It suddenly occurred to me that very little of the UK Roman roads have been dualled. Sure, there are some exceptions but the greater proportion remain single carriageway and were never much upgraded.

It can't be because of the expense of urban demoliton, because (again generalising) they go through far less urban area than the later "British" main roads.

Has anyone got a persuasive (and serious) explanation?

Nigel O

The vast majority of our pre-motor age roads have not been dualled, Roman roads are no exception. I'm not sure why it's surprising that most Roman roads haven't been dualled.

Even of roads which have been dualled, most of the dualling has been offline.

Actually, I wouldn't be surprised if a higher percentage of Roman roads have been dualled than non-Roman roads. Much of the A2, A12, A5, A1, A66, A14 (Cambridge-Huntingdon), A46 have been dualled - to my knowledge these follow the basic routes of old Roman roads.

Significant chunks of Ermine Street and Dere Street have been dualled as parts of the A1, and of the Fosse Way as part of the A46

Apart from the places where Roman Roads have been dualled, which others have given you, the main reason that more haven't been is because they wouldn't be useful. They often run through towns and villages, which are no good for dual-carriageways – the A12 shows what happens when you try to mix a dualled RR with bypasses: sudden curves. In other cases, they run through areas that are relatively sparsely populated and have no need of a high speed, high capacity route. In hilly areas, RRs often don't have good vertical alignment, going over hills rather than round them.

Which Roman Roads do you think should have been dualled?

The A38 heading north-east from Lichfield along Ryknild Street is a good example of on-line dualling of a Roman road.

How about the A5?

Several respondents have pointed out bits of Roman raods that have been dualled; I don't quite see how that carries my question forward.

Some respondents have said "Why should they be dualled?" or words to that effect. Well, lots and lots of other roads that started out as single carriageways have been dualled, haven't they.

Roman roads go in straight lines, or go over hills? Don't quite see why that is an objection.

Roman roads go through lots of towns? I think I disagree; of course they linked Roman townships, but a lot of those have not become large in the last millennium. Yes, I know some of them have; but hugely more of the medieval English roads go through (or earlier went through) what have become large tons.

But thanks everyone for your interest.

Nigel O

Looking around the local area, I see the A28 is single carriageway in either direction, the A20 has a small amount of dualling to Maidstone but not much, the A251 is single carriageway all the way to Faversham, and the A259 is infamous for being not just single carriageway, but full of 90 degree turns and level crossings. It doesn't get any better when you go over into East Sussex, which is particularly bereft of any dualled roads east of Lewes. So I don't agree that lots of mediaeval and later roads have been dualled online.

(PS: Shouldn't this thread be on the main roads board?)

Large sections of the A5 have been. Of the single-carriageway sections, most have been replaced by off-line motorway improvements, such as the M1, M6 and M54.

The A5 through the Midlands isn't exactly a prime candidate for major dualling anyway - it bypasses Birmingham, Wolverhampton and Coventry entirely - hence Telford's Holyhead Road being off the line of Watling Street even back in the 19th century. As for the section west of Shrewsbury, well, that's not part of the Roman Road anyway - that ended in Wroxeter, between Shrewsbury and Telford.

Yes, but the vast majority of non-Roman roads have not been dualled, so I don't accept the implication of your question that Roman roads are relatively unlikely to be dualled compared to other roads.

While you're on, here's some more routes which have been dualled which are more or less old Roman routes :
A30/M3
A417/A419
A120
A69 (section immediately west of Newcastle)
A74(M)

So there's actually quite a high proportion of old Roman routes which have been significantly improved.

Medieval roads and turnpikes tend to link towns together - either that is their raison d'etre or they were built that way to connect staging points (horse changeovers) for the mail coaches. Only with the arrival of thye motor car did the concept of a bypass make any sense.

But the Roman Roads often bypass all but the most major Roman settlements - compare Ermine Street north and south of Lincoln with the older "cliff roads" (B1398 and A607) which link the spring-line villages along the escarpment. Thus most Roman roads are not useful arteries but where they are, they can be used for considerable distances with only minor deviations, and an on-line dualling is often possible. Look at the A1, A12, A38, A46, A419.

Two interesting examples where Roman roads have been upgraded from unclassified to A road (albeit single carriageway) in recent years are the A15 between Kirton Lindsey and the M180, and the entire A1500.

Surely part of the explanation is that the socio-political geography of Great Britain has changed almost beyond recognition over the 2000 years since the Romans were here. In others words, what was a key and important route back then may not necessarily be so.

That can be a blessing as long as they aren't too far away - I dread to think what the A46 would be like if it went through Nottingham then onto Hull for instance!

As for hills being bad things - can you imagine the elephant racing which would ensue at about 30 mph on the uphill sections? Modern roads tend to avoid hills as they are the HGV's nemesis. The A46 was bad enough (pre-dualling) with a constant 40 mph nose to tail slog (no chance of overtaking as the other lane was the same) all the way from Bingham to Newark-on-Trent with that speed being scrubbed off on the up hill sections (and this is in the relatively flat countryside of the East Midlands).

The reason surely is very simple. With the obvious and massive exception of Londinium, the towns that were important in Roman times are no longer the ones most important today. Birmingham, Liverpool, Manchester, Newcastle, Leeds, Nottingham, Sheffield, Bristol- these all grew up around the industrial revolution and so were not served by Roman roads.


Rykneld Street from Letocetum (Lichfield/Wall) to Derventio (Derby)is probably the most important dualled Roman road today, because it links Birmingham not only with Derby but also the Sheffield/Leeds/Bradford area. Until fairly recently it also served as the 'missing link' between the M5 and the M1.

I beg to differ! Segedunum was a major Roman fort at the end of the wall, and was a major intersection of the North - South and routes West. There was even a Roman bridge across the tyne.

http://en.wikipedia.org/wiki/History_of ... _upon_Tyne

Manchester (the clue is in the name) was also significant in Roman times. Much of the A64 and A664 are the Roman road from manchester to York, and there are plans to dual part of the Roman Road between Chester and Manchester - the infamous A556 (M6/M56) link.

Yes, there's a Roman Road from Manchester to Ribchester as well. Where it runs through Blackburn with Darwen it has the imaginative name of Roman Road.

Bristol and Nottingham were among the most important towns in the country long before the industrial revolution, though admittedly long after Roman times as far as I know.

Newcastle too is one of those places like New College Oxford and the New Forest that are about a thousand years old.

isn't Bath Road (A4) in London a roman road???

Indeed, but it's only been called Newcastle for less than half the time it has been a major settlement.

The Warden and Scholars of St Mary's College of Winchester in Oxford, commonly called New College in Oxford, was founded in 1379, so not quite a thousand years old. It was new at the time, of course ...