The Bull Ring Uncovered: Excavations at Edgbaston Street, Moor Street, Park Street and The Row, Birmingham City Centre, 1997-2001

By Catharine Patrick and Stephanie Ratkai

The excavations in the centre of Birmingham uncovered evidence of habitation from prehistoric and Roman times, but the 12th to 19th centuries presented by far the most evidence, from artefacts, environmental samples and structural remains. The medieval industrial past was of particular interest, with tanning and the manufacture of hemp and linen all playing a large role in the city's prosperity. Metal working reached its peak in the seventeenth century, with brass founding becoming important from the eighteenth century onwards. Most of the artefactual evidence attests to Birmingham's industrial past, indeed the evidence for domestic life is comparatively scant, with an anomalous burial of two people at Park Street presenting something of a mystery. This volume presents insights into the early industrial past of this important city and is an invaluable record covering eight hundred years of occupation.

• Language: English
• Publisher: Oxbow Books
• Release date: Dec 12, 2008
• ISBN: 9781782978725

Book preview

1 Introduction

Catharine Patrick

---

Before the mid-1990s, archaeology in Birmingham did not have a high profile. Historical research (Holt 1985) had suggested that Birmingham was an important example of an industrially based medieval town but there was little in the way of excavated evidence to support this. Archaeology can provide an important and contrasting source to the traditional documentary record, even for the 18th and 19th centuries – both in terms of the evidence provided by standing buildings and in terms of the buried evidence of the lives of the urban residents. This evidence becomes progressively more important the further back in time one goes, particularly in Birmingham, where relatively few medieval documents have survived for the town.

The historic memory of the Bull Ring is not represented by buildings – these were replaced by the Bull Ring development of the ’60s and ’70s – nor is it represented by documents. The historic memory is instead represented by buried remains – yards, pits, wells, postholes, ovens, ditches and gullies – all of which can yield dating evidence and clues to the profession and status of the town’s inhabitants, and provide a glimpse of what it must have been like to live in the Bull Ring in the medieval period.

This publication focuses on archaeological sites at Edgbaston Street, Moor Street, Park Street and The Row, which are all located within the historic Bull Ring market area of Birmingham, close to the focal point of St Martin’s Church (Figs 1.1–1.4).

Geology and topography

This part of Birmingham developed on a prominent sandstone ridge about 1.2km wide. The ridge is part of the Birmingham Plateau, a geographical zone consisting of mainly Triassic rocks covered with glacial clays and gravels. The Bull Ring lies on this ridge (between 110m– 120m AOD) to the east of the conjectured Birmingham fault, overlooking a steep slope leading down via Digbeth to the lower lying and wet, marl-lined valley of the River Rea (VCH Warwickshire VII 1964).

The geology and natural topography of Birmingham are important for an understanding of the early development of the town. A good supply of water is essential to a medieval market town – it is needed for human consumption, for the watering of animals, and to drive mills, and is an essential ingredient, often in quantity, for many early crafts and industries. For early Birmingham, an important source of water was provided by springs that rose at the junction of the water-bearing Bromsgrove Sandstone Formation (formerly known as Keuper Sandstone) of the Birmingham ridge with the impervious Mercia Mudstone along the line of the Birmingham fault. These springs were the source of a number of rivulets that flowed generally southeastwards down the slope from the ridge and into the River Rea. Suitably modified and canalised, these springs and watercourses supplied the two medieval moats, Parsonage Moat and the Lord of the Manor’s (or Birmingham) Moat, which were amongst the earliest features of Birmingham’s topography. The watercourses also strongly influenced the layout of the early town. For example, the watercourse that connected Parsonage Moat with the Lord of the Manor’s Moat formed the southern boundary of the built-up area until the early 19th century (see Fig. 1.2).

Another major topographic influence on the development of the town was the rural road system which existed before the town developed. On the high ground where St Martin’s Church and the Bull Ring now stand, and where a market place was established in the 12th century, a number of local and long distance roads converged and then crossed the Rea floodplain via a single corridor, the Digbeth–Deritend–Bordesley route.

Together, the natural topography, watercourses and rural road network were important factors in shaping the development of the town. The town grew through the successive development of land parcels along the old roads and the insertion of new streets, generally roughly at right angles to the old roads, which created new land parcels for development. This street pattern then became fossilised and survived in recognisable form into the 19th century. Only with the coming of the railway, in the 19th century, and then the extraordinary road schemes and redevelopment of the 20th century, did this pattern become substantially obscured (see Figs 1.2–1.4).

Fig. 1.1 Location of Birmingham.

Moor Street and Park Street – both the focus of excavations described in this report – represent ‘insertions’ into the pre-existing road network. Edgbaston Street – the third major site described in the report – is likely to have earlier origins.

In Birmingham, as elsewhere, the typical medieval urban property plot, known as a burgage (‘town person’s’) plot, is long and thin, laid out at right angles to the street frontage. The rationale for this is simple: space on the street frontage was at a premium. The main building, perhaps a residence and/or shop, would be situated on the frontage, while the strip of land behind, the backplot, could be used for a variety of purposes – market gardening, keeping animals, industrial activities of various kinds, and waste disposal. Much of the archaeological evidence uncovered in the Bull Ring excavations relates to this sort of ‘backplot’ activity.

While the front of the plot would be defined by the road, the back end of the plot might be defined by a ditch or watercourse. Plots with a ready source of water to hand were particularly valuable. It has been noted that ‘the provision of watered plots is a recurrant feature of nascent urban settlements and, in particular, early markets’ (Baker 1995). Livestock could be grazed and watered close to the market – there would be a particular demand for such plots from the town’s butchers – while related industries such as the tanning of hides would demand a good supply of water.

Fig. 1.2 Location map showing excavated areas overlaid on 18th century map (Bradford 1750).

This pattern certainly seems to be true of early Birmingham. The plots fronting onto Edgbaston Street terminated in the watercourse connecting Parsonage Moat and the Lord of the Manor’s Moat; extensive evidence of tanning was uncovered in the archaeological excavations there. At Moor Street and Park Street, a large ditch marked the far end of the backplots fronting onto the market place and the upper end of Digbeth respectively. The ditch not only defined the original town boundary but also the edge of a deer park – Little Park or Over Park – that lay to the north and east of the early town. From the environmental evidence, a watercourse may have run along this ditch or it may simply have collected pools of standing water. Likewise, the environmental remains from a second ditch at Park Street, running at right angles to the above and defining the back end of the burgage plots running down from Park Street, indicate that it was also waterlogged. However, at Moor Street and Park Street the ditches were backfilled in the 13th century, which would seem to indicate that whatever the quantity of water they may have supplied to the burgage plots it was not considered of sufficient importance to keep them open. Park Street does seem to have had a number of water-filled ponds and tanks, which presumably provided some or all of the necessary water requirements. This is in stark contrast to Edgbaston Street, where the water-course was kept open until the late 18th or early 19th centuries. However, both the Moor Street and Park Street ditches produced plant and insect remains that provide important insights into the environment and the uses to which the ditches and adjacent land were put.

Fig. 1.3 Location map showing excavated areas in relation to the 1960s Bullring.

Background to the excavations

Edgbaston Street, Moor Street, Park Street and The Row represent part of the historic centre of Birmingham around St Martin’s Church. Research (Holt 1985; Baker 1995) suggested that evidence for the growth of Birmingham from the Middle Ages onwards was likely to be found in the immediate vicinity of the Bull Ring. The Lord of the Manor’s Moat, the smaller Parsonage Moat and their associated watercourses lie close by, and Edgbaston Street was one of the earliest streets to be laid out in the town. The notable persistence of property boundaries within the Bull Ring area suggested a high potential for the survival of archaeological deposits. It was predicted that there would be intense structural activity along the street frontages and a build-up of occupation deposits and rubbish pits within the yards and backplots to the rear. Excavation bore these suggestions out and extensive, well-preserved medieval deposits were located, generally at a depth of 1.60–3m below the present-day ground level.

The position of Moor Street and Park Street – close to the medieval and post-medieval market place – meant that archaeological deposits found there would be likely to reflect the area’s trading status. It was hoped that the archaeological evidence would help to date more precisely the insertion of the two streets into the town plan, and characterise the complexity and type of later developments.

The deposits and features identified by the archaeological investigations on all of the sites have revealed a sequence of development probably commencing soon after the granting of a market charter in the 12th century. In addition, information was gained pertaining to the economic activities which were crucial to the development of this part of Birmingham, beginning in the medieval period and continuing through the important transitional phases of the early post-medieval and later periods. The significance of industry to Birmingham in the medieval and early post-medieval periods cannot be emphasised too strongly.

The below-ground archaeology of the city centre has furthered our understanding of the chronology and form of Birmingham’s growth and has provided evidence to help to resolve vitally important questions concerning Birmingham’s early development. It has shed light on the historical development of this area from the Middle Ages up to the present day. The value of the archaeological resource in this area should not be underestimated, as Holt notes: ‘archaeology alone has the potential to offer a truly comprehensive early history of this part of Birmingham’ (Holt 1995).

Fig. 1.4 Location map showing excavated areas overlaid on modern Bullring.

Historical background

In 1086, at the time of the Domesday Book, Birmingham was only one of several small agricultural settlements within the area of the present-day city. Over the ensuing centuries this settlement evolved into a thriving trading, manufacturing and industrial town. The principal early stimulus to this development occurred in 1166 when the lord of the manor purchased a charter from the Crown allowing him to hold a weekly market and charge tolls. The charter probably legalised a pre-existing market and, as elsewhere, was no doubt obtained in an effort by the lord to create a new town around the market and enhance the value of his property through the generation of rents. Plots of building land, exclusion from tolls and privileged access to the market were offered to those who settled in the town (Holt 1985), and in the case of Birmingham this proved to be a very successful venture, signalling the rapid urbanisation of the settlement. By 1300, Birmingham and the wider region of Warwickshire had experienced a period of massive population growth and Birmingham had grown to become one of a network of market centres manufacturing and distributing goods. Lay Subsidies dating to the 1320s and 1330s show that Birmingham was the third largest town in Warwickshire after Warwick and Coventry (Holt 1985). However, Birmingham’s significance lay not in the presence of a castle (like Tamworth), nor as an administrative or religious centre (like Coventry, Shrewsbury or Stafford), but instead in its development as a trading and industrial centre for its evolving hinterland. The location of the triangular market place to the north of the manor house was probably part of this deliberate enhancement of Birmingham’s trading facilities.

The origins of Birmingham’s two moats – the Lord of the Manor’s Moat and Parsonage Moat – and their original relationship to each other are not clear but they are likely to have been important foci of rural development, and it has been suggested that they originally represented the manorial site and its ‘home farm’ (Baker 1995).

The 1166 charter refers to the market being held in the castrum and this may be the earliest surviving reference to the Lord of the Manor’s Moat. Stonework recovered during excavation of the moat (Watts 1980) has been dated to the 12th century, with parallels at Sandwell Priory, Wenlock Priory and Buildwas Abbey (pers. comm. Richard Morris). The market may have had a rival in Deritend c.1200 which, albeit temporarily, tried to capture some of Birmingham’s trade (Holt 1995).

It is likely that St Martin’s Church was built by the mid-12th century under the sponsorship of the lord of the manor. Architectural fragments dating to the 12th century have been found in the church, within a later rebuild. Descriptions, but no drawings, of these fragments survive. The economic success of the town allowed the church to be rebuilt on a larger scale around the middle of the 13th century (Holt 1985).

A survey of the lordship of Birmingham made in 1553 (Bickley and Hill 1891) details all the tenancies in the town, but it seldom gives the size of a property or its position within the town. The survey mentions a deer park and Tanners Row, which has been taken to mean Digbeth although the description would also be equally applicable to Edgbaston Street and Holme Park.

Nigel Baker (1995) has produced a preliminary town plan analysis of the development of Birmingham, mainly based upon interpretation of the Bradford map of 1750/51 (surveyed 1750, published the following year) and subsequent Ordnance Survey maps. His analysis has suggested that the principal features of the Birmingham town plan, which includes streets, street frontage lines, sites of public buildings, and property boundaries surveyed in the early 18th century, have probably not changed significantly since the area was first laid out for settlement following the granting of the market charter.

As described above, the growth of Birmingham was characterised by the successive development of land parcels along the major pre-urban roads, and the insertion of new roads laid out across the interstices of the existing network. Baker suggests that Moor Street and Park Street were cut through an already built-up frontage on the northeast side of the Bull Ring and that their purpose would have been to extend settlement behind that axial route (Baker 1995). The regular shape of the land parcel which these two streets define suggests that they were probably inserted as part of a single phase of town planning. The consequent creation of fresh building land may be seen within the context of an increased demand for house-plots within the rapidly expanding market town. Both Holt and Baker conclude that much of the central area of Birmingham, including the Moor Street, Park Street and Bull Ring area, was laid out sometime before c.1400. Given the limitations of the documentary and cartographic record, archaeological evidence is the only source for defining more precisely the early development of the town.

The founding of a Priory or Hospital at the northern limit of the medieval town was in keeping with contemporary urban development (Cullum 1993). As for the town itself, medieval documents for the Priory are scarce, although one document does record that land was given to the Priory in 1286. The Priory precinct was located at the northern extent of Dale End, which runs roughly parallel to Moor Street and Park Street. This is a street which represented a continuation of the main north-south axial route through Digbeth, the Bull Ring and High Street, and which fed north from the triangular market place. The later insertion of an arterial route from Dale End and High Street to the northwest (Bull Street) is likely to reflect the capacity of a priory to act as a stimulus for further planned medieval urban development (Palliser 1993).

Previous archaeological work

Prior to the 1970s, very little archaeological work had been done in Birmingham city centre. Construction work for the fish market to the north of the Bull Ring in the 1880s revealed a stone passageway and chamber, while a series of objects reputedly originating from the Lord of the Manor’s Moat, including a medieval ring, was already published (Oswald 1951).

Some examination of surviving below-ground archaeological deposits at the manorial moat site was carried out between October 1973 and June 1975, during development of the present-day Wholesale Market (Watts 1980). However, this development preceded the introduction of measures which have enabled large-scale archaeological excavation to take place as a condition of development (the Department of the Environment’s 1990 Planning Policy Guidance note 16). As a consequence, although much relevant information was recovered, the work resembled more a present-day ‘watching brief’ during construction than an open-area excavation. In this instance, Lorna Watts and a small team, working on behalf of the City Museum, monitored construction groundworks and recorded the survival of ‘islands’ of archaeology within the concrete pile foundation pits. Despite difficult conditions, this watching brief clearly demonstrated the survival not just of the ditch of the manorial moat, which was waterlogged and contained preserved wooden stakes, but also of substantial dressed sandstone footings and walls belonging to 13th-century structures on the moat platform. The walls survived to over 2m in height with a buttress on the outside of the south wall. No evidence for the date of the moat’s construction was found as most of the fills were post-medieval. Watts also noted that the 19th-century construction of the market had scoured away important occupation deposits from the moat platform itself.

Archaeological investigations associated with the Bull Ring redevelopment

Edgbaston Street was the first site to be excavated in advance of the Bull Ring development, with various episodes of excavation between 1997 and 1999. These excavations demonstrated the extensive survival of medieval and early post-medieval settlement and industrial features (dating from the 12th century through to the 20th century). The Moor Street site was the second major site to be excavated as part of the development, in 2000.

Smaller scale work took place elsewhere as the Bull Ring development progressed. In 1999 trial trenching immediately to the east of The Row Market found pockets of medieval survival between later 19th-century cellaring (Hovey 1999); this was followed up by a watching brief (Ramsey 2000). A watching brief (Patrick 2000 and this volume) was also carried out during the construction of a new street, The Row, which runs approximately east to west from Moat Lane to Upper Dean Street. This new street cuts across the northwestern side of the medieval manorial moat.

Trial trenching work was carried out at Manzoni Gardens and in the Open Markets in 2000 (Burrows and Mould 2000a; Burrows and Mould 2000b). Here all the archaeological remains were shown to have been scoured away by development, the only surviving early feature being a well cut into the sandstone ridge and located close to the present-day boundary of St Martin’s Churchyard.

A desk-based assessment in 1997 for the Martineau Galleries development (Litherland and Mould 1997) suggested the potential for survival of medieval deposits to the north of the Bull Ring. However, subsequent archaeological monitoring of geotechnical work and service pits in 1998 and 2000 found no surviving archaeology, but did record the natural sandstone ridge less than 1m below the present ground surface in Dale End (Mould 2001c).

In 2001, evaluation by means of trial trenching at Park Street was more productive, demonstrating extensive survival of medieval and later deposits. The evaluation was followed by area excavations in the same year and Park Street forms the third of the major sites described in this volume. At the same time as the excavations at Park Street, major excavations were also carried out near by in the churchyard of St Martin’s, in advance of landscaping works. The results of these latter excavations, which provided fascinating insights into the lives and deaths of the parishioners of St Martin’s in the 18th and 19th centuries (Brickley et al. 2006), form a companion volume to this one.

Collectively, these investigations make the most significant contribution to our understanding of Birmingham’s historic development since the recording of the Lord of the Manor’s Moat by Lorna Watts in the early 1970s.

Excavation and research aims

All of the major excavated sites offered an opportunity to study a more-or-less unbroken sequence of activity from the 12th century through to the 20th century.

The aims of the evaluation and excavation work, determined at the outset of the investigations, were to:

To a very considerable extent, these aims have been achieved.

Excavation method

The excavation method employed for all three major sites – Edgbaston Street, Moor Street and Park Street – was broadly the same. The modern overburden and concrete surfaces were removed by a 360 degree excavator, with a toothed and toothless bucket, under archaeological supervision. Spoil was stockpiled on site. The uppermost horizon of archaeological features and deposits revealed by machining was hand-cleaned and a base plan of features was prepared.

Machine excavation was carried out in two stages. The first stage was to remove modern overburden down to a distinct horizon noted at all three sites, the so-called Phase 3 ‘cultivation soil’, a dark layer of silty-clay sand dating to the 18th century, which generally sealed the earlier features. Any features cutting this horizon were excavated and the soil layer was itself sampled before being removed. The second stage was to mechanically remove the ‘cultivation soil’ to the top of surviving earlier post-medieval (Phase 2) and medieval (Phase 1) deposits and features.

Sampling by hand excavation comprised not less than 50% of discrete features. A higher percentage of discrete features was excavated where more information was required to achieve a full understanding of the date, character and function of an individual feature or group of features. Features of probable industrial function were fully excavated, whilst linear features not associated with settlement were sampled to determine their form, function and date, and to determine the stratigraphic sequence. Excavation of linear features associated with settlement comprised a minimum of 25%. All datable features were sampled for environmental analysis, with 20-litre samples being taken from non-waterlogged deposits and 30-litre samples from waterlogged deposits. Deeply stratified deposits or large features were multi-sampled.

Recording was by means of pre-printed pro-formas for contexts and features, supplemented by plans (at 1:20 and 1:50), sections (at 1:10 and 1:20), monochrome print and colour slide photography. Subject to the permission of the landowner, it is intended to deposit the paper and finds archive in an archive store approved by the Planning Archaeologist for Birmingham City Council.

At Edgbaston Street, Area B, the movement of skins and hides from one tanning pit to another over a period of almost 500 years had caused deposits to spill over from one feature to the next, making excavation in plan practically impossible. These features were instead recorded and sampled by a series of excavated sections, which is a departure from normal procedure. A similar approach was adopted for the series of pits and ditches at the Park Street site.

Phasing

To facilitate comparison between the sites, which are in fact elements of the same urban landscape, a common system of phasing has been adopted for Edgbaston Street and Moor Street (see below). Park Street is broadly the same but has subdivisions of Phase 3. Five phases of activity have been distinguished, mainly on the basis of the date of the pottery.

2 Land to the South of Edgbaston Street: Investigations 1997–1999

Catharine Patrick and Stephanie Rátkai

with a contribution from Steve Litherland

---

Background to the excavations

Today, the location of the Edgbaston Street excavations is underneath the new Indoor Market to the south of Edgbaston Street (Fig 1.4). Before redevelopment began here, land use comprised a multi-storey office block, two warehouses and a restaurant, which were demolished, and the rest of the street block was made up of rough car parking for the markets (Fig 1.3). The street block was still bisected by a narrow lane called Smithfield Passage that followed the line of a former medieval watercourse, and from here the ground level rose gradually towards the Edgbaston Street frontage.

A desk-based assessment of the site was carried out in 1995 (Mould and Litherland 1995b), and in 1997 two areas became available for evaluation by means of trial-trenching. Trench 1 was located at the western end of Edgbaston Street (Fig. 2.1). It extended to the south from the street frontage in order to test for surviving deposits relating to the former Parsonage Moat. However, several cellars had truncated any earlier remains and subsequent comparison and re-scaling of various historic maps with modern Ordnance Survey editions indicated that the moat actually lay a little further to the west, under what is now Pershore Street.

The second trench excavated at this time, Trench 5 in the overall scheme for site evaluation, was located immediately to the south of Smithfield Passage. Its purpose was to test for deposits relating to the watercourse that originally linked Parsonage Moat with the Lord of the Manor’s Moat. This trench demonstrated that the profile, primary fills and later backfill of the watercourse did survive. As a consequence, more extensive excavation to the south of Smithfield Passage followed on directly, with the aim of recovering further evidence of the watercourse. This involved the excavation of four further transects (Transects A–D) across the line of the watercourse (see Fig 2.1).

Four larger areas were opened up for excavation; Areas A–C lying to the north of Smithfield Passage and Area D to the south (Fig. 2.1).

Historical profile

by Steve Litherland and Catharine Patrick

As outlined in the introduction, Peter de Birmingham was actively promoting the status of his lordship in the 12th century, principally by gaining a market charter, and probably also by sponsoring St Martin’s Church. Edgbaston Street formed a primary component of his town planning, forming the lower end of a triangle of streets laid out around the market and St Martin’s Church. While it is possible that Edgbaston Street was a remnant of an earlier road system, after the laying out of the market place this thoroughfare connected the market, church and manor house with a network of local and regional routes serving Edgbaston, Dudley and Worcester to the west.

Edgbaston Street would have carried mainly local traffic from the southwest to and from the main axial route represented by High Street and Digbeth and it would have been one of the earliest streets to be developed in the town. Its limits were defined by Parsonage Moat to the west and the Lord of the Manor’s Moat to the east.

Edgbaston Street broadly follows part of the c.115m contour of the sandstone ridge that overlooks the Rea valley to the south. It is situated roughly two-thirds of the way up this hill and is also located just above the conjectural line of the Birmingham fault. Around the fault, several springs, issuing as water running off the hill, could not pass from the permeable sandstone into the impermeable Mercia Mudstone. These springs were harnessed to feed the Parsonage Moat situated at the west end of Edgbaston Street and also the Manorial Moat (i.e. the Lord of the Manor’s Moat) towards the east end (Fig. 2.2). The two moats were also directly linked by a watercourse, which by the later 18th century was linked to two other water channels, the Pudding Brook and the Dirty Brook (the latter basically an open sewer).

Fig. 2.1 Edgbaston St: location of trenches and excavated areas.

The watercourse linking the two moats formed the back boundary of an irregular series of properties running down the slope from Edgbaston Street, and it was the westernmost portion of these plots, situated between the former line of Gloucester Street to the east and the surviving line of Pershore Street to the west, that was targeted for excavation. The area immediately to the east of Gloucester Street lay under the former Rag Market building, where extensive cellaring had destroyed the archaeological deposits.

The provision of watered plots next to a market would have been particularly attractive to tradesmen such as butchers as livestock could be grazed and watered close to the market. The watercourse might also attract industrial enterprise, such as tanning. So the Edgbaston properties were likely to have been much sought after for their trading and market frontages and ready access to fresh water behind, and doubtless they continued to be so long after the de Birmingham family lost control of the manor in the 16th century.

There is evidence to suggest that the Black Death severely affected Edgbaston Street (McKenna 2005, 14). As a result, new tenancies were created by Fulke de Birmingham and there followed a rapid recovery, with properties given over to tanners, skinners, graziers, butchers, weavers, and flax and yarn dressers. The importance of a plentiful and easily available source of water to most of these trades is clear.

The 1500s were a difficult time for the administration of the town giving rise to three surveys, the most comprehensive of which, that of 1553, was only ‘rediscovered’ in the 1880s and formed the basis for a conjectural map compiled by William Bickley, who translated the document, and Joseph Hill, who wrote copious background notes to accompany the publication of the survey (Bickley and Hill 1891). While the precise location of some of the holdings outlined on the map has been questioned and further corroborative evidence is often desirable, nevertheless certain general points may be inferred from the 1553 survey with reasonable confidence.

By 1553, Edgbaston Street was certainly one of the more heavily developed streets in Birmingham, along with High Street, Corn Cheaping, Well Street (Upper Digbeth), and the Shambles in the market place. Most of the property in Edgbaston Street was held as free burgages, and some important families were represented who also owned other property around the town. These included the Holt (Holte) family, later of Aston Hall, who also held the Malt Mill, William Phillips (Phillippes, an ancestor of the Inges according to Hill), William Booth (Bothe, a knight), and Richard Wythal (who was probably a tanner).

As late as the early 19th century the limits of the town here were probably much the same as in medieval times. The Manorial Moat, Parsonage Moat and their associated watercourses still formed the southern boundary of the town. Throughout most of the 18th century the view from here remained overwhelmingly rural. Nearest the town were gardens and small enclosures, some of which were used as drying grounds by hide curers (note the ‘gibbetlike’ posts depicted on Westley’s view of 1732, see Fig. 2.2), and others contained osier pits where willow was soaked to make it supple for basket weaving. Beyond these were larger fields and enclosures that were once part of the Holme Park, which stretched from the Manor House down to water meadows by the banks of the River Rea.

By the time of Ackerman’s panoramic view of 1847 the block from Edgbaston Street to Bromsgrove Street had been fully developed (see Fig. 16.8). The oblique line of Smithfield Passage, which follows the line of the old watercourse, is still clear, although fronted on both sides by buildings.

Development up to the 18th century was hindered partly by the restrictive policies of the landowners involved. Dr Sherlock, Bishop of London (1678–1761), even went so far as to debar his successors from granting building leases by the terms of his last will and testament. This was because he thought that ‘his land was valuable, and if built upon, his successor at the extirpation of the term would only have the rubbish to carry off’. It took an Act of Parliament sponsored by one of Sherlock’s successors, Sir Thomas Gooch, to overcome this bar on development in 1766. However, a greater hindrance was probably the rather waterlogged nature of this area and the availability and development of more desirable land elsewhere in the town centre (pers. comm. Toni Demidowicz).

In 1731 the only development to the southwest of the Lord of the Manor’s Moat was a group of what appear to be farm buildings. However, this began to change once Sir Thomas Gooch, a major landowner, whose estates extended south from Edgbaston Street into the area which had been medieval parkland, was granted an Act of Parliament in which he was given permission to cut streets from his estate and lease out parcels of land. The fairly immediate outcome of this was the construction of Jamaica Row and Moat Row to the southeast and southwest of the Lord of the Manor’s Moat. This development was of mixed use comprising domestic retail and industrial buildings and is shown on Hanson’s map of 1778 (Fig. 2.3). However, development to the south of the Edgbaston Street backplots, beyond the watercourse, was minimal other than the division of land into garden plots. Clearly, as Birmingham expanded and burgeoned, there was a ready market for fresh garden produce in her markets. The situation still more or less pertained in the early 19th century, since Sherriff’s map of 1808 (Fig. 2.4) shows the area to be still predominantly gardens and osier beds. By the mid- to late 1820s this area was becoming built up.

Fig. 2.2 Detail of Westley’s Map 1731.

During the 18th century several trades are listed in the Birmingham directories for Edgbaston Street. Sketchley’s 1770 Directory indicates several butchers, a skinner, a fellmonger and leather workers, button makers, various shop keepers including ironmongers, grocers and drapers, tailors and peruke makers, a basket maker and several publicans. Unlike Park Street, metal-working trades such as smithing and brass founding were not well represented – in fact only one smith is listed on the whole street (pers. comm. Stephanie Rátkai). An apothecary and a surgeon, an attorney and an excise officer were also listed for Edgbaston Street along with several merchants, including Sampson Lloyd who had originally owned No. 18 Park Street (see Litherland and Rátkai, Chapter 4). The Edgbaston Street site included or adjoined the site of Hawkers glassworks which was built in 1777–8. The glassworks had gone by 1786–7, although Glasshouse Court is marked on Sheriff’s Map of 1808 (Fig 2.4). Documentary evidence from 18th- and 19th-century rate books (pers. comm. Toni Demidowicz) suggests that in more or less the same area as the Edgbaston Street excavations was a large holding under single proprietorship until the late 17th century. This holding was subsequently multiply sub-let, several lessees being skinners. At some point, probably in the late 18th century, the Welch family, also skinners (although termed ‘leather dressers’ in the 1808 Directory: pers. comm. Toni Demidowicz), managed to appropriate a large holding here, which is shown on Sherriff’s map of 1808 as Welch’s Skinyard. The term skinyard could be used for a variety of processes (see below).

Development continued during the 19th century, and it was at this time that the watercourses were culverted and these and other remnants of the medieval townscape disappeared. The driving force behind this was the rapid rise in the population of the town, which rocketed from around 23,000 in 1731 to around 170,000 in 1831. In 1815 the Manorial Moat was sold by Thomas Gooch to the town commissioners, the buildings were rapidly demolished (including the associated thread mill, once the Malt Mill, in Upper Mill Lane) and the moat filled in. Following a short construction programme, the Smithfield Market opened on the 5th of April 1817 for the sale of cattle, horses, sheep and pigs. A decade later the Parsonage Moat also disappeared under a turnpike road connecting Worcester Street and Bromsgrove Street, which is now called Pershore Street. A contemporary map (not illustrated here) shows how the new street cut through the southeast corner of ‘Reverend Curtis’ St Martin’s Parsonage’. Property deeds of the 1830s and 1840s describe the subsequent build up of structures along the newly created street frontage. At some point between the mid-1820s and the mid-1830s Gloucester Street, running south from Edgbaston Street, close to and more or less parallel with Jamaica Row, was constructed. By the time of Jobbins’ 1838 map, Gloucester Street was in place and the southern Edgbaston Street backplots had been cleared of buildings and replaced by narrow ribbonlike developments along the southern street frontage and along the north side of Smithfield Passage (pers. comm. Stephanie Rátkai).

Fig. 2.3 Detail of Hanson’s Map 1778.

Development of the Gooch estate in this vicinity continued into the 1830s and 1840s and became the home of a vibrant and diverse multi-skilled population working around the market in associated trades and in the growing industry of ‘Birmingham Toys’. Aspects of the documented history of these working-class people and their housing have been intensively studied as part of the conservation of the ‘back-to-backs’ on Hurst and Inge Street near by (Demidowicz 1994).

This speculative early Victorian development can be contrasted with the large Queen Anne and Georgian houses that continued to line Edgbaston Street. An advertisement from 1765 (Aris Gazette) described one of the houses, probably of this period, as a

handsome large commodious house, consisting of a large warehouse with a counting house behind it, two good parlours, a hall, two staircases, a china pantry, three large chambers, each having light and dark closet, many of each of them large enough to hold a bed, a spacious dining room, wainscoated, six good upper chambers with closets, a kitchen, pantry, four large cellars in one of which is a pump, a brewhouse with a pump, and an oven to bake bread, a good stable with a loft over it, a coach house and a large garden, with a canal, and other conveniences thereto belonging…

Fig. 2.4 Detail of Sheriff’s Map 1808.

One of the party walls of a house of this period survived adjacent to Gloucester Street until the latest development and provided useful information about potential archaeological survival along the Edgbaston Street frontage, which was proved by the excavation of Area A where several medieval features had survived. The status of Edgbaston Street increasingly declined during the 19th century; the large houses were subdivided and came to resemble in character the other properties in the vicinity of the market, which were clearly working class. However this decline also reflects a general trend witnessed at Park St for the ‘middle classes’ to move out of the Bull Ring area to new developments such as Old Square and Colmore Row (pers. comm. Toni Demidowicz) and later in the late 18th–early 19th centuries into the suburbs. This period marks the beginnings of the break from home and work being interrelated or interdependent, as it had been from the medieval period onwards, towards the division between home and the workplace. This process was however piecemeal (see Cattell et al. 2002, 23), with family concerns and small workshops still in evidence alongside small-scale concerns carried on in the courts, essentially one-man operations, and purely domestic occupation. This pattern is perhaps most famously seen in the Jewellery Quarter, Birmingham (Cattell et al. 2002).

Another reason for change was the massive growth of the urban population in the 19th century, associated with social changes brought about by the Industrial Revolution. The urban poor were forced to live in unsanitary, and often old, properties commonly situated in courts built behind the more substantial properties lining the street frontages. This situation only began to be seriously addressed during Joseph Chamberlain’s leadership of the Town Council, which corresponded with broader changes in the economic shape of Birmingham, particularly brought about by enhanced rail links. These firmly established the status of the 19th-century wholesale markets situated over the former Birmingham Moat, which continued to grow both in size and importance throughout the 19th century. Later, in the 20th century, the markets suffered set-backs associated with changes in the economic infrastructure of the country as a whole, and in addition they suffered from bomb damage during the war. The overall effect was such that by the late 1950s, when the economy began to expand again, the area as a whole was ripe for extensive redevelopment. Smallbrook Queensway became a crucial element of the Inner City Ring Road, while Edgbaston Street was downgraded further, essentially performing the modern function of a medieval back lane servicing the 1960s Bull Ring Shopping Centre. The wholesale markets, with one of the first grants of European Economic Community Regional Aid, were one of the last parts of the general market complex to be redeveloped in the 1970s, and it was during this work that archaeological investigations on the Lord of the Manor’s Moat took place (Watts 1980).

Excavation results

Phase 1: 12th–14th centuries

The original ground level sloped down southwards from Edgbaston Street, following the natural gradient of the sandstone ridge on which the town was established. Erosion of the soft sandstone produced deposits of soft sand and pebbles at the southern end of Areas A and B. Relatively thin patches of waterlogged organic material overlying the subsoil in Area C suggested that this area was substantially drier than Area D to the south. The presence of a tree bole (F352) suggested a former wooded landscape. In Area D the ground was wet, with patches of marshy ground. The area seems originally to have been wooded, as is evidenced by a number of tree boles. This interpretation is supported by the results of analysis of the plant macroremains and pollen (see Ciaraldi, Chapter 12, and Greig, Chapter 13).

Area A (Fig. 2.5, 2.6 colour & 2.7)

Area A lay towards the Edgbaston Street frontage. In the central third of Area A the subsoil (1002) was overlaid by a thin silt-sand layer (1074). A concentration of features survived against the eastern edge of the excavation area. A square pit (F123) had a recut (F105) and a smaller square cut at its base (F149) (Figs 2.5, 2.6 colour & 2.7; Fig. 2.8, S1). The sides of the pit were vertically cut, with no slumping, which suggests that it may originally have been lined.

The insect fauna and pollen remains associated with the fills 1011 and 1012 (see Fig. 2.8, S1) of this pit are informative (see Greig, Chapter 13, and Smith, Chapter 14). The insect fauna is typical of human settlement, but also includes species that are associated with decaying prepared timbers – strengthening the conjecture that the pit was originally wood-lined. Smith (ibid.) suggests that the pit held mouldering vegetation as well as decaying timbers, which could be consistent with a period of abandonment. Both the insect fauna and pollen suggest that the pit periodically served as a water cistern or tank. The high percentage of woodland pollen – which is unusual for an occupied site – might be accounted for by this pollen having been brought in with the timber for the lining, although Greig also suggests that it might be the result of a period of abandonment. The pollen analysis also showed the presence of grassland; the large numbers of bracken spores present may have resulted from the use of bracken for animal bedding.

In addition to a stone roof tile, the fills of the pit recut (F105) contained a substantial quantity of medieval pottery. The uppermost infilling (1010) contained an almost complete straight-sided cooking pot (Fig. 2.9, colour; Fig. 7.1.19), which may not have been a discard but may have been deliberately placed in the pit for an unknown function (see Rátkai, Chapter 7). The pottery evidence suggests that the pit was backfilled in the 13th century, possibly before c.1250.

Fig. 2.5 Edgbaston St: multi-phase plan of Area A.

Immediately to the east of the square pit, one half of a steeply cut sub-circular pit (F147), which again had a recut (F148), was recorded. The ceramic evidence suggests that this pit was early, with backfilling possibly occurring in the late 12th century but certainly before the mid-13th century. The recut F148 contained no finds. A series of three post-holes or possibly small pits (F134, F135 and F136) was aligned east-west. All of these features cut into a charcoal-flecked sandy silt layer (1042). A further four post-holes (F118, F121, F142 and F143) were recorded across the excavation area. Posthole F121 to the east of F110 contained two cooking pot sherds in Fabric cpj1 and cpj12 (Fig. 7.6.126) and F143, a small chip of reduced Deritend ware.

Fig. 2.7 Area A; phase plans.

Fig. 2.8 Area A and Area C sections.

Towards the western boundary, an oven (F109) survived as a clay-lined cut with a tile floor (Fig. 2.10, colour). A clay ‘plug’ was located near the base of the oven. The absence of charcoal and the presence of an unburnt clay lining completely covering the tiles suggests that the oven may not have been used. However, the upper backfill (1023) did contain a quantity of heat-shattered stones. Pottery comprised two whiteware jug sherds and two sherds from an iron-poor jug (Fabric ip8) which may be later medieval in date. Two fragments of worked bone, probably an offcut from the manufacture of knife plates (Bevan et. al., Chapter 8), were recovered from one of the oven’s fills (1028). However, bone-working debris was mainly confined to Phase 3 pits and the examples from F109 may therefore be intrusive. A second clay-lined feature (F110) may have been related to the oven (F109), but its only partial survival prevents further comparison and there were no finds from its fill.

Area B (Figs 2.11, colour & 2.12–2.16)

Area B was situated immediately to the north of Smithfield Passage adjacent to Pershore Street (see Fig. 2.1). The survival of Phase 1 activity extended over a large part of the area (see Fig. 2.11, colour and Fig. 2.12). A brown organic clay-silt occupation horizon overlay the subsoil. Towards the southern edge of the area, the orange gravel-sand subsoil (2247) was overlaid by bands of water-deposited sand (2017, 2018 and 2036). These may represent the continuation southeast of an early – partially silted-up – water channel (recorded as F213, F255 and F256; see Fig. 2.13), which fed into the larger watercourse linking the Lord of the Manor’s Moat with Parsonage Moat. The water channel had a rounded profile and appeared to be a natural feature. The insect fauna from 2036 suggest that the water channel was filled with decaying settlement waste. Pottery was only found in F256 and dated to the 13th century. Interestingly, sherds from the same vessel were found in the water channel and in one of the tanning pits (F229), providing a link between this water source and its use in the tanning industry.

Extending north from the water channel and watercourse was a small number of tanning pits (F224, F226, F227, F228, Fig. 2.14, S1 and F218/ F232, F229 and F233, Fig. 2.11). These were large, rectangular features measuring up to 5m in length and 0.80m in depth which were variously wood-lined, clay-lined, or simply cut into the subsoil. One of the pits (F224) had the remains of a wooden beam at its base. A number had stake-holes associated with them. The majority of tannery remains were contained within one plot of land – the limits of which were still bounded by Phase 3 brick walls.

Fig. 2.12 Area B; phase plans.

The plant macroremains and insect fauna from pit F218/ F232 confirm that water from the nearby watercourse was being used to fill the tanning pits. The presence of bark scleroids – which are consistent with the production of tanning liquor – and of ‘bark beetles’ associated with bark or timber attests to the presence of the tanning industry. The recovery of fragments of decomposed leather also demonstrates this material’s contemporary production (see Ciaraldi, Chapter 12, and Smith, Chapter 14). The range of insect fauna suggests that the pits would have remained open for a long period of time which is in line with the tannery continuing in use to Phase 3.

Further Phase 1 survival was recorded in a second plot of land to the east (see Fig. 2.12), including a large rectangular-shaped tanning pit (F237) and a cluster of features represented by an east-west aligned V-shaped gully (F265), its recut (F270) and a steeply cut sub-circular pit (F266), shown on Fig. 2.15, S1 and Fig. 2.16. Neither F265 nor F270 contained pottery but the clay-silt backfill of F270 contained leather offcuts. F266 contained three medieval cooking pot sherds. Insect fauna from the gully (F265) and pit (F266) was characteristic of human settlement and associated rotting waste.

Three pits (F267–F269) were recorded at the eastern limit of the excavated area (Fig. 2.12). Two of the pits (F267 and F268) were sealed by a thin layer of grey sand-silt (2230) before being cut by the third larger pit (F269). None of these contexts contained pottery or other artefacts apart from F268, which contained four sherds of pottery unlikely to be later than the 13th century. These comprised a mudstone-tempered ware jug sherd with roller stamp decoration and three cooking pots sherds (Fabric cpj1).

Area C (Figs 2.17, 2.18 colour & 2.19)

A cluster of three features – a north-south aligned gully (F350), a small pit (F352) and a larger pit (F351) – survived against the northern edge of the excavated area (Fig. 2.17; Fig. 2.18, colour and Fig. 2.19). The gully (F350) was filled with a fine grey-silt (3151) and was cut by the larger pit (F351). The truncated remains of three more pits (F306–F308) were recorded in section only. A whetstone or rubbing stone was recovered from the fill of pit F308. A thin occupation layer (3025) and an organic layer (3115, 3165) were recorded 2m to the south. A shallow rectangular pit (F348) and a post-hole (F349) survived to the west of a Phase 3 well (F327).

Fig. 2.13 Area B; photograph showing location of sections and major features, Phase 1.

Fig. 2.14 Area B; section of F224, Phase 1 and F208, Phase 2.

Fig. 2.15 Area B; section of F266, showing associated features, and of ditch F253.

Fig. 2.16 Area B; Phase 1 features, F265 and F266.

Fig. 2.17 Area C; multi-phase plan.

Area D (not illustrated)

A tree bole (F426) and two peaty, organic layers overlying the subsoil (recorded as 4031, 4032, 4049 and 4050) represent the earliest remains in this area. There were no artefactual finds from these layers and it is possible that these organic layers pre-date the Phase 1 establishment of the town. Charcoal and whipworm, indicators of human habitation, for example, do not appear until the middle sample in the monolith. The plant macroremains and pollen analysis show that this area was characterised by some woodland cover and wetland vegetation, although the insect fauna is more characteristic of open and disturbed land. There is clear evidence that this was a marshy area with pools of shallow water and that the area was subject to periodic flooding up to and including the first part of Phase 3. The presence of parasites in the pollen samples and specific species of insect fauna associated with settlement waste attests to a low level of human activity within the area.

Fig. 2.19 Area C; phase plans.

Transects A–D (Fig. 2.1)

A watercourse ran northwest-southeast across this area joining Parsonage Moat with the Lord of the Manor’s Moat (see Fig. 1.2). Unlike channel F213/ F255/ F256 in Area B (see above), the watercourse had a flat base and straight sides suggesting that it had been a deliberate construction, probably regulating the course of a preexisting natural watercourse or spring. Phase 1 primary and lower fills were represented by waterlogged, peaty deposits and gleyed black clays contexts (5043), (5051), (5110), (5401) and (5402). These were sampled extensively for waterlogged plant remains, pollen and insect remains (see Ciaraldi, Chapter 12; Greig, Chapter 13; Smith, Chapter 14) which suggest that the watercourse was free, but slow-flowing, possibly due to the reed beds and aquatic plants which were also recorded. The presence of a few trees was noted and this is supported by the evidence of tree boles in the adjacent excavated areas. The watercourse was adjacent to an open landscape with some hedgerows in Phase 1 and it is likely that this represents the backplots extending from Edgbaston Street and the open land to the south of the watercourse. The insect fauna gives a strong indication that animals were being kept on this land, supporting the suggestion that land at Edgbaston Street served as a stocking yard to rest and water animals prior to their sale at the Bull Ring market.

The insect fauna is characteristic of dense urban settlement, whilst specific human activity in this phase is indicated by the presence of cereals and hemp, which was used for making rope and canvas. Some sewage and urban contamination of the watercourse is also suggested by the pollen and insect fauna remains – the occupants of Edgbaston Street were not averse to throwing their household waste into the nearby watercourse.

Phase 2: 15th–16th centuries

Area A (Figs 2.5, 2.6 colour & 2.7)

In Area A, the only evidence of Phase 2 activity was a discrete layer (1083) against the eastern edge of the site, close to the Phase 1 pit and recut (F123 and F105, see Figs 2.5 and 2.6, colour). That there was some activity in this phase is evidenced by Phase 2 pottery found residually in Phase 3 contexts. This formed a minimum of 15% of the Phase 3 pottery. This may indicate that activity concentrated more towards the street frontage, evidence for which has been lost through later development. Alternatively, the plot could have been temporarily abandoned or the focus could have shifted to more industrial concerns at the back end of the plot, like those witnessed in Area B.

Area B (Figs 2.20–2.23)

The tannery established in Phase 1 continued in use throughout Phase 2, incorporating a number of tanning pits, most of which can be seen on Fig. 2.20 (F208, F209, F214, F217, F223, F230 and F279). Although the pits were cleared out at intervals, the entire contents of the pits were not always removed, resulting in multi-phase pit fills in some instances. A clay lining was recorded for two of the pits (F214 and F223). The wooden lining of F208 partially survived on its southern edge (see Figs 2.21 and 2.22, and Fig. 2.14, S2). Tanning pit F214 also had two stake-holes (F219 and F220) on its eastern edge, whilst tanning pit F279 had three (F276–F278, see Fig. 2.23).

Finds from the pits and associated fills offer glimpses of the various activities occurring on and around the area. The recovery of perinatal and unweened calves and of perinatal sheep from pits F214, F223 and layer 2196 suggests that livestock were kept on the site. Pit F208 contained a whetstone (see Fig. 8.12.2), which was broken at one end but which had a carved and decorated terminal at the other, possibly dating to the 15th century (see Bevan and Ixer, Chapter 8). The fill of this pit contains mainly early Phase 3 pottery, although there is some Phase 2 material in it and there is no reason why pit F208 could not have been dug in Phase 2 and continued in use through to Phase 3. The upper fill of pit F214 contained a fragment of seam i.e. a small strip of leather with stitching holes (see Macey-Bracken and Mould, Chapter 8). A palmate antler tine and an oyster shell fragment were also recovered from this pit and leather offcuts were found in F217.

Activity on either side of the tannery was limited. Only three other features, a pit (F222), a post-hole (F221) and a 1m-deep north-south aligned linear (F254), which heavily truncated the earlier Phase 1 water channel, were recorded to the west. Pit F222 contained two sherds from a coarseware jar of later 16th- or early 17th-century date. Pottery from the linear was probably more or less contemporary with that from F222. Some interesting ceramic evidence, including drinking vessels, and vessels for storage and food preparation (see Rátkai, Chapter 7), was recovered from a series of layers recorded across Area B including 2035, which sealed the backfill of tanning pit F214.

Fig. 2.20 Area B; photograph illustrating major Phase 2 features and sections.

Fig. 2.21 Area B; Phase 2 tanning pit F208.

Area C (Figs 2.17 & 2.18 colour)

Phase 2 activity is limited to a single occupation layer (3099; see Figs 2.17 and 2.18, colour) at the southern limit of the excavation area – immediately to the north of the watercourse. This layer contained three quite large sherds from a cooking pot (Fig. 7.5.117), a reduced Deritend ware cooking pot, and a late oxidised ware jug (Fig. 7.7.147).

Area D

No Phase 2 activity was recorded in this area, although two late oxidised sherds were recovered from the machining layer 4019.

Transects A–D

No Phase 2 activity was recorded in