Roman Coins About Britain - Septimius Severus, Caracalla, and Geta - More About the Bridge Coins of 208 and 209 AD

Back to
Does Britannia Feature on Severan Coins?
This page is part of the article
Roman Coins About Britain
Forward to
More About the Profectio Coins of 208 and 209 AD

“The whereabouts of these bridges has been the subject of much inconclusive theorizing”, From Dot to Domesday

We had a quick look at Severus’ and Caracalla’s Bridge coins earlier, but it didn’t really discuss where these bridges may be, which is what we’ll do on this page. There are no new coins about Britain here, so feel free to jump ahead to the next emperor, Victorinus.

Septimius Severus minted a small number of coins in 208 AD showing soldiers crossing a permanent bridge over a river.

RIC 4 786A (As)

Photo Copyright Numismatica Ars Classica NAC AG Auction 106 Part I, Lot 991

Caracalla minted a (now unique) medallion in 209 AD showing soldiers crossing a bridge made out of boats.

RIC 4 441 (medallion). An electrotype copy of the “TRAIECTVS medallion”. Photo © Trustees of the British Museum (CC BY-NC-SA 4.0)

It is commonly thought, although not fully accepted, that these refer to bridges built over the Rivers Forth and Tay in Scotland, which were major obstacles on Severus’ chosen route north. There are a number of competing theories about these coins, ranging from which coin refers to which river, where on the rivers the bridges would have been built, if they actually refer to bridges in Scotland, or if they even refer to bridges in Britain at all.

The Rivers Forth and Tay

Copyright Collecting Ancient Coins

Caracalla’s medallion of 209 AD is generally accepted as representing a bridge built in Scotland as part of his campaigns there, but there is some debate about which river, the Forth or the Tay, it was built over:

Severus’ coins of 208 AD are more controversial, with Reed proposing that they represent a permanent bridge over the Tay, but with Robertson, Elliott, and others, proposing that they represent a bridge elsewhere, such as over the Ouse in York, or even in Rome.

The Evidence

It’s not possible to prove exactly what these coins relate to, but we can look at the physical, historical and chronological evidence to see if we can rule out any of the proposals, and see if any of the remaining contenders seem more plausible than the others.

Topographical Evidence

In terms of topographical evidence, both rivers are easily accessible by sea, which means that the Roman navy in Britain, the Classis Britannica, would have had no problems getting boats, and therefore troops and materials, to the bridge sites.

The suspected bridge sites

Copyright Collecting Ancient Coins

The Forth is 1.78km wide at South Queensferry, but it can be spanned in two segments (1km and 0.78km). The Romans could build physical bridges of this size (Trajan’s bridge over the Danube was 1.135km) but the Forth is up to 60m deep and cannot be diverted like the Danube was. It’s doubtful that the Romans had the technology to build cofferdams to this depth and keep the water out. Locations upstream where it is easier to build have been ruled out by Reed for other reasons (although recent findings mean that might have to be revisited).

The Tay is narrow enough for a physical bridge where the probable bridge head camps are found (Carpow on the south bank and St Madoes on the north bank), and the marshes on either side of the river would probably require a physical structure to cross them. Some quick measurements from GoogleMaps and a nautical navigation chart tells me that it’s currently 390m of mud (including a submerged sand bank), 265m of river, and 270m of mud. Obviously there will be slight variations to this based on the exact route, and it might have been a bit different in Roman times, but it’s clear that you probably wouldn’t (or couldn’t) build a bridge consisting solely of boats here. A substantial physical structure is needed to clear the mud banks, and they make up the majority of the crossing. The Forth also has mud banks at the suspected crossing point, but they are a much smaller proportion of the overall length needing spanned, meaning a bridge there would mainly be made of boats.

Historical Evidence

It’s thought that the Roman army split north of the Forth, with Caracalla taking the larger portion north via an inland route, and Severus taking the smaller portion north via the Tay. If this is correct then Caracalla’s medallion is unlikely to represent a bridge over the Tay as he didn’t travel that way.

Note that Caracalla’s medallion is said to show the two emperors walking side by side over the bridge. This can’t be taken at face value unless the accepted route north by Caracalla is incorrect.

Chronological Evidence

The main argument for disputing that Caracalla’s medallion solely represents a boat bridge over the Forth, and that Severus’ coins represent a permanent bridge over the Tay, is chronology. In fact, that’s the reason why there is some doubt about Severus’ bridge coins relating to the Scottish campaigns at all. There are two issues to resolve; the absolute and relative dating of the coins. I’ll look at the absolute dating first.

Severus’ coins (asses, an aureus and a medallion) are dated 208 AD but they were almost certainly minted at the very start of that year because that’s when the bulk of asses and medallions were minted (“The Role of Medallions“, Curtis Clay, 2007). It’s arguable that the asses and medallion could have been minted later in the year (and that’s a real possibility; RIC 4 807 is an as from 210 with BRIT in the title. That was only adopted in 210 AD, so it cannot have been a Jan 1st issue), but the aureus was minted using an obverse die initially used in 207 AD (“A new bronze medallion of Septimius Severus in 208 AD“, Curtis Clay, 2007) which does rather suggest minting at the very start of 208 AD.

This means that the bridge being commemorated would have been built in 207 AD, which is at least a year too early for even Reed’s timeline, which is the earliest timeline for the Scottish campaigns. He places the preparations in Fife to 208 AD because that’s when Severus arrived in Britain.

At first glance this appears to rule out the physical bridge being in Scotland, but it doesn’t really. The location of Severus himself when the bridge was built is irrelevant unless you expect him to be there building the thing with his own hands. Instead, the only questions that are relevant are:

  1. were the Romans in a position to build a bridge over the Tay in 207 AD, and
  2. did they have a reason to build it?

The answer to these appears to be yes.

Carpow, the fortress on the south bank of the river Tay where the bridge would have been built, is thought to have been under continuous Roman occupation since 185 AD (“The date of the award of the Britannica cognomen to Legio VI Victrix“, Peter Warry, 2006). It was easily accessible by sea, and in the territory of the Maeatae who are thought to have been non-hostile at this time, having been bought off with payments from Rome sometime around 197 AD. The north bank of the Tay might have been hostile territory, but it’s reasonable to expect that the Romans could have held a small area around the bridge head.

Like almost everything to do with these campaigns, there are differing opinions about whether Carpow was occupied or even built before the campaigns, but regardless, it would not require much effort to land a force in friendly territory and re-occupy it, or start construction, on command. The Romans had three legions stationed in Britain before the campaigns, and they had to be given something to do to keep them busy.

It is thought that Severus started planning his invasion of Scotland in 207 AD, but it’s likely that he started earlier than that. After all, he was an ex soldier growing old and bored in Rome. Conquering the whole of Britain was his last big chance of glory, and he had to keep the British legions occupied to stop them revolting again. Regardless of the exact date, it wouldn’t take long to get an order to Britain ordering the construction of a bridge over the Tay which would provide a route accessible by sea into the area of the Caledonians where the campaign would be fought.

It seems, therefore, that the issue with the absolute dating of Severus’ coins is not an issue at all. The Romans had the ability and a reason to build, and hold, a permanent bridge over the Tay in 207 AD.

That leaves the relative dating of the coins. The issue here is that Severus’ coins for the permanent bridge over the river Tay were minted at the very start of 208 AD whereas Caracalla’s medallion of the boat bridge over the Forth was minted a year later at the very start of 209 AD. This appears to be out of sequence because the Tay is further north than the Forth. The obvious assumption is that the southern most bridge should have been built and commemorated first. There are a number of reasons why this might not be true:

  1. Boat bridges can be assembled much more quickly than permanent bridges, so work only needs to begin shortly before the bridge is needed. i.e. just before the army needed to cross it. Severus may have ordered work to start on the physical bridge at the Tay well in advance to ensure it was complete when the army finally arrived. He might also have ordered it to be built early to help secure the northern bank of the Tay where it narrows near Carpow. The fleet, and the fortress, would be at risk from the Caledonians once they realised an invasion was being planned
  2. The land based route to the north that Agricola used doesn’t require a large bridge over the Forth. Severus and Caracalla could have used that at the start of the campaign when they were just moving the army north for the first time. It’s not ideal because it took the Romans further away from their fleet and their base at Cramond, and it left their supply line exposed, but these wouldn’t have been significant issues as they were crossing into non-hostile territory. The boat bridge over the Forth was therefore not essential while the Maeatae were still non-hostile to the Romans. Alternatively, it’s not unreasonable to assume that the navy, who had 900 ships at their disposal (Elliott), could have ferried troops and equipment across at the start of the campaign. The bridge itself could then have been built later, perhaps to aid ongoing troop and supply movements, or to provide a shorter route for troops returning from the front
  3. Caracalla’s coin with the boat bridge was actually a medallion, and as such, was likely to be a new years gift distributed on the 1st of January 209 AD, possibly to those who had built or operated the bridge. Just because the “thank you” gift was dated 209 AD, there’s no reason to assume the bridge was built then. It could easily have been built in 207 or 208 AD and rewarded at the start of 209 AD when it had achieved its goals

These arguments show that the bridges could have been built in either order, and that the date on Caracalla’s medallion could be disconnected from the year the bridge was built.

Now it’s not my intention to suggest that any of these counter arguments are in any way correct. Instead, I just want to highlight the fact that there are multiple timelines possible for both building the bridges and issuing the coins, and we don’t know enough to say which one is correct. Therefore, we can’t say that the Severan coins must relate to a bridge in England or Rome just because of the early date, and we can’t say that the boat bridge must have been on the Tay just because Caracalla’s medallion had the later date. We just don’t know enough to say that Reed’s analysis is incorrect just because the coin dates are in the wrong order.

Alternative Theories

The analysis above shows that Reed could be correct. Severus’ coins could have been minted to commemorate a permanent bridge over the Tay, and Caracalla’s medallion could have been minted to commemorate a boat bridge over the Forth. That does not mean that they did. It’s a valid interpretation, but it’s not the only one.

Caracalla’s Medallion Represents Bridges over the Forth and the Tay

Simon Elliott suggested that Caracalla’s medallion represented boat bridges over both the Forth and the Tay. If the proposed route taken by Caracalla is correct, then the bridge over the Tay is a bridge he didn’t build, didn’t cross, and probably didn’t even see. In that case, would he bother commemorating it in a special medallion?

Desnier (“On the Bridge on a Coin of Septimius Severus – AD 208“. J. L. Desnier, 1997) said it is “obvious” that bridges built during the campaign, which I take to mean bridges across the Forth and the Tay, would be boat bridges. This would be a reasonable assumption to make based on the width (and in the case of the Forth, the depth) of the rivers, and the fact that permanent structures weren’t needed for a short campaign, but we’ve seen that the marshes at the edges of the Tay were wider than the river itself, so a large permanent structure would be needed on either side just to reach the boats.

It could be the case that the Tay bridge was a hybrid permanent/boat structure but that might be unlikely because of the nature of the fortress at Carpow. It’s on the south bank of the river, with hostile territory on the north bank. The river approach is narrow enough that it’s likely that a permanent bridge was needed here to allow the Romans to control both sides of the river and protect the fleet and fortress. While the campaigns might have been short, Carpow was built to last, and some long term way of controlling both sides of the river must have been planned.

I’m not convinced that Caracalla’s medallion represents both bridges.

Severus’s Coins Represents a Bridge That’s Not in Britain

Some people have suggested that Severus’ coins represent a bridge elsewhere, either in England or in Rome, but I’ve yet to find a convincing case made for any particular bridge. The historical sources don’t mention either emperor building bridges elsewhere at this time, and while an unknown bridge built in England (say York or Newcastle) as preparation for the campaigns is plausible, Desnier’s reasoning for the Milvian Bridge in Rome seems the weakest of all arguments so far. Despite saying:

“…even if it is not found on what we know is the triumphal route into the city…”

he goes on to say:

“…Septimius Severus could envisage the celebration of triumph in Rome in which the Milivan bridge … could figure as the gateway into Rome”

No other emperor seems to have included it on coins despite there being many celebrations of triumph in Rome. In fact, only Trajan (RIC 2 569 As, RIC 2 569 Sestertius, RIC 2 569 Dupondius) and Marcus Aurelius (RIC 3 270, RIC 3 1047, RIC 3 1048) had previously used bridges on their coins, suggesting that a bridge would have to be particularly significant to be worthy of celebrating on coin. Trajan’s bridge over the Danube was 1,135m long, and there’s no candidate physical bridge related to the British campaigns anywhere close to this long (the crossing at the Forth is 645m wider, but too deep to allow a physical structure at that time).

I’m not convinced that the physical bridge, if there is one, is anywhere other than Scotland.

Severus’s Coins Represent a Small and Simple Bridge

We are assuming that the permanent bridge on Severus’ coins represents a significant and ornate bridge somewhere, but it might represent something much simpler. It’s propaganda on a coin, not a photograph. The design itself was copied from one of Trajan’s coins, and Numismatica Ars Classica say:

“Severus’ coinage frequently took typological cues from the coins of earlier ”good” emperors as a means of connecting his reign to the less troubled times of the Antonine Age. By alluding to a coin associated with Trajan’s Dacian Wars, Severus’ Caledonian campaign and its constructions were given an elevated status and Severus was cast in the same light as the Optimus Princeps.”

Perhaps we shouldn’t take the ornate nature of the design, such as the triumphal arches decorated with statues, too literally. It’s possibly the case that Severus was using artistic license to commemorate a large bridge built in a much plainer style. Note that the bridge on Trajan’s coin was not representative of the bridge he actually built.

The reason this is important is that one of the reasons for rejecting a bridge in Scotland is that they didn’t have time to build a large ornate structure by this stage in the campaigns. For example, Robertson’s 1980 paper said:

“if the permanent, monumental, bridge was in Britain at all, it looks like a long established bridge, such as there would hardly have been time to build in either AD 208 or 209”

Another argument is that they didn’t have a reason to build an ornate bridge during an active campaign. I’ve dealt with the time argument earlier, but it becomes more realistic to assume a bridge in Scotland if it doesn’t have to be ornate, and in that case, a bridge over the Tay would be a candidate.

I believe this could be a plausible explanation.

Severus’s Coins Represents All Engineering Work Done in Preparation for the Campaign

“But as he advanced through the country he experienced countless hardships in cutting down the forests, levelling the heights, filling up the swamps, and bridging the rivers” Cassius Dio, Roman History LXXVII Chapter 13

“He especially saw to it that dikes were provided in the marshy regions so that the soldiers might advance safely by running on these earth causeways and fight on a firm, solid footing.” Herodian, History of the Roman Empire Chapter 3.14.5

An alternative proposal is that Severus used a bridge to commemorate all of the completed and planned bridges, dykes, causeways, earthworks, and other engineering tasks that both Herodian and Cassius Dio make a point to mention in their short descriptions of the campaigns.

If these tasks were celebrated at all, then it’s likely that it would be done with general coinage rather than Elliott’s suggestion of a single medallion issue by Caracalla.

I believe this could be a plausible explanation.


It’s probably fair to say that we’ll never know for sure if these coins represented bridges in Scotland, never mind knowing where they would have been. However, their dates coincide perfectly with the British campaigns which were such a large undertaking that it’s hard to believe that the coins could represent something else. The fact that they were celebrated on coins suggests that they were important undertakings related to the war, rather than a run-of-the-mill bridges constructed in safe territory.

The arguments against Severus’ coins being a bridge in Scotland do not stand up well to scrutiny. There was time to build a large permanent bridge over the Tay because it’s not necessary to insist it to be monumental, and not necessary to assume that work could only begin once Severus himself arrived at the building site, ready to clock on for a shift of hard graft. There was a reason to build a permanent bridge over the Tay, and reasons why a boat bridge wouldn’t have been suitable.

The arguments for Caracalla’s bridge to be over the Tay don’t fare well either due to the topology of the river. A one-off boat bridge over the Forth, where a permanent bridge would have been impossible and unnecessary, makes much more sense.

Article Navigation

Back to
Does Britannia Feature on Severan Coins?
Up to
Septimius Severus, Caracalla, and Geta
Forward to
More About the Profectio Coins of 208 and 209 AD
Please complete the required fields.