Railing against railings
There is a particular kind of official mind that looks at a busy road and concludes that the problem lies with the pedestrians. Rather than slow the traffic, calm the junction, or simply trust adults to judge a gap in the traffic for themselves, it erects a steel cage along the pavement edge and tells people where they may and may not cross. This is the guardrail: perhaps the purest physical expression of the precautionary instinct in British street design, and one of the few pieces of twentieth-century highway engineering that has since been officially recognized, in part, as a mistake.
Most pedestrian guardrail was never justified by evidence. It arrived as the default response of an engineering culture that treated restraint of the pedestrian as the safe option, and having installed it, rarely thought to ask again whether it still earned its keep.
Transport for London's own review, launched in 2005 and pursued with real vigour under Boris Johnson's mayoralty, found that a great deal of it did not. Of 204km of guardrail on the TfL road network, 63km, nearly a third, was removed in Johnson's first term alone, on the finding that it was doing nothing much except narrow the pavement, trap cyclists against turning vehicles, and encourage exactly the sort of climbing-over behaviour it was meant to prevent. A further 23km came down after that, and TfL now reckons some 110km still stands.
That 110km is the unfinished business. It represents guardrail that has survived not because anyone can point to a safety case for it, but because inertia is a powerful force in public administration, and removing things is always harder politically than installing them.
The argument against guardrail was always a good one, and it has not weakened with time. It substitutes a steel bar for driver attentiveness. A driver who sees pedestrians corralled behind a rail can reasonably assume none will step into the road, and drives accordingly, which is precisely why guardrail can increase risk at exactly the crossing points it was meant to protect.
It creates the crush point it was meant to eliminate. A cyclist turning alongside a rail-lined kerb has nowhere to go if a vehicle turns across them, but without the rail, they have an escape route.
It does not stop people crossing where they want to cross. It simply makes them do it while climbing over an obstacle, which is worse than doing it in the open. It costs money to install, money to maintain, and money, apparently, to argue about removing.
It is ugly. This is not a trivial complaint. A city's streets are its most visited public spaces, and cluttering them with barriers no one can justify is a tax on the eye that Londoners pay every day without ever having voted for it.
TfL should commit to reviewing the remaining 110km against the same risk-based criteria that removed the first 86km, publish a rolling timetable for doing so, and remove whatever fails to clear the bar, which, on the evidence of the first two rounds, will be most of it.
Where a genuine safety case exists, a rail that demonstrably prevents pedestrians straying onto a fast slip road with no realistic crossing point, it should stay. This is not an argument for recklessness. It is an argument for evidence over habit. The point of the original 2005 policy was never ‘no guardrail anywhere.’ It was a presumption against guardrail unless someone could show it was doing good. That presumption should simply be applied to the stock that inertia has allowed to survive.
Boroughs outside the TfL network, many of which retain far higher densities of guardrail than the TLRN ever had, should be invited, or where funding gives central government leverage, required, to run the same assessment on their own streets. Much of the ugliest and least justified railing in London sits not on the red routes Johnson's team reviewed but on ordinary borough roads that have never been looked at since installation.
The guardrail is a small thing, but it is a useful test case for a larger principle this Institute has argued for decades, that the state's instinct, faced with a hazard, is too often to restrain the citizens rather than to trust them. London found, when it actually checked, that the citizens climbing over the railings had usually been right, and the railings had usually been wrong. There is no reason to think the remaining 110km will tell a different story. The only question is whether anyone still has the will to finish the job that was so well begun.
Madsen Pirie