Newcastle (Pons Aelius) Bridge, Vallum and Fort (info needed

[D B Campbell]

Something that eludes me however, was Newcastle unique along Hadrians Wall to be set apart from the wall itself? And was it the only one to be built after the wall itself?

Carvoran lies behind the line of Hadrian's Wall; it has produced Trajanic material and is generally accepted to predate the Hadrianic system. Newcastle has never produced any material this early.

Like Brian said, the small size of the fort was offset by the milecastles and turrets nearby.

Not sure how a fort can be "offset" by milecastles and turrets. It's maybe there to supply/support the milecastles and turrets in that sector. More likely, there's a strategic purpose.

Also, I don't know much about the Antonine Wall, but how did troops garrisoned on those forts go beyond the wall? Were there milecastles?

Through the fort gateways. And there appears to have been a system of fortlets similar to the Hadrianic milecastles (although no turrets have ever been discovered).

And then there's the outer limes in Germany, where the forts all lie behind the frontier line and fortlets are provided haphazardly. (I think we need a "Frontiers" issue of Ancient Warfare magazine, Jasper! )

[MARCVS PETRONIVS MAIVS]

Carvoran lies behind the line of Hadrian's Wall; it has produced Trajanic material and is generally accepted to predate the Hadrianic system. Newcastle has never produced any material this early.

How far back did it lie?

Not sure how a fort can be "offset" by milecastles and turrets. It's maybe there to supply/support the milecastles and turrets in that sector. More likely, there's a strategic purpose.

Sorry Duncan, I did not make myself clear. What I meant by ''offset'' was that the small number of barracks was partly offset by the nearby milecastles and such which contained smaller barracks, as Brian explained once. Paul Bidwell has also expressed sympathy for this view. It may also have been possible there was another fort at Gateshead, which may have also been of smaller size, helping to house the rest of the contingent.

Through the fort gateways. And there appears to have been a system of fortlets similar to the Hadrianic milecastles (although no turrets have ever been discovered).

I don't understand. You mentioned beforehand the forts did not connect to the wall, so how did they exit beyond the wall through the fort gateways if the fort didn'c connect to it?

And then there's the outer limes in Germany, where the forts all lie behind the frontier line and fortlets are provided haphazardly. (I think we need a "Frontiers" issue of Ancient Warfare magazine, Jasper! )

I am yet to subscribe to Ancient Warfare but will do so very soon.

[PhilusEstilius]

I can not accept this idea that the fort at the Pons Aelius was built at a later date, I don't know who came up with this kind of thinking for we are clearly informed that the bridge was built by Hadrian and there would have been a fort to support it's defence. There are four major Roman river crossing points over the river Tyne and they all have a fort on the north bank of the river to give support. I think that when archaeology can show the NW and NE corners of this fort, then and only then can it be stated that the fort is behind the Wall.

[D B Campbell]

I don't understand. You mentioned beforehand the forts did not connect to the wall, so how did they exit beyond the wall through the fort gateways if the fort didn'c connect to it?

No -- I said the Antonine forts don't straddle the wall; i.e., they don't protrude the way many of the Hadrian's Wall forts do. So there is only ever one gate exiting to the north, unlike some of the Hadrianic forts which have three gates beyond the barrier line.

Most of the Antonine Wall forts are bonded into the wall; i.e., they utilise the Antonine Wall as their northern rampart. (There are notable exceptions, of course; e.g. Balmuildy was completed before the Antonine Wall reached it, and Bar Hill stands alone behind the Wall.)

This bonding of forts into the Antonine Wall is subtly different from the Hadrianic scheme. For example, at Housesteads, Hadrian's Wall buts against the fort corners and utilises the fort's northern rampart; if you "erased" Hadrian's Wall, you'd still have a perfect Roman fort. Whereas if you erased the Antonine Wall, many forts (e.g. Rough Castle) would lose their northern rampart! Others (e.g. Balmuildy) would exhibit protruding wing walls indicating that a barrier was expected.

But I fear we are straying off topic!

(And thank you for your PM, Yuri.)

[D B Campbell]

I can not accept this idea that the fort at the Pons Aelius was built at a later date, I don't know who came up with this kind of thinking for we are clearly informed that the bridge was built by Hadrian and there would have been a fort to support it's defence.

I'm not sure that we're "clearly informed", Brian. Unless you know of other evidence, all we have is the bridge's name, "Aelian Bridge" (Pons Aelii, or Pons Aelius), which is probably a reference to Hadrian. But not definitely a reference to Hadrian -- Francis Haverfield, the great epigrapher and father of Romano-British archaeology, noted that "Aelius" could equally refer to Antoninus Pius or even Commodus.

So it might be safer to say that the bridge is probably of Hadrianic origin.

There are four major Roman river crossing points over the river Tyne and they all have a fort on the north bank of the river to give support.

I think this is why an early fort has been postulated at Gateshead, where mid-second C material has turned up (but no structures afaik). Crucially, there is a flaw in your logic: the existence of a bridge does not prove the existence of a fort at Newcastle.

I can not accept this idea that the fort at the Pons Aelius was built at a later date, I don't know who came up with this kind of thinking ...

I am currently holding Archaeologia Aeliana vol. 31 (2002), entitled "The Roman Fort at Newcastle upon Tyne" (Thanks, Ross!), so I can probably shed some light on "who came up with this kind of thinking"!

It is at least suspicious that the earliest inscription from the fort dates from AD 213. Equally, the earliest stratified coins (i.e. excluding stray finds) dates from AD 260. The coin specialist has suggested that the finds are typical of an early third century foundation (R.J. Brickstock, "The Coins", Arch. Ael. 31, 2002, pp. 175-209, at p. 181).

But, in fact, the crucial evidence is the pottery assemblage, which is overwhelmingly Antonine, as opposed to Hadrianic. The construction levels of the fort have produced 65% BB2 (black burnished ware), normally associated with Antonine occupation and not found elsewhere in northern England until the AD 150s. (Remember: Hadrian's Wall was built in the AD 120s.) In addition, the (very) small quantities of samian ware are of a later 2ndC date which "may suggest that the fort was founded late in the century, or even slightly later" (P. Bidwell et al., "The Roman Pottery", Arch. Ael. 31, 2002, pp. 139-172, at p. 167).

On balance, the evidence, such as it is, supports the view that the fort was not a primary Hadrianic construction, but was added later, perhaps much later. Anyone who wishes it otherwise would really need to come up with some good arguments!

[PhilusEstilius]

I would still say yet that even with some of the archaeological finds that you have put forward Duncan, this still does not put any later date for the construction of this fort and indeed does not show in any way that the fort stands behind the wall. The information you give us from Haverfield is only his suggestions and nothing more, he appears to be like so many other archaeologists he makes statements and other such suggestions but does not commit himself in any way.

[D B Campbell]

I would still say yet that even with some of the archaeological finds that you have put forward Duncan, this still does not put any later date for the construction of this fort ...

That's archaeology, Brian. Without a sealed foundation deposit, we are limited to the balance of probabilities.

The experts who have investigated the Newcastle site recognise that the finds assemblage is quite different from a standard Hadrianic assemblage. Either the Hadrianic garrison was incredibly tidy and managed to erase all traces of their presence, or we take the finds at face value and accept that they suggest a later foundation than other forts along Hadrian's Wall.

... and indeed does not show in any way that the fort stands behind the wall.

That was not my purpose here. I was answering Yuri's question about dating, and your own statement, that

there is not a fort anywhere on Hadrians' Wall that is not attached to it in some way

(see above regarding Carvoran).

The main points, I would say, are: (1) frontier forts do not need to be connected to a frontier barrier, so there's no problem in having a detached Newcastle fort; (2) the terrain around Newcastle makes a link between fort and Wall problematic; and (3) the finds assemblage from the site strongly suggests an Antonine or even third century foundation.

The onus is on anyone who disagrees with these points to present a counter argument.
Namely, (1) why does the fort have to be attached to Hadrian's Wall (when other frontier forts are not), (2) how could the fort, physically, have been attached to Hadrian's Wall (which seems to lie some way to the north) , and (3) why is the material assemblage from the site much later in date than the assemblages from other Hadrianic forts?

As someone once said: Occam's razor is your friend. Don't fight it.

[mcbishop]

As someone once said: Occam's razor is your friend.

Certainly my principal weapon of choice.

Mike Bishop

[PhilusEstilius]

It becomes just a bit strange all this discussion about where the wall is when infact it has never been found in the city centre nor even it's foundation. therefore how the hell can anyone make statements of what is where with regard to the Pons Aelius.

[MARCVS PETRONIVS MAIVS]

how could the fort, physically, have been attached to Hadrian's Wall (which seems to lie some way to the north)

Hi Duncan,

According to that 1985 sketch you provided me (I again forgot who drew it), the scale seems to indicate the fort was located 5 meters behind the wall, if we interpret (as I did a few pages back) that the big black patch to the north is hadrians wall (the one he claims is the northen rampart of the fort) and the smaller black patch 5 meters down (which he doesn't label) as the northern rampart of the fort.

So according to this, I would not say 5 meters is too far behind. But again, like Brian said, untill someone discovers the wall itself, all we have is speculation, interpretation and conjecture.

In any way, would a fort so close to the wall not be connected to it in any way? I mean, 5 meters is nothing, would this not warrant a small curtain wall or bridge connecting to the wall so the troops would not have to go to the nearest milecastle or turret to man it?

Slightly back to topic though, the reconstruction is nearly complete and will be finished within the next week or two. Although my personal opinion (like Brian's) is that the fort connects to the wall (even if it was built at a later date, I see no reason why they would not have used the wall as the northen rampart - off course this is only valid if my interpretation that the fort is only 5 meters behind the wall is correct. It would not make sense to build a fort so close to the wall and not have it connect). Although that is my personal belief, the current evidence available (or lack thereof) would seem to suggest the fort was free standing.

As such, one must sometimes put personal feelings aside and accpet circumstances which are not to ones liking. I indeed hope future archaeologists prove me right, but the current reconstruction will feature a free standing fort, disconnected to Hadrian's Wall.

Again, thanks for all your imput and invaluable help.

Best Regards,
Yuri

[D B Campbell]

According to that 1985 sketch you provided me (I again forgot who drew it), the scale seems to indicate the fort was located 5 meters behind the wall, if we interpret (as I did a few pages back) that the big black patch to the north is hadrians wall (the one he claims is the northen rampart of the fort) and the smaller black patch 5 meters down (which he doesn't label) as the northern rampart of the fort.

Hadrian's Wall does not appear in John Nolan's 1985/6 site plan, Yuri. The "big black patch" is the fort wall, comprising a footing (measured at 2.10m wide) and a first course (1.70m wide). In 1992 excavations, a gravel street (the so-called via sagularis) was found running along behind, apparently 3-4m wide.

You can see that the fort rampart must have been running approximately NW-SE, whereas Hadrian's Wall generally runs WSW-ENE in this sector.

But again, like Brian said, untill someone discovers the wall itself, all we have is speculation, interpretation and conjecture.

Don't forget topography. The fort doesn't exist in a vaccuum, but must be sited on the ground. The problem at Newcastle is the nature of the ground. The position of the fort is isolated on a promontory, cut off from the north by the Lort Burn. Early observers saw the Hadrian's Wall curtain to the north of this burn, whereas the fort's northern rampart lies to the south.

That clinches it for me. There must have been a good 70m between fort and Wall.

[PhilusEstilius]

I would say Duncan that your compass headings are are a little mixed up some how, you have the fort rampart running NW-SE which cannot be correct. For if the excavation drawing is correct the walls of the inner buildings are shown to be east west and north south, therefore the north and south ramparts of the fort have to be E-W. Then of course we have to consider the headings you give for Hadrians' Wall, you are not wrong with the ENE where the Wall comes to the Pons Aelius from Segedunum however your WSW would put Hadrians' Wall into the river Tyne in the first mile after leaving the Pons Aelius. The direction of Hadrians' Wall leaving the Pons Aelius is indeed WNW as it travels up Westgate Road towards the fort of Condercum on the north side of the Tyne Valley where it continues in this direction until it reaches Wallbottle where it makes it's first major change to a westerly direction as it continues on to Heddon........I think that I shall stay with J Collingwood Bruce who considered that the Wall probably joined the fort at it's NE and NW angles.

[mcbishop]

I think that I shall stay with J Collingwood Bruce who considered that the Wall probably joined the fort at it's NE and NW angles.

Collingwood Bruce was following Horsley, as he readily admits, stating

'The Wall, passing through the site of St, Nicholas'-church, would, of course, be its northern boundary' (The Roman Wall, 1851, p.125)

Note the conditional - 'would' not 'was', because there was no evidence upon which to base this surmise, only what they could deduce by comparison with other sites (and produce the attached reconstruction). There certainly appeared to be empirical evidence for the course of Hadrian's Wall on Collingwood Street (which CB cites), but not of the fort being joined to it. Modern archaeologists using modern techniques, not hearsay, have excavated the north wall of the fort (see my posting somewhere several miles above) and found part of a building to the north of it.

At its closest, as Collingwood Bruce suggests above, Hadrian's Wall probably followed the line of Collingwood Street and passed close to St Nic's. A good 100m to the north of the north wall of the fort identified by its excavators (I suspect at least some modern scholars would place it slightly further south than that, but only by the width of a cathedral, still leaving 50m of clear air between HW and the north wall of fort). I see no reason to doubt the excavators' delineation of the fort, and plenty to carp at with Horsley's and Collingwood Bruce's conjectures, but they are not to blame over this as they simply did not have the evidence to work with, as somebody had been inconsiderate enough to plonk a medieval town on top of the fort; now, however, we have the north wall of the fort.

As Holmes tells Watson 'It is a capital mistake to theorize before you have all the evidence. Insensibly, one begins to twist the facts to suit theories, instead of theories to suit facts. It biases the judgment.' The north wall of the fort is an archaeological fact and it does not, on the available evidence, coincide with the line of Hadrian's Wall. Nor is there any evidence that there was any sort of spur wall linking the two.

Which ever way you cut it, application of a glinting, keen-edged Occam's Razor suggests the fort was not attached to the Wall.

Mike Bishop

[MARCVS PETRONIVS MAIVS]

Hadrian's Wall does not appear in John Nolan's 1985/6 site plan, Yuri. The "big black patch" is the fort wall, comprising a footing (measured at 2.10m wide) and a first course (1.70m wide). In 1992 excavations, a gravel street (the so-called via sagularis) was found running along behind, apparently 3-4m wide.

Oh, some pages back I presented a a sketch offering my interpretation of John Nolan's sketch and I thought you had agreed it could have been Hadrian's Wall. It may have been Brian though. I am short of time at the momment and cannot check properly. There are however some discrepancies in that sketch as has been previously pointed out, one of them being the fort seems to large, and its southern rampart would fall of a steep slope.

Don't forget topography. The fort doesn't exist in a vaccuum, but must be sited on the ground. The problem at Newcastle is the nature of the ground. The position of the fort is isolated on a promontory, cut off from the north by the Lort Burn. Early observers saw the Hadrian's Wall curtain to the north of this burn, whereas the fort's northern rampart lies to the south.

I remember reading somewhere that Hadrian's wall may have gone over these small streams with some sort of drainage system incorporated into the wall.

That clinches it for me. There must have been a good 70m between fort and Wall.

Like I said above, my interpretation may be incorrect. If this is true, then I have no trouble believing the fort is free standing. I only had doubts if the fort was so close to the wall (as I mentioned in my sketch)

[MARCVS PETRONIVS MAIVS]

I think that I shall stay with J Collingwood Bruce who considered that the Wall probably joined the fort at it's NE and NW angles.

In an ideal world we should simply be able to excavate the fort's corners and see whether there was a big stone wall attached to them or not. I feel this would make our life and this discussion much simpler :wink: