Fire: An ancient foe

One of the first things done by the first mayor of London in his first year in office was to issue a set of building regulations, intended to prevent the spread of a conflagration. Henry fitz Ailwin’s code, set out in 1189, is – once you translate it into English – pretty clear: “No chimney…to be built, unless of stone or faced with tiles or plaster, and not of timber…No house…to be roofed with anything other than lead, tile or stone. If any other exists, it may be pulled down forthwith by the constables…”

This represented a change. In an earlier time, fire had been treated as a mere destroyer of property. One seventh century law code had declared that “fire is a thief”. It takes away precious woodland. But moving into the cities means fire is not just a thief. It is a murderer.

In 1212, fitz Ailwin revised his code: there had been another fire. He needed to toughen up enforcement. After the Great Fire of London in 1666, the rules were changed again. “All the outsides of all Buildings in and about the said Citty be henceforth made of Bricke or Stone or of Bricke and Stone together except Doore cases and Window Frames.”

Enforcement was beefed up, too. Surveyors were to be appointed “to see the said Rules and Scantlings well and truely observed”. But these old codes, the products of lessons hard learned, are more than a historical curiosity.

First, they tell you an important story about our state. Defeating urban fire has been a critical role for longer than the English have been writing in English. A reduction in fire deaths over recent decades has resulted in complacency. But there were building codes before there was Magna Carta. The English fought city fires before they got into the habit of fighting the French.

And, second, as I read the 12th century proposals for keeping London safe and its 17th century strengthening, I find myself sighing at how clear the rules are and how well they served us. These codes were effective: London survived a literal incendiary bombing campaign.

Since the Grenfell Tower disaster of 2017, there has – rightly – been focus on the London Fire Brigade, and why the institution had not prepared its staff well enough for what went on to happen in west London. The first phase of the Inquiry, which covered their role in the blaze which claimed 72 lives, is the only part to be completed. But the blame is not theirs: it was the fire that killed, not the firefighters.

The catastrophe at Grenfell is the story of undervaluing a simple 800 year-old legacy of English urban fire-fighting: you never let the fire spread. The British have had rules that worried about so-called compartmentalisation long before we had rules on means of escape. And there are lessons in why this rule ceased to be observed.

At 00:54 on 14 June 2017, Behailu Kebede called 999 to report a fire on the fourth floor of the Grenfell Tower. Just 27 minutes later, at 1:21am, a resident on the 22nd floor called 999 – 18 storeys up. By 1:27am, the fire had reached the roof.

The cause of the disaster is pretty clear. When the Grenfell Tower was refurbished between 2014 and 2016, the builders affixed an insulation system to the building – the exterior cladding. On the inside of the cladding was a combustible insulation foam. These modern insulation materials are good at keeping buildings warm. But, as the Grenfell Inquiry put it, “they have a comparatively low time to ignition and can support rapid flame spread”.

More significantly, the builders affixed a set of aluminium panels on the outside to shield the building from the weather – specifically, Reynobond 55 PE. The PE stands for “polyethylene” – that is what was in the core of the panels. Polyethylene ignites at 377C, and can “flow whilst burning and generate burning droplets”.

How did it come to be that a 23-storey tower was permitted to be coated in lethal fuel? The answer is in the English building codes. These claim to have the same objective as our medieval forebears: contain the fire. But, in practice, the code was undermined.

The relevant part of the regulations, specifically “Approved Document B, Part II”, sounds medieval in its opening statement: “The external envelope of a building should not provide a medium for fire spread.”

The rules stated that any insulation applied to the exterior of a tall building should meet a standard known as “limited combustibility” – effectively being fire-proof. But the rules create an exception: external walls should meet that standard or “the performance criteria given in the BRE Report… [BR 135]”.

What that means is that you are permitted to use combustible insulation on the outside of buildings, so long as the specific design of your wall system has been tested using something called the BS-8414 test. You must build a large-scale model of your proposed design and have a fire engineer light a fire underneath it to check for fire spread.

That is where the regulations end. But the regulations do not exist in a vacuum.

First, the testing system itself is confidential. So I could commission 10 tests on my preferred wall design and it can fail nine times. But if it passes once, that is enough. And no-one would ever know about the nine failures.

But there’s more: in 2014, a building inspection body called the Building Control Alliance issued guidance to its members that, in the event that that you were building something similar to something already tested, you could ask a fire engineer to ratify that it was effectively the same.

By the time of the Grenfell Tower fire, we found examples of this process – so-called “desktop studies” – being used to sign off designs reliant on aluminium panels when the tests had been conducted using ceramic tiles. Oh, and these “studies” are confidential, too. So we do not know what it is in them, and how many are reasonable.

Some institutions went further. The NHBC, a charity, was advising builders that you could use some types of combustible insulation with some combustible outer cladding panels. And you need not bother getting a test done, nor a desktop study commissioned.

So we started with a building industry which was given just one carve-out from the Stuart-era principle – nothing that burns on the outer walls, you fools. But they expanded them, and governments went along with it. There were, in addition, mistakes in the drafting of the rules, which some builders ruthlessly exploited. The building inspectors were swept along in a tide of accepted practise which drifted ever further from the intent of the regulations.

In the wake of the fire, the government started enforcing a tough reading of the original rulebook. As of today, two years on from Grenfell, there are 318 high-rise buildings that are believed not to meet the original standards. A further 118 have already been fixed.

Since the building inspection process is shrouded in secrecy, we cannot check how and why decisions were made. We still do not know why or how the Grenfell Tower refurbishment got signed off by building inspectors. No-one has ever found a test certificate, nor a desktop study for that building. But £600m of public money has been allocated to re-cladding the dangerous buildings that are left.

The root of the problem is that, in 1985, the government switched from a “prescriptive” rulebook for safety to a “functional” one. We do not need to prescribe how buildings should be built, they said. Who cares how they are constructed, so long as they are safe?

The ambition is to allow flexibility for innovation. The principle is that you just need to be able to show to an inspector how something is safe. It is a noble instinct for the state to seek to constrain behaviour as little as possible. We also train engineers to think about outcomes, not processes – and a great number of good, safe buildings have been built that would not have met the needs of an older prescriptive rulebook. But a desire to be business-friendly has made the state wary of exercising its authority – even in ancient areas of responsibility.

I spent part of the day of the fire in 2017 looking at footage from other cladding fires – in France, in Dubai, in China. There had been warnings about the safety of some components being used on high-rises – including ones similar to those used at Grenfell. It was known that cladding could be extremely dangerous.

But while there were warnings, a principles-based system has no explicit and automatic means of turning those warnings into a widely adhered-to ban. In truth, our system allowed that any design could be used – just so long as the builders could mount an argument that the cladding was safe. A desktop study or a test result matters – not some news report from China.

Local authority building inspectors should have filled in that gap, you might say. And some are assiduous in asking for evidence and checking in on developers. One building inspector told me about his department asking a developer for evidence for the safety of a new fire door. They received a grainy VHS video in the post showing someone aiming a blow-torch at the door for 10 minutes. (“Please return when done,” read the hand-written label.)

But the system created a structure where the inspectors find themselves facing overwhelming pressure and, if they challenge a building’s safety, it could end in a trip to court where it will be “my word against that of whichever PhD engineer the developer’s hired to shoot me down”, as one put it to me. A rulebook that is too open to interpretation makes it harder for a building inspector in a hurry to stand their ground.

It is also clear that some inspectors wave things through. Since 1985, we allowed builders to select private companies to perform these building control inspections – so-called “approved inspectors” (AIs). There have been useful gains from this decision: the old public sector monopoly served both the public and the builders poorly. But there is an incentives problem with AIs: the Hackitt Review of building standards, commissioned after the fire, noted that some compete “to attract business by offering minimal interventions or supportive interpretations to contractors”.

The growth of the AIs, however, was part of a broader trend that had a significant impact in the story of Grenfell: the deskilling of the state. The British government lacks expertise in fire.

I was a reporter on BBC Newsnight when the fire took place. The piece we ran on the night of the fire revealed that the likely culprit was the polyethylene-core cladding. We then endured five days of pressure about the report. Are you sure? We were told by experts that we must be wrong. No sane person would ever have used that material.

Then we started reporting on “desktop studies”. Again, respectable builders and fire engineers explained that desktop studies are what you use when you are changing “brass rivets for steel rivets”, as one of them put it to me. You would never use them to swap out the insulation or panels in a cladding system. This must be a mistake.

Then, as time ticked on, we got asked by government to help. Could we source some sample desktop studies, for example? We realised we were – on these narrow questions – moving quicker than the government. This is not intended as a boast: the state was completely flying blind. Whitehall had no feel for what was happening on the ground.

They had no expertise on hand. For decades after the Second World War, the Joint Fire Research Organisation and then the Fire Research Station operated as independent fire safety analysts. But they were eroded down to being just a part of the Building Research Establishment (BRE), a state-run laboratory which was itself privatised in 1997. For 22 years, the UK government has had no fire safety laboratory of its own.

This was exacerbated by other decisions which meant that the “surveyors” of Charles II’s day – the people who were keeping an eye on building standards – have become a depleted force. One effect of the rise in Approved Inspectors was that there is no longer a big, deep bench of expertise which exists to serve the state’s policy agenda. There are fewer publicly employed building inspectors.

So when Sajid Javid, the communities secretary, was assembling an “expert panel” to advise him on fire safety in the aftermath of the fire, it proved difficult to find experts – or at least experts who were not compromised. Everyone has been paid by one side of an argument or another. It is too small a world. Sir Ken Knight, a genuinely esteemed expert who chaired the group, literally signed the testing certificate for Reynobond 55 PE.

The decline of the state building inspector, the privatisation of the BRE – and deep cuts to the civil service since 2010 – meant that the state’s understanding of fire had been allowed to wither away. And while regulation made safety the job of builders, not inspectors, that seemed not to matter. But we desperately need to rebuild this capacity. Fire is an ancient foe, but one that constantly changes.

We have had a few decades of fire safety success. The number of deaths from fire has plummeted. Some of this is the result of policy wins: the drop in smoking, in particular, combined with the rise in smoke alarms, fire drilling, fire education and the outlawing of the worst types of furniture fillings (an area, incidentally, where Britain is very prescriptive).

But there is a danger we need to confront. The fires of fitz Ailwin’s day are not the fires of today. So, for example, we have less wood but more plastic in our homes; we are surrounded by flammable polymers.

Underwriters’ Laboratories, an American institution, performed a test where they dressed two identically sized rooms in furniture from different eras, and then lit a fire to watch how fast it took for the whole room to reach “flashover”, the moment when the room’s surfaces all ignite. “The modern room transitioned to flashover in 3 minutes and 30 seconds and the legacy room at 29 minutes and 30 seconds.”

There are other changes we have made, too. We have tended to build more open-plan designs, which means that there is less containment inside buildings. Fires are less likely to be confined and are more likely to be well ventilated. And lightweight building materials often have worse fire performance than older alternatives.

The continuing decline of smoking might mean we are moving to a world where fire might become rarer. But it is a world where fires, when they happen, are more likely to be catastrophic. And where containment features which worked in previous decades might no longer function. Fire doors, for example, will not last as long.

A central problem is that the British state is too inexpert to follow these issues. No-one is in charge. Its usual method of operation is to commission occasional reports and research, but then to ask the businesses engaged in areas of business to consult on what should happen next. Or designing by disaster: waiting for something horrific to go wrong before anyone takes action. It took a secretary of state facing national political embarrassment to ban combustible material from high-rise walls.

Taking all these threads together, you can see the bones of what reform should look like. We need to empower the building inspectors to stand their ground – and keeps the dishonest builder (or lax inspector) in check. The Hackitt Review did propose big changes specifically to who should handle large, complex buildings – and it should help.

That report also proposed the need to build a pool of professional specialist fire engineers whose judgment and training can be relied upon. Too right – that requires robust institution-building to do accrediting and casting-out.

But we can go further: the only way to make it possible for inspectors to be able to stand up to developers is to make everything transparent, so that there is a chance that bogus research can be exposed. Any piece of research relied upon for a fire safety assessment must be public.

We also need the state to take an active role in the market. To check fire doors and the combustibility of everything from modern sofas to cars. It needs to take a lead. The developers spend money on research into products and design. We need to arm the inspectors with countervailing expertise. And if Whitehall will not do it, our big cities should.

Finally, we also need to be more relaxed about businesses being angry at government. Reform will probably mean a system where there is less discretion for builders – and a world where some types of components  are banned more easily. This would probably mean some building designs would no longer be possible – but we can live with that.

There are reasons to think of fire as a special category of problem: the consequences of errors in building design can take years to emerge, decades to fix and minutes to take a deadly toll. But there are elements of these issues which have broader applicability.

The British state is too blasé about outsourcing responsibility for functions or regulation and then allowing its own expertise to disappear. Effective regulation – even if it is functional regulation – requires a strong centre to keep an eye on what is happening. And the government is too nervous of standing up to business.

Banking regulation has gone through the same process. The rules on financial stability before the crisis gave too much power to institutions to mark their own homework – and, as in fire safety, the government before the crisis bragged about being light-touch and low-regulation.

We have a state that, denuded of expertise, lives in too much fear of upsetting vested interests. The Stuart building laws were blunt and crude. Give me that over pliability.

Photographs Getty Images, grenfelltowerinquiry.org.uk