MMC Musings

We may not like to admit it, but we’re run by our fast brains. Psychologists call it System 1 — the bit Daniel Kahneman describes in Thinking, Fast and Slow that reacts in fractions of a second, long before the slow, deliberate System 2 has got its boots on. It’s the part that looks after keeping us alive by deploying the ‘fight, flight or freeze’ reaction. Something startles or scares you – it fires instantly to keep you out of danger – you run, stay rooted to the spot or lash out.

That instant reaction is critical; if you’re strolling across the savannah a few million years ago and hear a rustle from the bushes, you’re not going to take off your backpack, get out your Big Book of British Sabretooth Tigers and check which one it is. Instead, you’ll scarper, look dead and hope it doesn’t eat you or beat it to death with your club.

System 1 is great – it stops you getting eaten by tigers. But it also deploys whenever you feel threatened by anything – whether it’s actually dangerous or not. And events that trigger it stay embedded in your mind, so you remember them almost as if they’re hardwired in. They become part of your belief system.

Odd as it sounds, that even happens in road safety.

Imagine… you’re standing on the pavement, chatting with a neighbour when a car comes whizzing past. It’s well within the limit but still too fast for comfort. “Those bloody speeding drivers!” you exclaim, angry and shocked. It’s a straight System 1 threat reaction and it completely bypasses your logical brain. It’ll be fight, flight or freeze or a combination of all three.

Not only that, but those nasty drivers aren’t from your village/town/street, so they’re strangers.  That makes it worse.  Your neighbour agrees and adds “They’re always speeding through here – someone’s going to get killed one of these days.  Someone should do something”. 

Emotional reaction

No-one normal, at this point, suggests using the modern, deliberate part of the brain and pulling the stats 19 records to see if there have been any accidents or check their causes. No-one deploys System 2 and drops the council an email to see what the results of the last speed survey was.

Instead, that System 1 experience bypasses your rational, thinking brain and becomes part of your emotional response and, before you know it, the local playgroup are standing with banners by the roadside demanding a lower speed limit.

That demand for a limit is emotionally-fuelled and so incredibly powerful. And it’s fed by decades of rhyming slogans like ‘Twenty’s Plenty’ and ‘Speed Kills So Kill Your Speed.’ And, at a superficial level, it makes sense – slower must be safer, surely? Our System 1 brains like it too – we don’t like fast, noisy things – they scare us.

Later that morning, you meet your local councillor in the corner shop and angrily tell her about the problem with speeding traffic. She probably has a couple of other similar conversations that same day.

Your councillor confers with her colleagues who have all had the same conversations.  The politician’s syllogism then comes into play:

“Something must be done!”

“X is something”

“We must do X!”

X, in this case, is the only tool in the box – an even lower speed limit. And in Oxfordshire, we’ve seen 60 become 40 become 30 become 20. It’s not going to be long before one of the local activist groups demands that cars should not be allowed to travel faster than the average speed of a bicycle.

Problem is, it doesn’t work – and for a whole host of reasons – but principally because it doesn’t fix the problem of people’s System 1 reactions.  

The argument for limits

Let me be fair to the other side here, because their strongest argument isn’t “slower must be safer” and it deserves an answer. It’s this: fear is a cost in itself. If people believe the road is dangerous, they behave as if it is – parents won’t let children walk to school, older residents won’t cross to the shop, the village gets cut in half by its own high street. On that view, a limit that makes people feel ‘safer’ has achieved something real even if the casualty statistics never move.

It’s a serious argument. And it fails on its own terms – because the fear doesn’t respond to the limit.

I saw it myself this morning when I was chatting to a friend in my own village outside the Co-Op. A car came past. It’s a 20mph limit and the driver was well within it (comfortably, I’d say), but in second gear, so revving and a relatively narrow road gave an exaggerated sensation of speed. My friend’s reaction? You can guess.

The number on the stick had changed. The System 1 reaction hadn’t – and won’t. If the case for the policy is that it reduces fear, and the fear is exactly the same at 20 as it was at 30, the policy has failed by the measure its own supporters chose. It’s happening in Oxfordshire already even with blanket 20mph coverage.

There’s no point in using logic or reason when it comes to speed – people just tune out

Because System 1 isn’t logical and actually doesn’t welcome logical intervention, you can deploy all the stats and figures you like, people simply glaze over and demand a lower limit. But it doesn’t matter how far you lower the limit, people will still get hacked off with ‘those speeding drivers’.

I’m reminded of JJ Leeming’s quote in his 1969 book Road Accidents, Prevent or Punish:

“In the three years before (introduction of the speed limit) there had been twelve accidents. In the three years after, there were nineteen… As was my usual practice, I reported this to the local Road Safety Committee. At the meeting the Parish Council representative said: ‘My council doesn’t mind if the accidents have increased. We have got our speed limit! The Road Safety Committee supported him!’

Ultra-low limits still won’t work

Let’s imagine the limit gets lowered from 40 to 30 or a 40 to a blanket 20. In most places, it’ll still be “Those bloody speeding drivers”.

Then, disappointed residents campaign for an even lower limit, calming, cameras, gateways, bumps – whatever. None of it will make them feel any safer because that emotional, System 1 reaction will still be there. We can’t take away people’s annoyance and fear with numbers on sticks and humps. (And yes – “numbers on sticks” is me playing the same emotional game I’m describing. It works, doesn’t it? That’s the point.)

The faith in those magic numbers on sticks to make us safer is remarkable, even when the evidence of our own brains contradicts it.

In Oxfordshire, the LibLabGreen Alliance comes along and promises free blanket 20s for everyone and we are where we are.  Add in their desire to use limits to promote active travel (even though they’re not measuring their impact) and the water is even muddier.

Emotional opinions not rational argument

So rather than being an informed, logical debate it becomes a process of “who can hit with the most heartrending emotional story”. Activist campaign groups are superb at it, and the stories are genuinely heartbreaking – but look closely at the cases wheeled out for the cameras and the limit is often a bystander: a driver over the BAC limit, drugged, in a stolen car, who’d have been lethal at any number on any stick.

The injury-severity curves are real – you’re more survivable hit at 20 than at 30 – but they only matter if the limit actually changes the speeds people drive at. Sign-only limits barely do, which is precisely why the DfT’s own guidance says limits should reflect what drivers are already doing.

It’s gone so far that OCC didn’t even bother carrying out speed surveys before they imposed blanket limits in Oxfordshire, let alone even taking DfT advice on setting limits at mean average speed (which in my view is insane as it criminalises half of drivers and especially the more competent: https://mmcmusings.com/2014/12/27/dolores-umbridge-in-charge-of-speed-limit-policy/).

Using the wrong tool for the job

The other big problem is County Hall’s current incumbents don’t understand even basic systems thinking.  Sorry – another link: https://www.linkedin.com/pulse/why-you-need-sack-dead-bloke-whos-running-your-mark-mcarthur-christie/ 

Councillors are using the tools for a complicated system on a complex one. When it’s that way round it always takes longer for the problems to surface, but when they do they’re usually worse.  

Sadly, I think we’re about to start seeing those problems surface – there’s so much more tailgating, frustration, anger, frustration overtakes… It’s even harder to cross the road because vehicles are bunching in 20 limits in a way they don’t in effectively-set limits.

And this is before we even get into the insane ‘driving by numbers’ limit change frequency.

So what would actually work?

Because “do nothing” isn’t my answer either. If you want traffic through a village to move at village speeds, there are tools with actual evidence behind them:

Set limits at the speed drivers are already doing, just as the DfT’s own guidance used to say – limits that match the road are limits people keep, without a camera in sight.

Build self-explaining roads. Drivers read the road, not the signs. Narrowings, surface changes, planting, buildings close to the carriageway – environments that feel like 20 produce 20, no enforcement required. It’s more expensive than a sign, which is exactly why councils don’t do it, and exactly why signs don’t work.

Use vehicle-activated signs where there’s a genuine hazard people need to slow down for. TRL’s research (Report 548) found them remarkably effective at cutting speeds at the specific sites that need it – targeted, cheap, and no criminalising of drivers doing sensible speeds on roads that can take them.

And measure. Speed surveys before, speed surveys after, casualty and collision causation data published against a proper baseline. If a scheme works, the numbers will say so. If a council won’t collect the numbers, ask yourself why.

None of this fits on a dimity little rhyming ‘bumper’Twenty’s Plenty for me’ sticker. That’s rather the problem.

What comes next?

I hope I’m wrong, but I think we’re about to see another very bad year for road safety indeed. Sadly, the response is unlikely to involve questioning the role of speed, speed limits and their enforcement. Instead, it’ll be “the medicine’s not working – make them take more!”

Postscript – July 2026

Update, 2026: I wrote the article above in February 2024 as a prediction. It was checkable, and now it’s been checked – see Oxfordshire County Council are spinning the road safety stats again for what the casualty data actually did, and what County Hall said it did. I’d have preferred to be wrong.