrorism data changed, and some claim that they were inflated. The NCST give this shiny graph of monthly fatalities from terrorist attacks from 1998 to 2004:


The MIPT data gives this for worldwide annual terrorism fatalities 2000-2006:


For the US and Western Europe only, they give this data (I’ve not included the roughly 3000 WTC attack deaths in 2001 so that the scale of the graph is right and you can see the detail more clearly) for 1968-2007:


The graphs indicate that terrorism is on the rise, but even taking the most pessimistic view of things the numbers are still microscopic on a global or even national scale. For comparison, in the US there are 40,000 road deaths a year, in the UK there are 3000 a year.

The upshot of all this is that based on the evidence of what has happened so far, the chance of being killed in a terrorist attack is negligible. If you want to worry about being killed by something, terrorism should be one of the last things on your list.

But what if things got much worse?

It’s possible that terrorist actions could get worse in three ways:

  1. they could become much more frequent
  2. they could get much more effective
  3. they could get hold of nuclear, biological or chemical weapons

The first scenario seems to be unlikely. Terrorist attacks are actually quite easy to pull off, and yet there aren’t very many. In America, you can buy high powered guns legally. All a terrorist would have to do is buy one and go crazy in a shopping mall. If you had a large number of potential terrorists, this would be a highly effective strategy but it hasn’t happened. This suggests that the number of willing would-be terrorists is actually fairly small, or that they are generally incompetent. There’s also no particular reason to think that attacks should become much more frequent than they are. The Iraq war has certainly done a lot to increase the number of terrorist attacks worldwide, but even doubling and tripling these numbers leaves the danger tiny.

The second scenario also seems unlikely to bring any great change. Increased numbers of terrorists might mean an increase in expertise and so an increase in effectiveness, but on the basis of what we’ve seen so far, this wouldn’t make an enormous difference. The largest and most effective terrorist attack ever was the WTC attack which killed some 3000 people, but far from being an indication of a new and more effective terrorism, it was more likely a freak event that was much more effective than expected. In a video released in December 2001 (although not everyone believes it is genuine), Osama bin Laden says:

(…Inaudible…) we calculated in advance the number of casualties from the enemy, who would be killed based on the position of the tower. We calculated that the floors that would be hit would be three or four floors. I was the most optimistic of them all. (…Inaudible…) due to my experience in this field, I was thinking that the fire from the gas in the plane would melt the iron structure of the building and collapse the area where the plane hit and all the floors above it only. This is all that we had hoped for.

In other words, what they expected to achieve with the attack was a tiny fraction of what they actually managed to achieve. That said, it could also have been much worse than it was if more people had been in the buildings for example. Along these lines, it might happen that at some point in the future another terrorist plot happens to succeed as well or better than this. It’s unlikely, but even if it did happen, it still probably wouldn’t make an awful lot of difference. To put it in perspective, a disaster on the scale of 9/11 would have to happen once a month before the number of deaths was the same as the number who die in car accidents every year in the US.

So the only possible scenario we really need to worry about is that terrorists might get hold of nuclear, biological or chemical weapons. This is not something I know anything about, so here I’ll just quote Mueller:

Chemical arms do have the potential, under appropriate circumstances, for panicking people; killing masses of them in open areas, however, is beyond their modest capabilities… Biologist Matthew Meselson calculates that it would take a ton of nerve gas or five tons of mustard gas to produce heavy casualties among unprotected people in an open area of one square kilometer. Even for nerve gas this would take the concentrated delivery into a rather small area of about three hundred heavy artillery shells or seven 500-pound bombs. A 1993 analysis by the Office of Technology Assessment of the U.S. Congress finds that a ton of sarin nerve gas perfectly delivered under absolutely ideal conditions over a heavily populated area against unprotected people could cause between three thousand and eight thousand deaths. Under slightly less ideal circumstances—if there was a moderate wind or if the sun was out, for example—the death rate would be only one-tenth as great. Although gas was used extensively in World War I, it accounted for less than 1 percent of the battle deaths. In fact, on average it took over a ton of gas to produce a single fatality.

Properly developed and deployed, biological weapons could indeed (if thus far only in theory) kill hundreds of thousands—perhaps even millions—of people. The discussion remains theoretical because biological weapons have scarcely ever been used. Belligerents have eschewed such weapons with good reason: biological weapons are extremely difficult to deploy and to control. Terrorist groups or rogue states may be able to solve such problems in the future with advances in technology and knowledge, but the record thus far is unlikely to be very encouraging to them. In the 1990s, Aum Shinrikyo, a Japanese cult that had some three hundred scientists in its employ and an estimated budget of $1 billion, reportedly tried at least nine times over five years to set off biological weapons by spraying pathogens from trucks and wafting them from rooftops, hoping fancifully to ignite an apocalyptic war. These efforts failed to create a single fatality—in fact, nobody even noticed that the attacks had taken place. For the most destructive results, biological weapons need to be dispersed in very low-altitude aerosol clouds: aerosols do not appreciably settle, and anthrax (which is not easy to spread or catch and is not contagious) would probably have to be sprayed near nose level. Explosive methods of dispersion may destroy the organisms. Moreover, except for anthrax spores, long-term storage of lethal organisms in bombs or warheads is difficult and, even if refrigerated, most of the organisms have a limited lifetime.

Nuclear weapons, most decidedly, can indeed inflict massive destruction, and it is certainly reasonable to point out that an atomic bomb in the hands of a terrorist or rogue state could kill tens of thousands of people. But it may also be worthwhile to note that making such a bomb is an extraordinarily difficult task and that warnings about the possibility that small groups, terrorists, and errant states could fabricate nuclear weapons have been repeatedly uttered at least since 1947, and especially since the 1950s when the ‘‘suitcase bomb’’ appeared to become a practical possibility. It has now been three decades since terrorism specialist Brian Jenkins published his warnings that ‘‘the mass production and widespread distribution of increasingly sophisticated and increasingly powerful man-portable weapons will greatly add to the terrorist’s arsenal’’ and that ‘‘the world’s increasing dependence on nuclear power may provide terrorists with weapons of mass destruction.’’

So why does nobody say this?

Some people have said this sort of thing, but not many and rarely people in prominent public positions. The mayor of New York, Michael Bloomberg, recently got in trouble when he said:

There are lots of threats to you in the world. There’s the threat of a heart attack for genetic reasons. You can’t sit there and worry about everything. Get a life…. You have a much greater danger of being hit by lightning than being struck by a terrorist.

US Senator John McCain wrote:

Get on the damn elevator! Fly on the damn plane! Calculate the odds of being harmed by a terrorist! It’s still about as likely as being swept out to sea by a tidal wave. . . Suck it up, for crying out loud. You’re almost certainly going to be okay. And in the unlikely event you’re not, do you really want to spend your last days cowering behind plastic sheets and duct tape? That’s not a life worth living, is it?

So why is the perception of the terrorist threat so out of proportion to the real threat?

Well, one reason is that we’re very bad at estimating risk. We tend to overestimate the significance of dramatic and unusual events. News reporting multiplies this effect, because by its very nature it reports the news, the unusual, not the run of the mill risks we face every day which are actually much more significant. This begins to explain it, but it doesn’t really excuse our ignorance. Terrorism has been a major world issue for almost 6 years now, plenty of time for a more realistic view to have gained acceptance.

There are all sorts of reasons why it might not have, but the one I want to focus on is that it isn’t in the interests of anyone who might have been able to lower the perception of the terror threat to have done so.

The media aren’t exactly in the business of making things sound less dramatic and exciting than they really are. It is in their interests to exaggerate the threat. Perhaps more importantly than that, it is certainly not in their interests to portray the threat rationally and calmly. How boring would that be?

Similarly, politicians have much to gain from exaggerating the threat, and much to potentially lose by being reasonable about it. If they told everyone that everything was OK and they needn’t worry, what would the reaction be next time there was a terrorist strike? It wouldn’t matter if there were only a few casualties, fitting the general prediction that terrorist attacks will occur but won’t add up to anything like the number killed by everyday threats like cars and bathtubs, they would immediately be ridiculed and voted out. Not only that, but in the immediate aftermath of an attack, doing nothing is simply not a politically viable option, even if it is actually the least worst one. Actually, I’m not at all convinced that this need be so, but being reasonable about the threat, for a politician, is decidedly the more personally difficult and risky approach.

Politicians also rely on their advisers considerably, and these advisers are typically unelected and have much to gain from the increased control that the threat of terrorism can buy. Increased powers make the job of the police easier. Huge databases on us make the jobs of civil servants easier. Increased funding makes the jobs of the secret services easier. They have everything to gain from exaggerating the threat, and the politicians listen to what they say.

Finally, there is a whole sector of the economy that produces technology for fighting the ‘war on terror’. Billions of pounds are spent on this technology, and so the companies involved have very good reason, and ample resources to spend on lobbying and exaggerating the threat.

A word on policy

Most of what the government does in response to terrorism is actually ineffective or even counterproductive. Not only are they wasting billions on this and destroying civil liberties that define the nature of our society, but they might even be making us less safe.

A couple of examples of this. In the few months after 9/11, many people in the US would drive around the country instead of taking internal flights. As a consequence, it was estimated that 1000 more people died in traffic accidents than normal for that period of time. Bad advice to avoid planes and our inability to assess risks accurately cost 1000 lives, more than a third of the death toll of 9/11 itself.

After the Tube attacks in London, a policy of random searches of bags was instituted on the New York subway. On an average day, there are 4m journeys on the NY subway, so the chance of catching a terrorist if there were one (and there hasn’t been one yet) would be tiny. Could those police officers have been used more effectively in fighting ordinary crime? Nobody has done the calculations, but it seems likely.

Some have argued that although these measures are strictly speaking quite ineffective, they might have a deterrent effect. Terrorists might be put off from trying to blow us up because of the danger of getting caught. But is this really a sustainable point of view? Someone who is willing to blow themselves up in killing a few people would be put off by a tiny chance of having their bag searched? What would they have to lose? Even if their bag was searched, they could set off their bomb as the police officer approached.

What about the extreme risks associated to, say, a nuclear terrorist attack. Surely it’s worth doing anything we can to avoid that? The thing is, for new laws and powers to be effective, two things would both have to happen. Firstly, the terrorists would have to have the expertise and materials in the right place to pull of a nuclear attack that worked. That’s unlikely, as we’ve already seen. On top of that, the police would have to be able to stop this plot using the new powers, but not using the old powers. Bruce Schneier claims that in all cases of foiled terrorist plots so far, they were foiled using traditional methods of investigation rather than new anti-terror powers. Most of the time both of these things aren’t going to happen – either the terrorists wouldn’t be able to pull off the nuclear attack, or the police would have been able to foil the plot using the powers they already had.

Let’s think about some numbers. There’s no way of working out these probabilities exactly, so these numbers are just to give an idea. Suppose there’s a 1 in 100 chance that the terrorists had the materials and expertise to pull off a terrorist attack, and that there’s a 1 in 100 better chance that the police can stop it with the extra powers. This would mean that these powers would be effective only 1 in 10,000 times. If the chances were both 1 in 10 rather than 1 in 100, the new powers would be effective only 1 in 100 times.

The point is that when we’re thinking about whether or not to pass new laws or grant new powers to the police, we shouldn’t be thinking that these laws will stop a nuclear attack therefore they are justified even though they’re extreme, we should be thinking that these new laws might help to stop a relatively unlikely terrorist attack that might have been stopped anyway.

Is that worth undermining the nature of our society for?

Absolutely not.

(Note: this entry was edited slightly from the original posting to improve the last section, but no substantive changes were made.)

It’s back! and this time it’s still bullshit
June 7, 2007, 4:02 pm
Filed under: Civil Liberties, Politics, Security, Terrorism

John Reid wants to keep people in jail without trial for three months – one month just doesn’t seem to do it for him. Gordon Brown seems to agree.

This is the wrong thing to do for the same reasons that it was wrong last time – neatly summed up by David Davis “We do not protect our way of life by undermining our way of life”. The threat of terrorism is fairly small (about the same as the threat of drowning in your bathtub or being struck by lightning), and in a democratic society policy is not decided by what the police want or think they need – that’s a police state.

I wonder if we’ll have such good arguments this time round as last time?

Janet Anderson (Rossendale and Darwen) (Lab): … In their recent evidence to the Joint Committee on Human Rights, the police referred to one case in which, if they had printed out the computer data that they had recovered, it would have made a pile 66,000 ft high. That is the sort of challenge that the police face and they need our help to meet it. (Hansard, 9th Nov 2005)

ZOMG! Thousands of feet high? That’s millions! We need to give the police 3 years of detention without trial, not 3 months!

Lib dem peer Lord Carlile kicks off this round with his active imagination:

I can imagine that there may well come to be cases – and I’m not saying that there have been any yet – in which the need to protect evidence, to discover what the evidence is, to de-encrypt computers, to find people may not be achieved within 28 days.

Ah, ‘de-encryption’, if only there was a word for that. Let’s see now, hardware-based decryption of a 128-bit key was estimated to take 11,000,000,000,000,000,000 years in 2005 (excitingly, down from 84,000,000,000,000,000,000 years in 2000). I hope Lord Carlile will be pushing for this length of detention without trial.

Whatever happens, I’m sure we’ll be able to trust the BBC to report accurately on it.

Fact of the day
May 3, 2007, 11:51 pm
Filed under: Politics, Risk, Terrorism

In the three months following the WTC attacks, over 1000 people died because they travelled by car instead of by plane.

(From this interesting article.)

Comments Off on Fact of the day

Rational responses to terrorism
April 30, 2007, 7:05 pm
Filed under: Politics, Terrorism

When I say that we are over-reacting to the threat of terrorism, I am occasionally challenged to explain what an appropriate reaction would be, how we can quantify it, etc. Documents such as Europol’s EU Terrorism Situation and Trend Report (via Lenin’s Tomb) are a good start. (See also this article for a view of the situation in the US.)

The tables below list all the terrorist incidents in 2006 (the first one), and the number of arrests made (the second one). Sorry I’ve included them as images because WordPress seems to have difficulty with copying and pasting tables in and I didn’t want to write them out by hand.

The first thing to notice is that of 498 terrorist attacks, only one was identified as ‘Islamist’. On the other hand, close to half of the arrests made were of people suspected of ‘Islamist’ terrorism. On the other hand, ‘Separatist’ terrorism (such as ETA in Spain) accounted for 424 of the 498 attacks, but accounts for less arrests in total than of ‘Islamist’ terrorism.

I think it’s reasonable to conclude that our anti-terrorism resources are not being used efficiently.

UPDATE (2nd May 2007): Actually this conclusion is not such a reasonable one to make because the statistics below don’t tell you about the number of casualties per incident. See the comments for more details.

The tables and the rest of the report make interesting material to study. Most of the 2006 incidents were due to Basque separatists in Spain and Corsican separatists in France. Unfortunately, the report doesn’t give figures for the number of casualties, wounded, and the amount of property damage caused by these incidents, but it does repeatedly refer to the fact that casualties were usually low. Given this, it becomes appropriate to consider terrorism as an economic cost, and therefore the resources given to combat it should be proportional to the damage caused. Similarly, restrictions on civil liberties, legislative changes, etc. cannot be justified based on the costs of terrorism compared to other forms of damage to society (cars, pollution, etc.).