From the specific to the general

Imagine a grim, appalling, but unfortunately all-too-common scenario. A primary school is under attack from a deranged gunman. Shots have been fired, and the gunman stalks the school corridors, looking for children to kill. In one of the classrooms, a nine-year-old child is cowering under his desk with his teacher, both hearing the approaching footsteps of the gunman.

As the gunman opens the door to their classroom, we freeze time, and imagine two possible alternative scenarios. In the first, both teacher and pupil are unarmed and defenceless. In the second, the teacher has a firearm, which he has in a holster.

Given these specific circumstances, I’m sure most of us might consider it would be better – at that specific moment – for the teacher to be armed with a gun, than to be unarmed and defenceless. With a gun, he might, at least, be able to surprise the gunman, leaping up from his hiding place and firing several rounds at him, incapacitating him. That would certainly be better than the alternative of being effectively powerless as the gunman enters the classroom.

So, given these specific circumstances, we could reasonably think that is a good idea for a primary school teacher to be armed with a gun.

But would any of us then draw the conclusion that it is a good idea to arm primary school teachers in general? Just because our particular teacher might benefit from having a gun in the specific circumstances of a gunman approaching him down a school corridor, do we then think it makes sense to for all primary school teachers to be equipped with an easy-to-access handgun, throughout the school day?

That – to me at least – actually sounds like a pretty dangerous idea, even if I might have agreed that it would definitely be better for the teacher to have a gun, than to be unarmed, under the specific conditions of our thought experiment. I certainly wouldn’t be easily persuaded that the ubiquitous arming of primary school teachers is consequently a good idea, and I would also be resistant to accusations of being inconsistent.

This is because there are lots of reasons why arming primary school teachers is a bad idea. Those reasons don’t become any less compelling if a primary school teacher might benefit from a handgun in a classroom when the school and its pupils are actually under attack.

To make this even clearer, we could go further and imagine the nine-year-old child is alone in the classroom as the gunman approaches. Again, we might agree that the child having easy access to a handgun at that specific moment might be a good idea, while being appalled at the notion of all nine-year-olds arriving at schools equipped with handguns.

Anyone who is arguing that arming primary school teachers (or nine-year-old children) is a good idea in general, based on a specific, isolated scenario like the one outlined here is effectively performing something amounting to sleight of hand.

Logic.

It’s a pleasingly simple argument that unfortunately misses out all the reasons why arming nine-year-olds with guns when they go to a school might not be a good idea (negatives), and that also misses out all the other ways we can potentially stop nine-year-olds from being shot (positives).

But of course many people in the United States do actually think like this, at least when it comes to arming teachers. Many schools think it’s a good idea for teachers to be armed. For these people, the potential negatives of teachers being armed aren’t being considered, or are outweighed by the persuasive argument that teachers would then be able to protect themselves once an attack is already underway, something they would not be able to do without a gun. Likewise one powerful way of reducing the likelihood of children being at risk of being shot at in the first place – gun control – is completely unthinkable for a large swathe of American society.

We wouldn’t employ this kind of logic in Britain, would we? Perhaps the closest analogy might be knife crime.

Logic.

Unpersuasive, right? While we might agree that a teenager being attacked with a knife would – in those specific circumstances – benefit from a stab-proof vest, we wouldn’t think that all teenagers should therefore be walking around in stab-proof vests, or even just those teenagers in areas with particularly high knife-crime. The correct response would be to take positive steps to prevent teenagers from being stabbed in the first place, rather than fitting them all with cumbersome protective equipment – while also taking into account the direct negative consequences of having to wear that protective equipment.

The examples don’t even have to involve violence.

Logic.

Just because we accept that a pedestrian directly under a falling brick would be better off with a hard hat, we shouldn’t then be compelled to accept that all pedestrians should therefore walk around wearing hard hats. We’ve missed out all the other potential ways we can stop bricks striking people on the head, and all the potential negatives of compelling people to walk around with hard hats on.

And yet. And yet. It is precisely this form of argument that is, unfortunately, extremely common in discussion and debate around cycling safety in Great Britain (and indeed across much of the world).

Logic.

Certainly, if I was cycling along and I found myself confronted with a motor vehicle heading unavoidably towards me, I may think that – under those specific conditions – some polystyrene might lessen my chances of injury. But as with all these examples here, that would be a poor basis for arguing that anyone cycling around should always wear a helmet. Doing so misses out all the positive ways we can stop people being hit on the head by motor vehicles, and all the negative consequences of forcing people to wear helmets when they want to cycle somewhere.

The problem with all these forms of argument is what Jack Elder calls – in this excellent Twitter thread that inspired this post – the conflation of the specific with the general. (Click to expand the thread)

 

Possibly contentious assertion: the cycle helmet debate is unhelpful in exactly the same way as the gun control debate: because people confuse systemic effects with specific cases.

— Jack Elder (@jackelder) March 12, 2018

Something might be a good idea under specific, exceptional circumstances, but that’s not a sound basis for suggesting it’s therefore a good idea in general. The aforementioned sleight of hand involves removing that distinction.

This photograph from a recent academic conference on emergency medicine managed to attract 129 retweets, and was shared many more times, by doctors, nurses, students and paramedics, all agreeing with the expressed sentiment that cyclists should definitely wear helmets, at all times, on the basis of the statistics shown in the chart.

Doubtless, in the specific circumstances of people suffering an injury serious enough for them to be admitted to hospital, those injuries could well have been less severe if they had been wearing protective equipment. But is that a sound basis for arguing that all people should wear protective equipment at all times? It most certainly is not. It ignores all the sensible ways we should be preventing injuries from happening in the first place, and ignores all the negative consequences of compelling people to wear protective equipment at all times in an attempt to mitigate serious but rare injuries.

It would be reasonably easy to draw up a similar chart showing the proportion of people admitted to hospital after being injured by falling objects who suffered various forms of head injury, split between those people who happened to have been wearing a hard hat at the time of their injury, and those who weren’t. The ‘hard hat’ group would likely have suffered a smaller proportion of the listed injuries than those people who hadn’t been wearing hard hats. It would be ridiculous, however, to draw the conclusion from that chart that people in general should always wear hard hats. We would have made the mistake of  thinking that because personal protective equipment might reduce the chances of an injury occurring among the population of people who have already been admitted to hospital, personal protective equipment should therefore always be worn – without asking any reasonable questions about

• how those injuries occurred,
• how they could have been prevented,
• whether compelling people to wear protective equipment might have negative unintended consequences,
• or indeed whether compelling people to wear personal protective equipment has any effect on injury rates at a population level.

We can speculate about why it is helmet-wearing among people cycling that is repeatedly subject to this superficial level of argument. My hunch is that – among other possible explanations – it originates with a failure to conceive of ways in which we can prevent, or lessen the severity of, head injuries to people cycling around, beyond strapping helmets to them. We simply can’t imagine a world in which people cycling around don’t suffer the risks posed by motor traffic, or a world in which cycling could be a genuinely low-risk activity.

Consequently, if we think exposure to motor traffic danger is somehow inevitable, immutable, our responses to cycling injury are inevitably going to be limited to protective equipment. In this sense, our response is analogous to the gun control debate in the United States, where for some people the idea of reducing or even removing access to guns among the general population is effectively unthinkable. Consequently, for these people, guns and the danger posed by them are also inevitable, so much so that arming teachers becomes the only logical response. We shouldn’t be constrained by our failure to imagine alternatives.