Swindon Borough Council are considering getting rid of their speed cameras. Roderick Bluh, the Council Leader, offered a truly awful argument for doing so on today’s BBC lunchtime news:

When you fine a motorist for speeding, he’s already been speeding. Would it not be best to invest and make sure that speeding is prevented in the first place?

There’s nothing wrong with the idea that it’s better to prevent a crime than to punish one. If we can implement measures (e.g. advertising, improved signage, etc.) to encourage drivers to slow down, then that’s great. However, whatever measures we take to reduce speeding are only going to be partially successful; there will still be some people out there who break the speed limit. So what should we do about them?

There’s no reason why, having done all we can to prevent people from breaking the law, we can’t also try to catch and punish people who break the law. The first problem with Bluh’s argument is therefore that it restricts the options, trying to force us to choose between prevention and punishment when we can in fact have both.

The second problem is that it ignores the fact that speed cameras act as a deterrant. If you know that if you speed you’ll get caught and fined, then there’s a good chance that you’ll slow down. Bluh’s argument thus generalises from the fact that speed cameras don’t stop some crimes (those that they detect) to the idea that they don’t stop any crimes. Speed cameras may not stop people who are fined from speeding, but they do stop plenty of people who aren’t fined from speeding.

Even without those problems, however, there would still have been reason to worry about Bluh’s argument. If its logic worked, and we should get rid of speed cameras because by the time we catch and fine someone for speeding it’s too late to prevent them from speeding, then we should get rid of more than speed cameras. When the police catch and imprison a murderer, his victim is already dead; should we therefore forget about trying to catch and imprison murderers? Bluh’s argument isn’t just an attack on speed cameras, it’s an attack on crime detection and punishment in general, so goes far too far to be plausible.

Whatever decision the Council reaches, hopefully they won’t get rid of the cameras on the basis of the argument above.

Today the media have picked up on claims that the violence on Britain’s streets is a result of women consuming alcohol while pregnant. The story seems to have started with Scotland’s chief medical officer, Dr Harry Burns, reporting the connection to MSPs.

Pre-natal exposure to alcohol is thought to be linked to a wide range of physical and mental defects, including growth deficiency, learning difficulties, epilepsy, and (at a lower level) poor memory and general clumsiness. Children exposed to alcohol in the womb are even reported to have distinctive facial features, narrow eyes and thin upper lips. The name for these phenomena is Foetal Alcohol Spectrum Disorders (FASDs).

The idea that alcohol exposure damages unborn children is perfectly plausible; after all, it clearly damages adults. There’s still a little work to be done to show that alcohol consumption during pregnancy is a significant cause of violent behaviour on the streets of Britain, however, and certainly not enough in the media reports to do so.

The first thing it would be useful to know is whether the explanation being offered is sufficiently powerful. What proportion of pregnant women drink sufficient quantities of alcohol to cause FASDs? What proportion of children whose mothers drink that much alcohol display violent behaviour? How does this compare to the proportion of children whose mothers don’t drink that much alcohol who display violent behaviour?

Without knowing these things, we can’t know how much violent behaviour could be attributed to FASDs if the claim of a causal link were accepted, so can’t assess its importance as a possible cause of anti-social behaviour.

No doubt at least some of this information is out there (even if it didn’t make it into the media reports). However, there was a slightly concerning comment made in The Herald that the only large study of the prevalence of FASDs was in Italy. For a start, Italian drinking habits are rather different to those in Britain, so there has to be some doubt as to whether Italian findings are applicable here. The lack of corroborative studies also raises a doubt as to whether the connection is as well understood as the reports first made it seem.

Before questioning the significance of FASDs as a cause of violence, however, we should question whether they are a cause of violence at all. Assuming that a correlation between pregnant women drinking and their children tending towards violence has been found, how else might this correlation be explained?

One possible explanation would be that women with a disregard for authority are both (a) more likely to disregard medical advice not to drink while pregnant, and (b) more likely to rear children with a disregard for authority (and so with anti-social tendencies). In that case, a mother drinking while pregnant wouldn’t cause her child to be violent; rather, the mother’s drinking and the child’s violence would share a common cause: the mother’s disregard for authority.

There are other theories that could explain the data too. Perhaps living in a rough neighbourhood tends to drive women to drink and to draw children into gangs. Perhaps children who can steal alcohol from their parents are more likely to get drunk and so get involved in fights. Etc, etc.

Of course, there’s a chance that the claimed connection is real and we’re experiencing social problems in Britain because of pregnant women drinking. There’s also a chance that the scientific community has good reason to believe in this connection.

However, the evidence made available in the national press doesn’t show that there’s a correlation between pre-natal alcohol exposure and violent tendencies in this country, doesn’t show that the best explanation of such a correlation would be that pre-natal alcohol exposure causes violent tendencies, and doesn’t show that pre-natal alcohol exposure causing violent tendencies would explain the levels on violence on Britain’s streets.

For these reasons, the information presented in the media falls well short of demonstrating the connection between drinking and violence that is claimed.

In October 1999, Warren Blackwell was convicted of indecent assault. He had, it had been alleged, raped a woman at knife-point on New Year’s Eve. He served more than 3 years in prison for this alleged crime. The conviction was overturned, well after his release, this month.

Mr Blackwell’s conviction was quashed because the credibility of his accuser was undermined. His initial conviction rested in large part on the testimony of his alleged victim. Once an investigation by the Criminal Cases Review Commission had damaged her reputation, her testimony was no longer deemed sufficient grounds for a safe conviction. The verdict on Blackwell was changed, belatedly, to ‘not guilty’.

The accuser, it turned out, had a history of making rape allegations. She had previously accused at least five men of rape, but never before had her claims been substantiated.

It would be dangerous to argue from the mere fact that a woman has alleged rape on a number of occasions without her alleged attackers ever being prosecuted to the conclusion that she is a serial liar. The conviction rate for rape is only 5%, and this is plausibly due to difficulties in proving guilt beyond a reasonable doubt rather than because most allegations are spurious. A woman might genuinely be raped several times without getting justice.

However, in this case there was enough to call the accuser’s credibility into question. Not only had she made an unusual number of accusations in the past, but at least one of those accusations had been conclusively disproven. She had a history of making false allegations, a track-record of lying.

This case illustrates the importance of reputation in establishing a witness’s credibility. If someone is known to have lied in the past then their credibility is reduced, particular regarding similar claims in similar situations. Once the facts about Blackwell’s accuser were revealed, the case against him was substantially weakened, and he was rightly treated as a victim of a false accusation rather than as an assailant.