Back in the spring of 2019, a driver lost control on the A5 near Church Stretton in Shropshire. The road was bone dry, visibility was perfect and the alignment was as straight as an arrow for over a mile. The speed limit was 60 mph but the vehicle was actually going over 70 mph when it crossed the centreline and ploughed head-on into oncoming traffic. The driver didn’t survive.
That bit of the A5 follows the route of Watling Street – one of the old Roman military highways that were engineered in the decades following the AD 43 invasion to move legions quickly from the Channel ports to the frontier zones of Wales and the north. Nearly two thousand years later its still moving vehicles along at a pace that’s far too fast, and people are still losing their lives on it.
I’ve been a UK traffic calming expert for years and I’ve spent a long time studying the relationship between road geometry and how drivers behave. The more I look at Britain’s collision data, the more I notice a worrying trend: the roads we inherited from the Romans are the most like deathtraps in the country for speeding-related crashes. This article explores the rather dark legacy we’ve got and wonders whether we have a moral duty to adapt ancient infrastructure to keep modern drivers safe, or whether we should be preserving it as a bit of history.
The discussion that follows will combine archaeology, transport engineering and a lot of soul-searching. It will look at why straight, wide roads with clear views and good drainage seem to encourage people to drive at breakneck speeds even when they’re supposed to be going slower. It will tackle the conflict between giving these roads a heritage listing and the need for them to be safe. And it will use case studies to show how traffic calming measures like speed bumps and speed tables can be brought in without sacrificing the roads’ historical charm.
The Romans built their roads to move armies quickly. And, amazingly, two millennia on those same roads are still moving traffic along faster than it probably should be. We’re driving on infrastructure designed for taking over conquerors – and we’re dying on it.
The Roman Road Network in Britain and Today’s Crash Map
By AD 200, the Romans had built over 2,000 miles of major roads right across Britain. These weren’t just rough tracks – they were highly engineered highways built to get troops, goods and officials from one place to another at the speed of light.
The main routes included:
- Watling Street: Running from the Channel ports at Richborough and Dover through Londinium (London) to Viroconium (Wroxeter) in Shropshire – roughly following today’s A2 and A5
- Ermine Street: Connecting London to Eboracum (York) along what is now the A10 and A15
- Fosse Way: Linking Exeter to Lincoln via a diagonal route parallel to the modern A46 and A429
- Dere Street: Running from York to the Scottish border as today’s A68
- London–Silchester–Exeter route: Serving the south west legions
These routes didn’t just vanish when the legions pulled out. Estimates say that 20–30% of the UK’s main road network still runs on the foundations laid by Roman surveyors a couple of thousand years ago. The A5 through the Midlands, the A1 in North Yorkshire, the A66 across Cumbria – all follow paths cut by the Romans long before the first motor cars rolled onto the scene.
The Collision Evidence
The Department for Transport’s STATS19 data from 2015-2022 shows a pretty scary trend. Rural single carriageways – many of them on the old Roman routes – account for around 60% of fatal crashes in Britain, even though they only make up 20% of total road mileage. Speed is cited as a contributing factor in 25-35% of the cases where people have been killed or seriously hurt on these roads.
A 2021 report from the RAC Foundation noted that Roman straights crank up the risk of speeding in rural areas by a factor of 2-3 compared to roads with a similar classification but with some curves.
Important to note: correlation does not equal causation. Not every crash on a Roman road was ’caused’ by the fact it was a Roman road. But the engineering that went into these routes – their geometry, their width, their drainage – all play a part in making modern speeding more likely.
Why Straight, Fast Roads Invite Speeding – No Matter What the Limit
Years and years of traffic engineering and human factors research has shown us that drivers set their speed according to how the road feels – not according to what the speed sign says.
This phenomenon – sometimes called the ‘self-explaining road’ principle – means that drivers look at the road’s width, curvature, views and roadside features to decide what feels like a safe and sensible speed. Posted limits and warning signs have relatively little effect unless the road’s geometry is working with them to send the right message.
Roman Design Priorities
The Romans built their roads with a single priority in mind: get the job done quickly. And that meant designing the roads to be as straight and wide as possible, cutting across any obstacles that got in the way. Their design criteria included:
- Straight alignments: Tangents of up to 20 kilometres, slicing across the landscape with no effort at all
- Raised agger: Cambered embankments 1-2 metres high and 5-10 metres wide, designed to get rid of water in a hurry*
- Consistent width: Most roads were about 4-6 metres wide, ideal for a couple of carts to pass each other as well as a legionary marching in formation
- Good drainage: Side ditches and a layered construction system ensured that the roads didn’t become waterlogged
These features were a big part of the reason that Roman legions could march 20-30 miles per day, and that wagons could travel steadily in all kinds of weather without getting bogged down. The roads were built with speed in mind – Roman speed, that is – with carts probably averaging about 5-6 miles an hour, while mounted messengers might have reached a top speed of 25 miles an hour.
Modern Consequences
The net result was that these roads sent a pretty clear visual message that speeds of 60-70 mph were acceptable, even when the speed limit was just 30-40 mph approaching a village.
Take a look at a typical Roman-aligned rural B-road as it enters a village – the 30 mph sign appears, but the road itself hasn’t changed much. The carriageway is still wide and straight, with sightlines stretching out for hundreds of metres. Unless drivers have already seen the sign and are ignoring it, most of them will roll in at 40-45 miles an hour because nothing about the road says “slow down”.
It’s a well-known fact that traditional speed limit signs and enforcement are rarely enough to bring speeds down on their own – you need physical or visual changes to the carriageway to make a real difference.
Heritage vs Safety: When Scheduled Monuments Meet Real-World Safety Data
In 2018, highway engineers proposed a safety scheme for a section of the Fosse Way in the Cotswolds, a road that had seen multiple serious accidents over the previous decade. Several of those had involved drivers going at excessive speeds, and the proposal included speed humps at village approaches and kerb build outs to create the illusion of narrowing.
But the scheme got rejected.
The Fosse Way runs through a Scheduled Ancient Monument, and Historic England officers were worried that the proposed road humps would damage the Roman embankment preserved beneath the surface for nearly 2,000 years. The compromise they eventually agreed on relied on visual narrowing and gateway signage, which is a lot less effective at actually slowing people down.
This case highlights a broader tension in transport planning on heritage routes.
The UK Heritage Framework
Roman roads in Britain are protected under a number of overlapping designations:
- Scheduled Monument status: So any works that might affect the monument need consent from Historic England
- Listed Building proximity: Grades I and II* buildings can constrain road alterations if they’re near by
- Conservation Areas: Local authorities have controls in place to protect the visual character of the area
- National Planning Policy Framework (NPPF): Preservation is usually the priority, but safety can override that if there’s a good reason
These protections are in place for a good reason – Roman roads are irreplaceable archaeological assets, and the agger, the drainage systems, the original stone surfaces all tell us a lot about engineering practices that shaped a continent.
But they also put highway engineers in a tricky spot.
Documented Conflicts
There are several UK locations where this preservation/safety dilemma is playing out:
- Fosse Way, Cotswolds: The fact that the road is a Scheduled Ancient Monument made it impossible to put in speed cushions and deep excavation for traffic management
- Roman causeway near Hadrian’s Wall (A69): In 2020, safety interventions were limited to gateways, in order to avoid cutting into the embankment
- Chester (Deva Roman fortress): The medieval fabric built on top of the Roman street grid makes it hard to realign kerbs, so shared space design is the only option
- Lincoln High Street: The fact that the Roman north-south axis is buried under medieval buildings limits how much you can do with the horizontal layout
The issue becomes a real conundrum when you look at the collision history on one of these protected routes – 5-10 KSI incidents per decade, for example – and wonder who gets to decide how much change is acceptable. When does the duty to pedestrians and cyclists crossing a Roman street outweigh the duty to preserve the stones beneath their feet?
Case Studies: How to Calm Traffic on Roman Roads Without Damaging Them
The good news is that you can actually implement traffic calming schemes on Roman alignments without damaging the archaeology. The following case studies show how various traffic calming measures have been implemented in a way that is sensitive to the road’s past.
Case Study 1: Fosse Way Through Ettington, Warwickshire (2016)
This Roman-aligned village street had seen four KSI collisions between 2011 and 2015, with speeding identified as a factor in most cases.
The traffic calming tools deployed included:
- Village gateways with a 3 metre visual narrowing using build outs
- Textured entry treatments in contrasting materials
- Two speed tables (100mm high, 6m long) at the school crossing and main pedestrian crossings
- Buff-coloured asphalt on the table surfaces, which mimicked local stone
Historic England was happy with the scheme because it didn’t involve any deep excavation that would have damaged the agger. The speed tables used permeable setts to let water drain just like the Romans intended. The visual impact was deemed minimal, with materials and colours that complemented the Warwickshire stone vernacular.
Case Study 2: Watling Street (A5) Near Towcester, Northamptonshire (2019)
Full Width Speed Humps were knocked back on this semi-rural bit of road because of how they’d look and the historical significance of the area. The road gets a lot of big vehicles including tractors and whatnot, and traditional road humps would have needed digging pretty deep. Some traffic calming devices like speed cushions or partial humps are designed to only take up part of the road, so buses and emergency vehicles can drive over them without losing their speed.
Alternatives to speed humps that were considered:
- Four chicanes using two metre build outs on one or both sides of the carriageway
- A surface of coloured resin-bound gravel to give drivers something to look at
- Paint on the road to make drivers understand where they should be going
- Gateway signs at the points where the speed limit drops to 40
The chicanes deflected drivers by about 20 mph. Average speeds across the stretch of road where we were monitoring dropped by around 10 mph. Injuries from crashes were down by 60% over three years.
Heritage people said that we weren’t spoiling the look of the place. The scheme showed that slowing people down doesn’t always require putting up big vertical bits – sometimes just making the road curve in a funny way does the trick.
Case Study 3: York – the old Roman road, Via Principalis (2022)
York is a real challenge when it comes to balancing the old and new – the Roman roads are still under the old medieval street pattern. The Via Principalis is the main road that runs north-south through the old fortress and it’s a nightmare for drivers and walkers alike.
We did a few things to calm it down:
- Used cobbled speed tables (80mm high) on four key junctions – we made them out of reclaimed stones
- Made the corners a bit tighter (4 metres) so drivers have to slow down when turning
- Got rid of some of the unnecessary signs and traffic lights
- Reduced the number of parking spaces so the traffic lane is that bit narrower
Result was an average speed of 18 mph and crashes down by 70%. The good bit is that we managed to keep the character of the place – the flat stone surfaces looked like they belonged on the street, not like some alien road furniture.
Using vertical deflection on old roads – speed humps and speed tables, but not too many
Vertical deflection is one of the few things that actually works to slow down traffic on straight rural roads. But when you’re dealing with old roads, you need to be careful not to mess up the character of the place.
Looking at the different types of deflection
Speed tables are good on old village streets because they double up as formal or informal crossings for people who can’t drive.
Adjusting the design for old roads
Implementing flat top humps on Roman roads needs some special care:
- Limit the number of devices – don’t put too many in one place
- Get the ramp profile right – 1:15 gradients are better for noise and vibration
- Keep drainage going – 3-5% crossfall will keep the Roman camber working
- Choose the right surface – stone or Quicksetts finishes on the plateaus
- Consult with the experts – archaeological assessment and geotech drainage checksÂ
The A466 Wye Valley, a Roman route that just happens to run through the Welsh borders, got three speed humps (90mm, sinusoidal profile) back in 2018 along with some lane narrowing and some new signs. Afterwards, monitoring showed:
- Average speeds plummeted by 10 mph
- The number of KSIs halved between 2019 and 2023
- Not a single issue with drainage
The trick to making vertical deflection work on heritage routes is to treat every fix as a tiny tweak to address a specific problem spot, rather than some grand scheme that tries to overhaul the whole road at once.
Rethinking Speed on Roman Roads: Enforcement, Culture and Design
You can’t put a band aid on speeding just by tweaking the road. Enforcement and culture change are essential – but they work way better when you marry them up with designs that pretty much enforce themselves.
The Limits of Traditional Enforcement
Fixing speed cameras on long straights on Roman routes seems to get people to slow down in the immediate area – but as soon as they hit a bit of a break, they’re back up to speed. We saw this on the A5 in Shropshire where fixed cameras only managed to get average speeds down to 55 mph on a 50 mph stretch – it was like they were just playing a speed game.
Average-speed cameras on the other hand have done a way better job. A 2022 pilot on the same stretch of the A5 knocked the mean speed down by 8 mph by measuring the time it takes to get from one point to another, rather than just measuring the speed at one point.
Self-Enforcing Design
The most effective way to get drivers to slow down is to combine enforcement with road design that just encourages slower speeds:
- Narrowing the road so drivers don’t feel as comfy cruising at high speed
- Adding some street trees, parking bays and buildings right next to the road to create some visual clutter
- Using mini roundabouts or chicanes to give drivers a little nudge off their high-speed groove
- Using block paving and some surface treatments to give drivers a visual clue that they’re leaving a high-speed zone behind
- Installing some traffic signals and traffic circles at key junctions
This sort of design works particularly well on roads that have long lines of sight – that’s when you tend to get drivers taking risks.
Policy Trends
The UK is slowly moving towards adopting lower default limits on roads that are hazardous – that is, where drivers tend to speed up and slow down a lot:
- 20 mph is now the limit for 85% of councils on built-up areas and residential roads
- 30 mph is the default on villages where houses front the road
- 40 mph is increasingly common on rural roads where the road bottlenecks
Devon County Council is leading the way here by getting communities on board with 20 mph limits on historic routes and adding some speed reduction measures along the way.
Combined Approaches
The A68 Dere Street safety scheme is a great example of what can happen when you put a combination of physical measures and enforcement into place – like the speed tables at settlement entrances, and some temporary average-speed enforcement along the way. Driver monitoring showed that once the enforcement was taken away, the drivers just kept on going slower – the physical changes had reset expectations.
Local authorities and parish councils have to play a big role in all of this. Community groups along Roman routes often bring a lot of passion to the table when it comes to road safety – they’ve seen the damage that speeding can do, and they’re often willing to put up with some visual changes to make it safer.
Philosophical Fault Lines: Do We Tame Roads Built for Empire?
There’s something pretty awe-inspiring about standing on a 1st-century Roman alignment and knowing that Roman boots once trudged along this very same path. The engineering is undeniable – straight lines carved across the countryside, embankments that have survived twenty centuries…
But there’s also something grim about going to a funeral for someone who’s been killed on the same road – about talking to families who’ve lost someone because a road built for conquest still works exactly as designed.
The Core Ethical Question
Do we owe more to the Roman engineers who built these roads – or to the people whose safety is at risk on them today?
There are two philosophical positions that frame the debate:
Preservationist Purism says that Roman roads are irreplaceable archaeological heritage, so we should preserve them at all costs. Visible changes should be kept to a minimum, and the road should only be altered in extreme circumstances.
Consequentialist Safety says that human life is more important, and if traffic calming tools can prevent deaths, then that’s what we should do. Stones just aren’t more valuable than people – even if we’re talking about ancient stones.
A Middle Ground
I’d suggest a third way: reversible, legible and well-documented intervention.
This approach says that some changes are necessary, are justified, and that we have to do them in a way that’s reversible and won’t damage the road’s origins. That means:
- Accepting that some change is necessary and justified – and making sure it’s reversible too
- Using shallow-depth construction to protect buried archaeology
- Designing it so that any changes blend in with the original road – not clash with it
- Documenting every single intervention, so that future researchers can see what we’ve done
By that standard, we’re not “desecrating” Roman roads when we install speed tables or heritage-compatible surfacing – we’re just updating them: adapting a road built for one purpose for another.* Are we disrespecting history or showing respect for it by calming Roman roads with speed humps or textured speed tables?
- Do you think the Romans themselves – pragmatic military engineers who were constantly on the lookout for ways to improve their designs in response to the local environment – would have frozen their infrastructure against the possibility of making it better?
- Is there more room in our way of thinking about heritage to allow for a kind of dynamic preservation that lets us protect the archaeological value of ancient roads while at the same time allowing safety-critical modifications to be made?
The Romans built roads for a reason: to help their legions move quickly and conquer new territory. Their values just aren’t the same as ours: they wanted to move armies, not people. They didn’t worry about one-way streets or Vision Zero commitments to wipe out death and serious injury. Their roads were designed to serve the needs of an empire – not the people using them today.
But a genuine respect for history might actually mean adapting the old infrastructure to make it work for us in a way that it never did before. A road built to move armies at top speed shouldn’t be holding people back – it should be something that we’ve made safe for them to use.
Conclusion: Turning a Military Corridor into a Safe Street
Roman engineering was incredible – it’s what underpins some of Britains worst speeding problems today. The roads were designed to get legions from A to B as fast as possible – and that’s exactly what they still do. As a result, rural single carriageways have a disproportionate share of fatal and serious collisions all over the country.
But we can’t just wipe the history away. The Roman road network is an incredible achievement in its own right – one of the biggest infrastructure projects in history that was completed without all the fancy tools we use today. It’s worth studying, documenting and yes, preserving.