Part 3

Monuments of Rome

10

Bread and circuses (c. ad 70–80)

The amphitheatre

The years following Vespasian’s accession witnessed a remarkable urban florescence (Table 10.1). This was set in motion by a series of building projects designed to deliver an architecture of ‘bread and circuses’, bringing to mind Juvenal’s famous critique of Rome’s dependency on such imperial handouts. ¹ London’s first great public building was its amphitheatre (Fig. 10.1). Archaeologists identified this structure when its eastern entrance was uncovered at Guildhall Yard in 1988, giving coherence to earlier discoveries in the area. ² The arena was sunk into a hollow formed by a tributary of the Walbrook that had been landscaped to form an elliptical gravel-floored stage. This was enclosed by a timber wall, possibly with a main entrance on its southern side. The east gate was about 3.4 metres wide, closed by a pair of doors with a small wicket gate to one side and a wooden tank fed by water-pipes inside the gateway that may have housed a fountain. An earthen bank enclosing the arena formed a cavea for tiers of timber seating rising to the line of an outer timber wall. It has been estimated that there was enough space to seat around 6900 spectators, little short of London’s likely adult male population of the time. The ambition of the project makes one wonder if it were not also designed for audiences swelled by troops in transit.

Fig. 10.1 The Flavian amphitheatre (chiefly GAG87: after Bateman et al. 2008). Drawn by Justin Russell.

Table 10.1 A suggested timeline of events affecting London in the period ad 70–125

The passage through the cavea was formed by timbers felled in ad 70 and ad 71, and a pile beneath the main threshold was felled ad 74. The building programme might, therefore, have stretched over several years. The distribution of dated timbers is consistent with groundworks starting c. ad 71, using wood procured the preceding winter, with the final fit-out delayed until after c. ad 74. Works may have been suspended for a couple of years between raising the cavea and installing the seating and fittings. Alternatively, the entire construction programme may have been compressed within the period c. ad 74–5, but drawing on timbers procured a few years earlier. Some timbers were marked by stamps, including ‘ICLV’ and ‘MIBL’, likely to have tracked an official procurement exercise. We don’t know whether these identified civic, procuratorial, or military suppliers, although the decision to build would have relied on the authority of the provincial government.

This early Flavian construction, perhaps Britain’s first amphitheatre, confirms London’s importance to Vespasian’s project. ³ The building was a symbol of the new regime, presenting an architecture typical of military and veteran communities. ⁴ Notwithstanding vast differences of scale and sophistication, London’s amphitheatre shared ideological and political purpose with Rome’s Colosseum: itself a gift from the emperor Vespasian to his capital city. Both were the product of the reconfiguration of urban performative ritual that followed Vespasian’s ascent to power. Festivals associated with the Imperial cult were placed at the centre of religious and civic life throughout the cities of the western empire, in which arena audiences were entertained by games (munera) involving gladiatorial combat. ⁵ This theme was popular in the decorative arts of Roman Britain showing that such performances were a significant part of provincial culture. ⁶ Roman games normally started with a ceremonial procession that collected the gods from temples and carried them proudly to the arena. ⁷ This made the amphitheatre the principal destination for religious pageants, and a setting for theatrical events. ⁸

The amphitheatre was also the chief venue for public execution, where Roman authority was affirmed in violent spectacle. Killings supervised by Roman soldiers could involve beheading, burning, crucifixion or being thrown to wild beasts. ⁹ London’s amphitheatre was therefore a direct manifestation of Vespasian’s assumption of control, marking an end to the chaos that attended Nero’s fall. This public building, the largest in London at the time of its construction, was placed on the town’s northern limit. Here it defined a ceremonial area where ritualized civic violence could be accommodated without polluting the urban templum, perhaps giving monumental shape to an earlier execution ground (above p. 110).

At the same time as the amphitheatre was planned, unusual building activity took place on the banks of a tributary of the Walbrook some 300 metres to its east. Excavations at Drapers’ Gardens found two palisade fences using timbers felled in the winter-spring of ad 70/71. The fences were more than 2 metres high, formed from upright radially cleft pales with spear-shaped (hastate) upper ends and fastened together by horizontal rails nailed into place. These were the product of military-style engineering, forming one or more irregular enclosures on the northern boundary of the settlement. ¹⁰ We have no direct evidence for their purpose, but it is tempting to suggest an association with the amphitheatre. One might imagine captives and wild beasts destined for the arena being held in pens and stockades here on the town borders.

An administrative complex in Southwark

Southwark also warranted early Flavian attention (Fig. 10.2). The channel between the north and south islands was canalized by post-and-plank revetments that reduced its width to about 15–20 metres, creating quays and moorings for rivercraft. ¹¹ A small jetty on the south side of this channel used timbers felled in ad 72, and the revetments that formed the north bank of the channel are dated as early Flavian by pottery within the land reclamations. The canal served a major building complex on its northern bank. Excavations by Carrie Cowan at 15–23 Southwark Street revealed parts of a massive early Flavian courtyard building, measuring at least 50 by 30 metres, laid out with unusual symmetry over three main wings. ¹² This was one of London’s earliest and largest masonry buildings, commanding the southern borders of the settlement. The principal rooms were set behind a 3.2-metre-wide ambulatory around an east-facing courtyard. The flint and ragstone foundations rested on closely packed timber piles that had been felled in ad 72 and ad 74. It replaced one of London’s more significant Claudio-Neronian buildings, itself on the site of a pre-Roman settlement, and the occupation here generated a concentration of objects associated with the army (pp. 63 and 78).

Fig. 10.2 The Southwark Channel and ‘mansio’ (15SK80: based in part on Cowan 1992, adapted to include more recent information on the revetments associated with the Flavian canalization of the watercourse). Drawn by Justin Russell.

The location, date and character of this complex mark it out as unusually important. Cowan suggests that it may have been a mansio, or praetoria, offering facilities for high-ranking imperial officials travelling through London on public business. These administrative compounds were also places for marshalling goods obtained through taxation and rent, typically laid out around large courtyards with ambulatories providing covered space for the loading and unloading of waggons. Richard Hingley has questioned this identification, pointing out that it might have been the house of a wealthy Londoner. ¹³ This is not beyond the bounds of possibility, but it would have been an exceptional residence, unparalleled in the domestic architecture of the period. The building’s layout, with its symmetrical projecting wings, finds closer parallel in an early second-century building at Chelmsford that is also thought to have been used as a mansio. ¹⁴ These establishments were distinguished from contemporary townhouses by their scale, symmetry, location, masonry architecture and courtyard layout. They were also provided with richly decorated interiors and bathhouses, all of which were eventually provided at 15–23 Southwark Street (p. 263).

Praetoria were found not only at roadside locations but at entrances to cities where large official entourages could avoid the congested city streets. ¹⁵ The borders of Southwark were an eminently sensible place for such a facility, meeting the needs of those arriving at London from the south, commanding river traffic and the approaches to London Bridge, and as base for the administration of the south bank settlement. Vespasian’s administration promoted centralization, transferring responsibilities for collecting indirect taxes from agents (publicani) to government officials, and is credited with a policy of improving communications. ¹⁶ A new transportation network was developed around London in the 70s, when new roads were built alongside the Thames and penetrating into the ironworking districts of the Sussex Weald (p. 174). This may have contributed to the development of an industrial area, possibly associated with ship-building, between the mansio and Thames (p. 212). ¹⁷

The canalization of the Southwark channel and building of the supposed mansio were contemporary with the construction of the amphitheatre. Materials for both projects were procured as early as the winter of ad 70/71, and work was still ongoing at both sites in ad 74 perhaps after interruption. There are also intriguing similarities in the later histories of the two complexes. Both witnessed several phases of alteration and improvement before becoming the sites of small late antique cemeteries (p. 382). These might hint at common institutional links between these buildings controlling the opposite borders of town.

Improvements to port and city

London’s Neronian port remained busy throughout the Flavian period. ¹⁸ Landing stages built onto the Thames foreshore immediately downstream of London Bridge added to handling capacity. These were investigated at Pudding Lane where an open framework structure with a front wall of horizontally stacked timber baulks extended 57 metres along the waterfront. ¹⁹ This supported timber decking reached by gangways from the quays behind, and was built with timbers procured after ad 69 but probably before ad 77. ²⁰ Simple post and plank revetments of this date stretched further downstream at Three Quays House and Sugar Quay, combining to suggest that quays now extended 620 metres along the riverfront. ²¹ Behind them, the glassworker’s shop (above p. 101) may have been converted into a mosaicist’s workshop. In the neighbouring unit three lead ingots stamped as property of the emperor Vespasian from British lead-silver mines, each weighing about seven kilos, were buried upside-down beneath the floor. ²² They are likely to have come from mines in the Mendips, obtained on behalf of the procurator and shipped to London for use in construction works or industrial production. We can only guess as to why they were never recovered from their place of safekeeping.

Several streets were added around London’s borders in early Flavian expansion (Fig. 10.3). These included roads built over the long-redundant ditches along the northern and western perimeters of the Claudian enclosure. ²³ If this land had been part of the defensive circuit it may have remained part of the emperor’s patrimonium, in which case the procurator would have been responsible for releasing it for development. The northern road carried eastwards beyond the former city limits to a junction with the road to Colchester. ²⁴ It then continued into the countryside, where ceramics found at St Clare Street suggest that road surfaces were laid between ad 70 and 80, heading towards a riverside establishment near Shadwell (p. 176).

Fig. 10.3 A proposed reconstruction of London c. ad 74. Drawn by Justin Russell.

Another east-west street was inserted mid-way between the amphitheatre and the ‘decumanus’, extending the line of a Neronian street on Cornhill by bridging the Walbrook near the Bank of England. Primary street surfaces along this route, at Old Jewry and Ironmonger Lane, were pottery dated to the late first century. ²⁵ This road joined onto an earlier, unmetalled, route which skirted the sacred area at Blossom’s Inn: the first metalled surfaces of this stretch of road are dated by the presence of a locally produced (VRW) mortarium with a counterstamp of Matugenus, dated c. ad 80–90, at 120–2 Cheapside. ²⁶ It then veered north-west into the countryside, becoming the 2-metre-wide cambered track recorded at 7–12 Aldersgate Street. ²⁷ This probably carried wheeled traffic past the farm at Clerkenwell towards a crossing of the Fleet, offering a short-cut onto the road to St Albans (Verulamium). Another important route out of town headed towards Westminster along the line of Fleet Street. This crossed the Fleet over a bridge built no later than the early Flavian period: facilities built alongside this crossing incorporated a plank cut from a tree felled ad 67–73. ²⁸

Later in the 70s the west bank of the Walbrook was reorganized following the introduction of two new north-south aligned streets meeting at the street junction at One Poultry. ²⁹ Drains built in contemporary engineering employed freshly carpentered planks felled in the winter of ad 77/78. These streets gave access to properties along the west bank of the Walbrook, where land was prepared for housing by massive terraces formed by timber box-revetments. This was a major landscaping exercise, dendrochronologically dated ad 73–81. The scale of these works finds parallel in the construction of the Neronian waterfront quays, suggesting a public venture drawing on advanced engineering capabilities. One of the buildings, possibly a two-storied granary, incorporated several timbers felled ad 78 and one felled ad 79. ³⁰ It appears that the main landscaping works along the banks of the Walbrook took place ad 78, and building plots were filled out with new structures the following year. Further engineering took place south of the Thames a year later. Excavations near where Watling Street crossed from the mainland onto the south island exposed a zone of timber posts felled in the winter of ad 79 or spring of ad 80. ³¹ The timbers were arranged in rows radiating back from the waterside and perhaps supported a raised riverside boardwalk.

The roads built in the 70s opened up new areas for property development. Income from these developments might have contributed to revenues managed by the procurator, possibly underwriting the costs of celebrating the munera. The commissioning of London’s amphitheatre entailed a commitment to forward expenditure, and it was common practice to secure resources for public festivals by consecrating landed property in a foundation bequest. ³² It is also possible that some properties rewarded veterans and others who supported the Flavian cause. Early Flavian investment also involved building roads to reach satellite sites within London’s hinterland, including riverside sites at Westminster and Shadwell, and the construction of quays at the base of the river Fleet and in Southwark.

Mills and bakeries

Food security must have been an acute concern during London’s uncertain infancy. The larger charred grain assemblages found in London consisted chiefly of cleaned (de-husked) spelt wheat brought into town in bulk after processing and cleaning. ³³ In the early city this grain was ground into flour in rotary hand-mills made of lava from the Eifel Hills in Germany, imported via the Rhine. These querns are rare on domestic sites and milling is likely to have been associated with bakeries. Exceptionally large numbers have been found in the Cheapside area, where bakeries and granaries stood alongside the ‘decumanus’ before the revolt of ad 60–1. ³⁴ More than 1,000 fragments of broken quern stones were used to form cobbled surfaces around a water-tank dated c. ad 75–80. This assemblage also included three fragments from animal-powered millstones, supplementing the evidence of an hour-glass donkey-mill found nearby in Princes Street. ³⁵

Grain would have been unloaded at quays along the lower Walbrook and granaries have tentatively been identified on the opposite side of the road to the quernstone pavement (Fig. 10.4). The first, built c. ad 78, was replaced by a structure with a raised wooden floor using timbers felled ad 79–92/4. Cleaned free-threshing wheat, with some spelt and a little barley, was found here in a ‘grain bin’ and large quantities of carbonized grain were found at the National Safe Deposit Company site in Bucklersbury. ³⁶ A bakery was tentatively identified at One Poultry, 50 metres west of the cobbled surfaces with discarded quernstones. ³⁷ This unusual Flavian structure housed several ovens, along with debris suggestive of food-processing. It occupied a prominent corner location formed by the amalgamation of several earlier properties. The core of the building was built of brick or masonry set over coursed flint and mortar foundations, and a narrow portico was set along both street frontages. It was located on the midway point of the route from the forum to the amphitheatre, making it a suitable point for selling or distributing food to processions on feast days. Other bakeries have been identified around London’s central forum. Three circular tile-lined ovens suitable for baking were found in a late first-century timber building opposite the south-east corner of the forum at Fenchurch Street, and two others in a contemporary building at Birchin Lane. ³⁸

Fig. 10.4 Flavian buildings in and around One Poultry following alterations of c. ad 78 (ONE94: after Hill and Rowsome 2011). A water-mill may also have stood on the north side of the bridge over the Walbrook. Drawn by Justin Russell.

In an important initiative a series of water-powered mills were introduced to London before the end of the first century. Andrew Wilson describes seven large millstones from London, including three from Bucklersbury House where the Walbrook channel appeared to narrow and a mill may have stood. ³⁹ Excavations at the Bloomberg site found two rare survivals of oak machinery that may have derived from this mill. One may have been part of the mill wheel, consisting of a paddle blade sculpted from a sawn oak plank. The other was part of the lantern gear used to drive the rotating shaft set through the millstones, represented by an oak disk pierced by six round holes for gear bars and a square axle hole for the drive shaft. Vertical mills require right-angled gearing to convert the water-wheel’s rotation into the horizontal rotation of a spindle to turn the lower millstone. Another complete millstone from a water-powered mill was found at Pinners’ Hall in Great Winchester Street suggesting that a second mill was located higher up the Walbrook valley. ⁴⁰ Stephen Myers’ doctoral dissertation shows that the Walbrook provided sufficient head to power watermills at both locations. ⁴¹ The river arrived at London’s northern boundary some 7.50 to 8.00 metres above OD, but reached the Thames between 0.50 and 1.50 metres OD: a drop of 6–7 metres. A typical late first-century overshot waterwheel had a diameter of 2 metres, requiring a fall from water offtake to riverbed of less than 3 metres. On this basis, the head of the Walbrook was sufficient to drive two overshot wheels in series.

A water-mill may also have stood where the Fleet flowed into the tidal Thames. Excavations between 1988 and 1992 identified two small eyots on the east bank of the Fleet, just north of Ludgate Circus (Fig. 10.5). ⁴² Reports prepared by Bill McCann describe a substantial jetty and timber-framed warehouse on the downstream eyot associated with pottery dated c. ad 100–120 and crop-processing waste from the parching and threshing of spelt wheat. Features believed to have been associated with a substantial tidal mill were found on the upstream, northern, eyot. Here a 2-metre-deep channel, perhaps a mill race, may have fed a mill pond. The channel also contained charred spelt wheat and chaff, showing that threshing took place nearby. Dendrochonological analysis shows that the waterside revetments used timbers felled ad 76–99.

Fig. 10.5 A proposed reconstruction of the late first-century topography at the mouth of the river Fleet (VAL88: after McCann 1993). Drawn by Justin Russell.

It is not clear if London’s water-mills were built in a single programme although this is distinctly possible. The mill near the Walbrook bridge was probably built c. ad 78, since this is when the banks of the Walbrook were re-engineered in ways consistent with the insertion of a mill leat. The arrival of this new technology helps to explain the mass discard of hand-mills c. ad 75/80. Whilst the watermills may have been operated by private contractors, their construction was integrated with other aspects of hydraulic engineering and relied on advanced engineering associated with military and public contracts. The introduction of water-powered mills would have radically transformed the feeding of the city. They were more than five times as productive as animal-powered mills but cheaper to operate, and Myers estimates that two mills on the Walbrook could have produced flour for up to 5000 civilian consumers. With those on the Fleet, London’s water-mills could have met a third of the needs of the entire city. Their construction would have eased the problems of keeping London’s population fed, supporting distributions of bread from adjacent bakeries. ⁴³

City limits and defences

A massive late first-century ditch found at Baltic House during the construction of the ‘Gherkin’ skyscraper was of a suitable scale and character to have formed the northern boundary of the enlarged city. ⁴⁴ This V-shaped ditch was originally about 6 metres wide and 2.45 metres deep, almost twice the size of contemporary city ditches, with a square-cut ‘ankle-breaker’ slot at the base. It was open long enough for weathering to occur, and contained pottery dated ad 70–100. No trace of a rampart survived, but upcast from the ditch allowed the construction of a substantial bank about 125 metres forward of the line taken by the Claudian defences. The ditch hasn’t been found anywhere else, largely for want of investigations in appropriate areas, making it difficult to know whether it was part of a town boundary or a more local feature. Guy Hunt has suggested that ditches at Cooper’s Row, on the eastern side of town, might also have defined a late first-century urban boundary. ⁴⁵ These were, however, smaller and of uncertain date. Despite the failure to identify other stretches of the ditch-system, its character and location is consistent with its interpretation as an early Flavian defensive earthwork.

Other towns in south-east Britain were provided with formal boundaries at this time. A V-shaped ditch of similar dimensions was dug to enclose parts of Verulamium, perhaps c. ad 75–80, before being backfilled by the middle of the second century. ⁴⁶ Late first- or early second-century ditches also drew boundaries to the town at Silchester (Calleva Atrebatum). ⁴⁷ In all three towns the earthworks appear to have been discontinuous, suggesting that their symbolic role was more important than their defensive one. The provision of defences implies an elevation in status, and it has been argued that the bank was built when London was granted a formal urban charter. ⁴⁸ It is also possible, however, that London was important enough to warrant urban defences because of its association with the provincial government regardless of its formal status.

London’s new Flavian streets also included an unusually important metalled road that cut a swathe through housing at 10 Gresham Street. ⁴⁹ The late Neronian enclosure here (Fig. 9.5) was cleared to make way for a 7.5 metres wide road flanked by substantial drains. This was significantly wider than normal for contemporary city streets and later subsidence shows that the aggregate for its metalled surface was quarried from adjacent gravel pits without regard to the needs of subsequent house-builders. Since a perfectly viable Claudio-Neronian street lay only 25 metres to the west the new road was probably inserted to take traffic towards an important destination rather than service local houses. ⁵⁰ This important destination was later the site of the second-century Cripplegate fort (Fig. 19.2). Although excavations have yet to find certain trace of an earlier fort, the site was comprehensively cleared of vegetation before dumps of imported brickearth containing Flavian pottery established a level building platform over an area of at least 18,000 square metres. These works took place long before the Hadrianic fort was built. This area was also bounded by a V-shaped perimeter ditch, up to 2.4 metres wide and 1.6 metres deep with an ‘ankle breaker’ that was found at Barrington House where it may have enclosed an early Flavian encampment preceding the Hadrianic fort. ⁵¹ Pottery suggests that it was backfilled before the end of the first century. It has not been possible to identify features associated with a fort interior, but a similar absence of evidence characterized the Neronian fort in the south-east quarter. ⁵² The presence of a temporary camp to the north-west of the settlement might explain why a principal road was built to give access to an extensively landscaped site bounded by a ditch of a type associated with military engineering. The large V-shaped ditch found at St Bartholomew’s Hospital, discussed in Chapter 7, might have defined a similar enclosure. We can surmise that the area of higher ground north-west of the city housed passing military units at various points in London’s early Roman history. ⁵³ Whether or not Vespasian’s forces used temporary camps on the north-west outskirts of town, they had continuing access to the Neronian fort in its south-east corner. This was an increasingly neglected feature, but wasn’t levelled until c. ad 85. ⁵⁴ The failure to convert the redundant fort into housing in the mid-70s may have been no more than institutional inertia, but it is more likely that the site was kept vacant for the sporadic needs of troops and goods transiting London. These forces could have brought engineering skills and physical labour to aid the construction of the amphitheatre, mansio, river channels, quays, terraces, water-mills, new roads, and town defences that were built in the busy 70s.

Vespasian and London

These early Flavian changes have implications for our understanding of imperial policy in Britain. Vespasian was proclaimed emperor in the summer of ad 69 but didn’t arrive in Rome until the autumn of ad 70. According to Josephus this was the year that he appointed Petillius Cerialis, probably his son-in-law, to govern Britain. ⁵⁵ Cerialis was dealing with the aftermath of the Batavian revolt in Germany when appointed, and this probably delayed his departure to Britain until the spring of ad 71. This coincides with the date that work started on assembling materials for the construction of London’s amphitheatre and the supposed mansio/principia in Southwark, and perhaps also the renewal of water-lifting devices at Blossom’s Inn. ⁵⁶

Britain was important to the new regime. Both the emperor and his legate had previously served here, the former at the time of the Claudian conquest and the latter during the Boudican revolt. ⁵⁷ Vespasian was the first Roman emperor to rule without the benefit of dynastic legitimacy, and success in Britain offered a way of grafting Flavian achievement onto earlier Julio-Claudian victory. ⁵⁸ Britain was the place where the divine Caesar and Claudius had consolidated their authority as guardians of Roman destiny, and where Vespasian gained early military glory. These factors help to explain why London, the symbol of Roman power in Britain, was the precocious beneficiary of Flavian patronage. It is difficult to see how resources for the changes described here could have emerged from private or civic munificence within London’s diminished late Neronian community. The scale of the engineering and political character of the monuments suggest the involvement of the administration. The likely patrons were senior officials, acting to consolidate Flavian political authority.

Many scholars are reluctant to credit the Roman army with works undertaken on behalf of Britain’s civilian communities, offering corrective balance to earlier studies that were unduly dismissive of civilian capabilities. The army was, however, a versatile tool in the implementation of imperial policy, and easily deployed onto construction projects when not on active campaign. ⁵⁹ The particular circumstances of London’s Flavian reconstruction might have warranted such help. A parallel can be found in the actions of Vitellius, Vespasian’s principal rival for power in the year of the three emperors, who sent a legion to help build amphitheatres at Cremona and Bononia. ⁶⁰ This precedent may have encouraged the Flavian commitment to provide munera, as Vespasian adopted this aspect of Vitellius’ political strategy. This is not to say that other agencies were not involved in the construction works undertaken in London, where the administration could enlist the support of a range of public and private contractors.

Although elements of London’s urban renewal may have been planned on Cerialis’ appointment in ad 70, dendrochronology describes subsequent peaks of activity c. ad 74 and c. ad 77/78. Political and military considerations may have contributed to a cyclical pattern of investment in London’s infrastructure at these dates. ⁶¹ Flavian preparations aimed at the completion of the conquest of Britain involved exaggerated movements of labour and capital. Julius Frontinus replaced Cerialis late in ad 73, a few months before supplies of timber were obtained to complete the fit-out of the amphitheatre and supposed mansio. ⁶² Similarly, Frontinus was succeeded by Agricola in the summer of ad 77, which arrival was promptly followed by the procurement of timbers for the landscaping of the lower Walbrook valley and related urban improvements. ⁶³ These works included the construction of water-powered mills and new arrangements for feeding the city. Comparable, if not entirely synchronous, investment cycles can be reconstructed from the evidence of major injections of fresh supplies of imported coin in the periods ad 64–7; ad 71–3; ad 77–8 and ad 86–7. ⁶⁴ There is the hint of a correlation between the arrival of new governors commanded with expansionary goals, exercises of procuring timber for major construction projects in London, and injections of new coin. Political impetus for the improvement of London may have stemmed from the arrival of new provincial governors, each ambitious to leave a stamp on the leading provincial city.