Roman Britain was a single province from AD43 to sometime after Septimius Severus won control of it from Clodius Albinus, that is about AD198. Then two provinces were created: Britannia Superior (capital London) and Britannia Inferior (capital York). The use of the ranking adjective was also used for Germania and Libya. In Germania it referred to the relative height of the land, and we still use the terms High German and Low German.
However, the north of England is not lower than its south, but the reverse, so we can only suppose that the terms meant ‘better Britain’ and ‘worse Britain’. Had the terms meant ‘hither’ and ‘further’ (from Rome) they would have been citerior and ulterior. But they weren’t.
We know nothing about the boundary of the two provinces, other than each capital must have lain within its territory. It is customary to divide Britannia at the Mersey-Humber line (which runs from Liverpool on the West to Kingston upon Hull in the east), but there is no evidence to support that.
Another possibility is the approximate line of the Severn-Humber economic divide (SHED), a much more significant and ancient division. This is clear from Later Pre-Roman Iron Age (LPIRA) contrasts and remains potent to this day, as seen in the work of the Social and Spatial Inequalities Group (SASI Group) (see www.sasi.group.shef.ac.uk/maps/nsdivide/) at the University of Sheffield. For anyone not familiar with the geography, the Severn Estuary runs into the Bristol Channel, a wide tidal piece of sea which
In the LPRIA period, coins were used south and east of the SHED line (the Lowland Zone, or LZ) and not north and west of it (the Highland Zone, HZ). Similarly various types of pottery were made and distributed in the LZ, and no pottery has been found in the HZ. It is also the divider for LPRIA forts versus no forts, with all forts again found in the LZ south and east of the line (Millett, 1992, various pp). Todd notes that certain architectural features such as souterrains exist only in the HZ, north west of the SHED line (Todd, 2008, p.20) North and west of the SHED line agriculture is to this day dominated by stock rearing, as opposed to the mainly agrarian nature of south and east.
The SHED line was also the line of the Fosse Way, for a generation the effective border of the Roman Empire in Britain, so the earliest Roman province was the LZ, or a part of it.. Naturally it formed the tribal borders of a number of Brythonic political units. Consequently it also forms an isogloss, because the people to the north and west of that line would operate in different discourse communities to those south and east of it. They would not talk of minting coins, how to make or sell ceramics, or techniques for building or rebuilding earthworks.
The line itself runs approximately along the Jurassic Ridge, which is a watershed for local rivers. It may have had a negative impact on settlements, as here is the highest level of the natural radioactive gas Boron. This runs along the SHED/ Jurassic Ridge line and it is likely that fewer settlements would be made along that line, since the gas permeates through floorboards and has a long-term adverse effect on health, even today. Gas, river access and proximity to potentially hostile neighbours would have reinforced the divide.
All of this leaves me in no doubt that this was the border between Britannia Inferior and Britannia Superior, as overseen by Septimius Severus after beating Clodius Albinus. He stripped the governors of the power to command armies and stripped the armies of the power to levy taxes, thus creating checks and balances.
At some point between AD198 and AD305, the term ‘Britannia’ came to be applied to Britannia Inferior only, and the LZ became known as ‘Caesariensis’. Birley suggests this may have coincided with the reception of the Caesar Constantius Chlorus in London in AD296 (Birley); at some point London was renamed Augusta and this may have followed the split. During the reign of the Tetrarchy Caesariensis was split into Flavia Caesariensis (FC, capital Lincoln) and Maxima Caesariensis (MC, capital London), while Britannia split into Britannia Prima (BP, capital Cirencester) and Britannia Secunda (BS, capital York). We can see this in operation at the Council of Arles, where there were bishops from London, York and Lincoln and clergy from Cirencester present; perhaps the bishop was too old or too ill to travel. It was normal for the church to have a bishop in every provincial capital and a metropolitan for every diocese, and London doubled as the seat of the vicarius, so perhaps the bishop of London was a metropolitan (Verona List).
Maxima Caesariensis was the only province of the four to be a consular province; such provinces were held by ex-consuls and were therefore a privileged set. It may be that its tax income was allocated to Maximian and that of FC to Flavius Constantius Chlorus, which is why it was clearly a lot smaller. Because there was a bishop at Lincoln, I would propose that the basic shape of Flavia Caesariensis was that of the medieval See of Lincoln. Since an Anglo-Saxon king could have multiple bishops, but a bishop could serve only one king, this large bishopric which runs from the Thames to the Humber to me represents the original Mercia of the Tribal Hidage and Flavia Caesariensis.
There is a mystery later in the fourth century, and that is Valentia. Its name suggests it cannot predate the reigns of Valentinian (imp. 363-375) and Valens (imp. 363-378). Ammianus Marcellinus says that the province was created after the suppression of the Barbarian Conspiracy (AD 367). We know that this is true because the province exists in the Notitia Dignitatum and (after the event) in the calendar of Polemius Silvius. It too is a consular province, indicating its importance. It is unlikely that a consular province would be a part removed from an ordinary province; more likely is that it would have been a part of Maxima Caesariensis. If MC is south-eastern Britain and FC is the area around Lincoln, I propose that the there is only rational place for Valentia.
I cannot see any reason why a new province would be established in the north west as a subdivision of Britannia Secunda. The provincial governor did not command troops, he only raised the funds to pay for them. A north-western province, roughly modern Cumbria, would have been quite poor and have no other function other than as a military base. Nor can I see an ex-consul traipsing up to Carlisle (Luguvallum) as a civilian leader in what Peter Heather would call the ‘middle of bloody nowhere’.
My argument is that Valentia was a reward, not an imposition. As such, I can only see it as a subdivision of Maxima Caesariensis. My best guess is that it comprised what are now Norfolk, Suffolk, Essex, Kent, Surrey and Middlesex, with a small area of what is now East Sussex. The coastal areas of this match the reach of the Count of the Saxon Shore. This was a special military command designed to ensure the orderly settlement of ‘Saxons’ in Britain following the events of AD367. I completely refute the idea that it was called the Saxon Shore (Litus Saxonum) because it was the shore from which the Saxons were forbidden, because they were generally forbidden to be on any shore!
The Governor of Valentia would have ruled the province from London, which was part of the tribal area of the Cantii; the four kings of Kent referred to by Caesar ruled were East Kent, West Kent, Surrey and Middlesex. The Saxon kingdom of Kent seems too to have held overlordship over Essex. It will have been noted that the governor of Maxima Caesariensis also worked in London, as did the vicarius who coordinated the whole island, plus, I would guess, officers who reported to the Praetorian Prefect of the Gauls. In addition, now that Britain had a metropolitan, an archbishop of London, he and his clergy would have worked there too.
London in the fourth century was rather different from the vigorous city of previous centuries. At some point it built a fourth wall along the river side, an area which had previously been open land lined with wharves, thus cutting the city off from its own riverfront and landing places, both military and civilian. This wall survived till Norman times, when it was demolished, trapping a coin of that time under its rubble. Late Roman London ought to have been a secure place, although it failed to stop Pictish ships from slipping past the watchtower at Shadwell and seizing the city.
At some point, as I have said above, London was renamed Augusta, the city of the Augustus. This was a singular honour, although it cannot have been taken as such too widely, because the name didn’t stick, unlike similar names in Gaul and Germany, for example (Autun was Augustodunum and Augsburg was Augusta Vindelicorum). I suggest that the reason for the change in name was first to assure the importance of the city to the Empire, and second because it was the base for the consular governors of two provinces, for the vicarius who reported directly to the emperor, for the bishop of London, for the staff of the Pretorian Prefect of the Gauls and probably the headquarters staff of the Count of the Saxon Shore. As there was a mint in London until after the reign of Magnus Maximus (d. 388), this may be associated with the enclosed site. The later location of the royal mint within the Tower of London may reflect an earlier industrial site in that area, as the Normans built the White Tower on top of the Roman citadel, clearing away the strata which had built up in the interim.
London’s soil was cleared of rubbish and relaid over older demolished buildings as ‘black earth’. This has sometimes been interpreted as rotted timber buildings, but if so, the residents of such buildings were so careful they never dropped a chicken bone or a small coin. I think that within the walls the normal inhabitants were moved out (to the extramural settlement later known as Lundunbyrig, perhaps) and an imperial-divine precinct created for the great and the good and their crews. They even built a very large church, probably a cathedral, dedicated it is believed to St Paul, which was discovered on Tower Hill in 1995. The church of Allhallows by the Tower dates from 675 and contains Roman material. It seems likely that St Martin in the Fields, now in Trafalgar Square was a Roman church, and if so its dedication must postdate St Martin of Tours (d. 8 November 397) and may predate AD410. Maybe that was for people displaced from Augusta. The dedication of St Pancras Old Church to an obscure Roman saint may suggest that, as legend insists, it was an old church just outside the walls for local residents, its its later boundaries were from Oxford Street to Highgate, suggesting that a local noble with a sizeable retinue patronised it.
It is notable that there were churches to St Pancras and St Martin at Canterbury when St Augustine arrived in AD597. Augustine brought relics of St Pancras with him to Britain, and since his task was to reopen the See of London (rather than starting a new one at Canterbury), he had presumably intended them for the extramural church near London.
In summary, I see Valentia as the fifth province usually cited, subdivided from Maxima Caesariensis to include the territories bounded by the Saxon Shore forts and run from an imperial precinct set within the walls of what had been London.
Haselegrove, C. (2008) ‘Central and Atlantic Britain’ in Todd, M. (ed.) A Companion to Roman Britain, London: Wiley.
Hind, J. (1975) ‘The British 'Provinces' of Valentia and Orcades (Tacitean Echoes in Ammianus Marcellinus and Claudian)’ Historia Bd. 24, H. 1 (1st Qtr., 1975), pp. 101-111.
Millett, M (1992) The Romanization of Britain: An Essay in Archaeological Interpretation, Cambridge: CUP.