I think I understand how the Roman's built the Pont du Gard, because it's simple in principle. Once they'd cracked how to build a Roman Arch (semi-circular), the rest is boiler-plate. Big and impressive but not high-end engineering compared with a Victorian elliptical arch or the 1915 Çanakkale Köprüsü, where the centre span of the bridge is over 2km long.

To me the surveying that decided the route the aqueduct would take is much more impressive, but I think I know how if might have been done using relatively simple techniques. (Simple when you know how!)

Beyond my understanding is how the Romans laid out long straight roads between far distant points. Looking at the Fosseway, we have a straight road connecting Lincoln to Ilchester, with a kink at the end extending to Exeter. On the almost dead straight section lie Leicester, Cirencester and Bath.

Once the bearing between the two ends is established, a straight road is easily laid out by aligning poles. (New front poles are aligned visually with the two poles immediately behind, and the road stays straight with good accuracy.) But how did they establish the bearing in the first place? Today the bearing of Ilchester from Lincoln can be got in a few minutes from a modern map, but the Romans didn't have that sort of map. Or compasses. Short straight stretches of road are easy, but does anyone know how the Romans surveyed a long straight road between very remote locations? It's a long way from Exeter to Lincoln!

Building physical objects like the Pont du Gard is amazing enough, but I'm more impressed by the thinking that decided where the aqueduct needed to go relative to a water source and a town, given that difficult country lies between the two. They will have worked out how much it was going to cost too…

Dave

Dave

In my experience Roman roads are straight where you can see from end to end, but only straight-ish elsewhere. Much straighter than the rolling English road made by a rolling English drunkard, though. One factor was the freedom that the invaders had to put the road where they wanted, in a countryside which was, effectively, empty of roads.

And have you considered why the straightness was needed? In flat country, it does give the 'quickest' or shortest route, but was that the only reason? One question is 'How were Roman vehicles steered?, or Have you ever seen a picture of a Roman Wagon with any steerring scheme at all?'

I await your answers …

Cheers, Tim

There were probably very few villages and towns already in England when the Romans arrived. Just scattered Bronze Age settlements with huge areas of forests between them. There were some tracks joining these settlements, often along ridges; but nothing like the Saxon and later density of villages and tracks we see now.

A lot of the route-finding might have been by view-points; less effective of course on plains, but many Roman roads do go over hills rather than round them.

(As I found when using the Fosse Way from fairly near home to the Warwickshire Exhibition Centre!)

The map Dave shows has Dorchester marked on it. This (as its name suggests) was a Roman settlement only a few miles inland from Weymouth, which was not there then but the Romans did use the Wey estuary as a harbour.

These were linked to the Fosse Way somewhere near what is now Yeovil, more likely Ilchester; its route now the A37 then onto Weymouth as the A354. Among the many Roman traces around Dorchester is parts of its water supply. This was an aqueduct taking water from the R. Frome a few miles upstream of the town, and at one point loops round the flanks of a small side-valley now carrying part of the Dorchester bypass.

I invite Bazyle, and anyone who thinks he is right, to offer evidence of the claim that there were villages and towns all linked by roads, before the Roman invasion.

Tim

I tought u just needed 3 poles and a length of string…..not Poles as in Polish….

we cant even build HS2……

The trams in Edinburgh are about 8 miles long and took 10 years

Hadrians wall is 73 miles long and took about 7 years

What really impresses me is that there weren't really that many humans about 2000 years ago and yet the Romans built these vast projects across all of Europe and maintained a food supply system to keep it all going

Their organising and co-operation skills were 1500 years ahead of anyone else

Rod is right – thanks. Anything that was even slightly heavy went by boat. Look at the placing of the major Roman towns in the UK – almost every one is as far up the river as the boats could get. And of course, that was where a river crossing developed, often a ford to start with, sometimes a floating bridge.

Exeter, Dorchester, Winchester, Chichester, Colchester, Gloucester, Chester, Lancaster – and that is just settlements still carrying names based on Roman activity there.

Cheers, Tim

Laying out the road's line would have been relatively easy given clear countryside, once they know which way to go, but much of Britain was still heavily forested at the time. So the difficulty would be working through the woods when all you see is the trees.

I think they did have wheeled carts, but single-axle. Otherwise they would have used pack-animals.

Although military purposes were first and foremost, the Romans and Romano-British would have used the roads for commerce and civilian "traffic" too.

'

Ilchester is near Yeovil, about half-way between Bath and Dorchester, and marked the junction between two roads: the Fosse Way's continuation to Exeter, and the road to Dorchester, thence South to Weymouth. The Roman port is thought to have been up-river at Radipole, at least for small vessels, about 3 miles from the estuary.

To join Aqvae Svlis to Lindinis (Ilchester) means climbing over Eastern Mendip and a smaller but quite sharp ridge, giving the surveyors long views ahead; but then crossing a wide plain, an inland continuation of the Somerset Levels, to the fort and village established by the Romans on that plain's Southern fringe. West is more plain for some distance then the country becomes mainly hills, but I don't know how the Romans dealt with those. South reaches Dorchester by going over the Dorset Downs (along now, the A37).

The A354 from Dorchester goes over the Ridgeway escarpment, and the Romans' steep direct line is now just a track. The modern road deviated via a hairpin-bend, supplanted in turn in 2012 by a new road through a big cutting that might have been deeper and less steep if the LSWR or GWR had not got there first and put a railway tunnel though the hill.

So several land-forms for the Romans to cross, with a lot of forest cover and some very moist areas that might have been rich in willow woodland.

'

What is notable is how the post-Roman to modern roads sometimes deviate abruptly from their Latinate guides, even where the latter is still evident and the deviation has no obvious topographical advantage.

An example is on the A37, coming South from Shepton Mallet. A couple of miles from Ilchester bypass, the road turns very sharply right then wiggles round a bit before regaining the proper Via Lindinis; which runs straight on from the first bend, as a farm track. A second, modern deviation thanks to the bypass makes the original pop out in the village as a surfaced but "No Through Road" lane.

as long as you have a vague idea of the relationship of 2 towns and enough people and gromas it works. Say you want a road from Newcastle to Chester. You know that Chester is south and west of Newcastle, so team A sets off south from Newcastle and team B sets off east from Chester. They leave observers at suitable high points along the way. Sooner or later their paths will cross, and they can start iterating to get all the observers in line. I bet this second phase wouldn't take as long as you think. All the observers are moving at the same time, and are converging on the correct line. If they run out of high points they might have to set up in between stations, but they will be getting closer all the time anyway. Even if you don't know the east/west relationship you can have another team and a third line to the west from Chester. You can get a good enough initial South and East from the sun, Once you've set off you can back sight to keep in a straightish line, I've done that in thick mist on mountains, it's surprising how well it works.

Tim –

Re placing towns on rivers. For trade and travel that certainly makes sense where the river is navigable but not all Roman towns are on estuaries or inland waters that are. It also makes sense if the river can supply reasonably clean water. Hot and cold running water, indeed, in Bath – which is also far enough down the R. Avon for boats to arrive from the R. Severn.

London and Exeter are probably among those at the lowest bridging point or ford but also sensibly close to the sea.

Ilchester and Dorchester are close to the Rivers Parret (I think) and Frome respectively, easy to span, but not so close for flooding to be a serious risk. In fact Dorchester is quite high above the river, necessitating a long aqueduct from some distance up-valley for the water-supply. Both settlements are many miles inland and neither river is navigable.

Dorchester is some 20 miles up-stream from the Frome's outlet in Poole Harbour. It nearest sea-port appears to have been Radipole, near Weymouth.

Chichester adjoins an estuary but I don't know how feasible it was as a port. Places like Chichester and Radipole may have accommodated only small coastal craft and fishing-vessels, not larger ones from France or round Iberia.

The Fosse's route suggests to me that the Romans were carefully finding their way SW from Bath between a big area of marshes (the Somerset Levels) and a lot of hill country though they still shortened the distance by crossing Mendip . If the land was a bit higher and drier we have expected the road to have been more direct.

…

Incidentally there is one road in that region that puzzles me by its location, and I don't think it's Roman. It is the A39, which descends from the Polden Hills then determinedly but needlessly climbs steeply over a hill spur at Puriton, instead of taking the obvious contour round it, still above flood level!

…

The original question though was about building materials.

I wonder if we can make concrete last 2000 years but won't because over the last 100 years or less there has developed an idea that a building is old and ripe for demolition at only 50 years; removing any need to consider quality for longevity.

This might be re-inforced by the comparative building methods. Until the late 19C big public buildings were slow and laborious to build, so it would have made sense to make them to last as long as possible. Whereas now there seems an assumption that it is now so relatively easy and rapid to erect big buildings that it is too easy to think them disposable.

So why make concrete that will still hold up 1000 years hence, when it's likely to be converted to hardcore in well under 100?