Artefacts and the Study of Life in Roman London

  • Details
  • Transcript
  • Audio
  • Downloads
  • Extra Reading

From jewellery and domestic utensils to weaponry and religious objects, artefacts provide a vivid connection to London's Roman past. 

This lecture explores the ways in which archaeologists can analyse and interpret these objects, presenting some results of the latest research into the artefacts from excavations in the City of London and the new insights they provide into the lives of the first Londoners.

Part of the Mondays at One Archaeology Series.

Download Transcript

05 October 2015

Artefacts and the Study of Life in Roman London

Michael Marshall

These days it is not uncommon to hear people worry that as a society we are all together too concerned with things and material possessions rather than people. This is no doubt a legitimate concern but it is important to remember that this interest in ‘things’ is hardly a new one and how could we not be concerned with our homes and workplaces, clothes, personal possessions, tools and utensils, means of transport, treasured mementoes and cold hard cash. All of these are very literally a part of the fabric of our lives. For archaeologists, who study material culture from the past, whether we are uncovering the layout of ancient streets or a sherd of pottery, that that is exactly what we are dealing with, fragments of human lives.

The rediscovery of Roman London is an ongoing process which in the 21st century principally takes the form of developer funded excavation, undertaken by MOLA and by other professional archaeologists, or research into the huge body of evidence that has already been recovered. If we are simply looking for traces of Roman lives then a walk through the galleries here at the Museum of London or a visit to the Archaeological Archive will demonstrate that we have a lot of evidence to work with. The big challenge is finding ways to look at these artefacts and to extract information from them so that we can use them to understand the past.

Paul Bahn memorably described the whole process of archaeology as a kind of jigsaw from hell which will never be finished, where half of the pieces are missing and where someone has pinched the lid so you cannot look at the picture. On bad days I can empathise. But today I hope to draw on some examples from my own work and from the work of others, not to try and tell some comprehensive account of Londinium but rather to show that there are patterns there to be found amongst the fragments and progress to be made in piecing together and understanding the lives of Roman Londoners through things.

My talk will fall into four sections:

1. Some examples of finds of great intrinsic interest which reward close study and give us opportunities to explore different aspects of Roman life

2. Some ways in which we can interrogate and compare larger groups of objects, in this case Roman dress accessories, in order to finds out something about how they were used and who used them

3. Some examples where the archaeological context helps with the interpretation of objects and vice versa

4. Finally I will zoom out and consider a few patterns in the distribution of artefacts on a wider scale to see what small finds can offer to our understanding of an big city

Artefacts that tell stories

An amber pendant in the shape of a gladiator’s helmet

This small carved amber pendant was found in amongst the mud on the banks of the River Walbrook during recent excavations at Bloomberg London. Though stylised it is beautifully made in the form of a helmet with a crest on top, a flange around the neck and a grille visor all clearly visible. This is not the helmet of a Roman soldier but rather the helmet of a gladiator and there are sufficient details to suggest that it probably belongs to the murmillo type who fought in the arena armed with a shield and short sword and was particularly popular in the early Imperial period.

It is not surprising to find gladiatorial imagery in London. Our amphitheatre was established around 70 AD and to judge by the excavated portion it had the capacity for thousands of spectators. Enthusiasm for the spectacle of the games is evidenced not only by the size of this venue but also by the fact that gladiators are depicted widely amongst finds from the city and while our amber amulet is unique these other objects provide a clearly defined context for it. Examples include mould blown glass cups, ceramic oil lamps and decorated red samian tableware. It is probably no accident that while our amber piece was individually made, these are all made in moulds they were probably objects produced for a mass audience.

Often these can be simply regarded as genre scenes but in some cases we might go as far as to call them sporting memorabilia, perhaps made to celebrate specific bouts, and the names of gladiators appear above their heads. This cup found at Old Jewry and dating to the first century AD depicts the battle between a murmillo and a thraex two very popular types of gladiator. However, it is the only example known from this particular mould. An otherwise unattested gladiator Burdo has dropped his shield and raises his hand in supplication to the audience while his opponent stands ready to deliver a killing blow. Perhaps then this unique design is a testament to a career quickly ended by defeat or by death.

There are a number of sources of amber in Europe but during the Roman period the most important was the Baltic sea area and amber was collected there and transported significant distances to workshops elsewhere, particularly around Aquileia in northern Italy. Such an origin is perfectly possible for our gladiator’s helmet pendant. Classical sources tell us that amber was not only valuable but was also thought to possess beneficial magical properties. So for example Pliny (Naturalis Historia XXXVII.30) and Artemidorus (Onirocritica 2.5) mention that for women to dream of amber rings signalled good fortune and that amber amulets could be used to protect young children from illness.

Might our example be such an amulet? I think there is good reason to think that it was. A perforated amber object of similar size but different subject is known as part of a group of amulets in the burial of a young child at Colchester and similar patterns can be found further afield including another example in the form of a gladiator’s helmet found in a grave in northern Italy (Calvi 2005). The choice of the gladiator helmet imagery might have been appropriate as it is a mode of protection and also because gladiators, when fighting, are striving to survive and against death. That gladiators were thought to be symbolically potent in London at this time is perhaps suggested by an extraordinary cremation burial from Southwark. Here a woman was buried alongside a group of tazze incense burners and lamps depicting both gladiators and the Egyptian deity Anubis guardian of the dead. At the time of the excavation some speculated that the burial might be that of a female gladiator but at best the suggestion is unproven. Laying that issue to the side for a moment we might consider that the inclusion of the fallen gladiator imagery alongside that of Anubis in this burial indicates that gladiators were symbolically associated with issues of life, death and protection (see Bateman 2008). We can never prove it but it seems likely to me that this gladiator head was probably a protective amulet bought for a child by wealthy parents perhaps because they were sick. The child may have been an enthusiast for the heroes of the arena and enjoyed it for its imagery but it was also probably hoped that it would keep them safe and well.

A wax and wood writing tablet

In London archaeologists are blessed and cursed in equal measure by copious amounts of soggy dirt. Not always fun to excavate, waterlogged deposits alongside the river Thames and the Walbrook stream do, however, preserve organic objects made of wood, leather or textile that are rarely available to us elsewhere. This includes the Roman precursor of the etch-a-sketch or IPad, the wax and wood writing tablet. Scribes used a metal stylus to score through black wax page and reveal the pale wood below. The text could then be erased by smoothing the wax and the tablet reused. When we are lucky the original message can survive impressed into the wood beneath and, a bit like the impression on a telephone pad, can be retrieved.

Pain staking work on this tablet from 1 Poultry by Dr Roger Tomlin allowed him to read the text and provides some incredibly vivid insights into the Roman London.

Vegetus Montani imperatoris Aug(usti) ser(vi) lucundiani vic(arius) emit mancipioque accepit puellam Fortunatam sive quo alio nomine est natione Diablintem de Albiciano LEG[...] (denariis) sescentis ea(m)que puella(m)que de qua agitur sanam traditam esse erronem fugitivam non esse praestari quod si qu[i]s eam puellam de qua agitur par[tem]ve quam [evicerit] cera quam pe[r geni]um [imperatoris] Caesaris scr[ipsit iura]vitque [...]ARIS

'Vegetus, assistant slave of Montanus the slave of the August Emperor and sometime assistant slave of lucundus, has bought and received by mancipium the girl Fortunata, or by whatever name she is known, by nationality a Diablintian, from Albicianus [...] for six hundred denarii.

And that the girl in question is transferred in good health, that she is warranted not to be liable to wander or run away, but that if anyone lays claim to the girl in question or to any share in her, [...] in the wax tablet which he has written and sworn by the genius of the Emperor Caesar


(Translation and transcription after Tomlin 2003; Hill and Rowsome 2011)

This then is the sale document for a slave and an important reminder of the role that slavery has played in London’s history from the very beginning. It is couched in formal legal language and clearly lays out the terms of sale and liabilities of those involved giving us some sense of the legal framework. We can put the sum of 600 silver denarii in some context as we know in the late 1st century under Domitian (AD 81 – 96) a soldier’s annual pay had risen to around 300 denarii so we are looking at around a significant expense, which would equate to around two years wages and for many people probably rather more.

The tablet is also a reminder of how complex the legal status of individuals could be in the Roman period. Slaves could own other slaves. They could also hold relatively important positions in society and indeed the fact that a slave of the Emperor is in London may indicate that he is here on some kind of official business. But while it was possible for some slaves to lead successful lives and eventually gain their freedom there is little doubt that for others life was grim and cheap. While Fortunata could supposedly be trusted not to run away London has produced quite a number of shackles or manacles at least some of which are likely to have been used for imprisoning slaves and which speak for a significant section of Roman society whose labour probably produced a lot of the material things that we study but who are otherwise now mute.

Around 400 writing tablets have recently been recovered from excavations in the Walbrook valley and are now being studied by Dr Tomlin. Analysis of the wood from which they are made by Karen Stewart has shown that, like earlier finds, the great majority are made from silver fir (abies alba) a non-native timber. Analysis of their size and manufacture strongly suggests that they were being recycled from the staves of silver fir casks or barrels which were imported from continental Europe probably containing alcohol. These were not worth returning empty and so were often recycled either by reworking them to make other objects or using them to line wells.

The Poultry writing tablets and other inscriptions and pieces of writing equipment also offer us routes into understanding literacy and its role in Roman London. Wide ranging surveys of writing equipment by Hella Eckardt (2014) and John Pearce (2004) show that London has one of the biggest corpus of writing equipment in Roman Britain rivalled only by intensively excavated military sites with comparable excellent preservation such as Carlisle and Vindolanda. Indeed certain classes of objects such as high quality metal inkwells are heavily concentrated here a fact which may indicate the wealth of some of those involved and also a degree of ostentation associated with the practice. Gwladys Monteil’s work on the distribution of more humdrum samian inkwells (2008) shows that that their use was initially focussed on the area of Roman London’s town centre to the east of the Walbrook and that they then spread. In general there is a sense that they may cluster around areas where we would expect commercial and military administration to be taking place and this is supported to some extent by the content of the writing tablets which is frequently of a commercial nature and in several instances is written by identifiable soldiers. Literacy was probably far from universal but seems to have been an important tool for achieving economic success and holding and exercising power. New work on writing tablets by Dr Tomlin is currently underway and those interested in a wider study of inscription in Roman London are referred to an excellent survey article by Nick Holder (2007).

An enamelled early Roman flask from London

This small hexagonal enamelled copper-alloy flask from Moorgate is one of my favourite finds of recent years. Through close examination and comparison to similar finds we can begin to understand it better. So while we already have quite a large portion of it the probable full form can be reconstructed with reference to a near identical example from a 2nd century cremation burial at Corbridge in northern England (Casey and Hoffman 1995) and that example also confirms the stylistically implied dating. In total there are a little over a dozen examples of the type from the whole Roman Empire and their distribution is concentrated in Britain with a few outliers in Germany and on the Black Sea coast. This suggests that like several other forms of enamelled vessels they were actually made in Britannia and the decoration combines a classical derived design on the shoulder, not dissimilar to that from the ink well lid on my previous slide, with a repeated pattern of slender trumpet scrolls on the sides that clearly has its origins in Celtic art (Hunter 2012).

The exact site where these flasks were manufactured cannot yet be determined but there are some good reasons to believe it lies in northern England. A series of 2nd century enamelled pans inscribed with the name of forts from the western part of Hadrian’s Wall was perhaps made somewhere around Carlisle, while moulds for related enamelled vessel types of similar date have been found at Castleford in Yorkshire (Cool and Philio 1998). Enamelled finds belonging within this tradition are known from London and other sites in the south east suggesting that there was clear market for them.

The contents of this flask are long gone but based on its small size, its narrow neck and the fancy decoration we might suspect that it contained some kind of liquid, perhaps an expensive one to be used in small volumes. Pairs of holes in the shoulder are similar to those used to attach carrying handles to other types of vessels such as oil flaks for bathing and all together it seems most probable that it was some kind of toilet flask perhaps containing unguent, perfume or oil associated with toilet practices with their roots in the Mediterranean.

This then is a fascinating Romano-British object its form and function meeting new, very Roman, needs while its decoration harks back to Iron Age tastes. London has much to offer those studying the development of this sort of hybrid culture as while the city seems to have been a new foundation of incomers from elsewhere in the Roman Empire it is clear that a taste soon developed for British Iron Age style objects amongst sections of this immigrant population or amongst locals who were moving into the town. Initially these are the very same types of objects decorated in this style in the Iron Age and fragments of weapons, horse gear and drinking vessels are all know from the city. Some of these come from contexts suggesting a close association with the Roman army and links with earlier traditions of male military culture (Hunter 2008).

Other objects were modified in an ad hoc manner to play new roles in this new cultural milieu such as this Iron Age silver coin which may have been worn as a piece of jewellery. However, whether this was a retained token of identity worn by a Briton in an alien city, a trophy claimed by a Roman soldier or simply a souvenir is not easy to determine. What the Moorgate flask demonstrates is that by the 2nd century craftsmen working in an Iron Age style were deliberately adjusting their output to meet the needs of new Romano-British consumers.

From assemblage to ensemble

Perhaps the easiest way to explore dress traditions and their relationship with ancient people is through iconographic and literary evidence but these sources can often be skewed in their coverage towards certain sections of society, particularly the literate and wealthy. In fact both are fairly scarce for Roman Britain. For the later Roman period in situ finds worn by the deceased in inhumation burials increasingly become a valuable source of evidence but these are not common in the early Roman period. Instead I hope to show that even the more run-of-the-mill evidence, objects routinely discarded by ancient Londoners and found in rubbish pits or mixed up layers, can provide interesting insights into how they dressed.

Brooches from the Middle Walbrook valley

Brooches are amongst the most common types of objects found on Romano-British sites particularly those of the 1st and early 2nd century and Roman London has produced more than 800 examples. They have an extensive literature devoted to their classification, dating and distribution (e.g. Riha 1979; Feugère 1985; Bayley and Butcher 2004; Mackreth 2011).

It is clear that brooch types are variably distributed across the Empire and their range varies from the strictly local to the inter-provincial. So while some such as the Aucissa type can be found from North Africa to northern England others such as Dragonesque or Wirral brooches are concentrated in specific counties or regions and are very rare outside of Roman Britain. The influencing factors are probably quite varied ranging from the location and output of workshops to culturally defined patterns of taste and the degree of mobility of the population in a brooches ‘homeland’.

Some brooches may have been deliberately worn as badges of identity, in the later Roman period iconographic sources and grave associations show that some were clearly badges of rank, but even if this conscious element is excluded it is clear that in many instances it is possible for us to distinguish between a brooch in a ‘British’ style, principally made and used on this side of the channel, and others that are characteristic of areas elsewhere within the Empire. As such these common dress accessories are useful tools for exploring connections between places.

Recent excavations at Bloomberg London have produced a large assemblage of around 170 brooches. Here those on the left are continental types while those seen on the right are British types. If we consistently classify brooches then we can compare sites or larger areas in terms of the popularity of different types. In this example we have sub-divided the brooches from the site into broad categories (after Plouviez 2008) and compared the resulting profile to a British mean and a sample of earlier finds from London. The results are interesting as while the Bloomberg brooches have obvious similarities to the others there are also major differences. Chronologically early types are extremely important indicating intense activity during this period. A notable lack of later brooches is probably a reflection both of a shift towards clothes types that didn’t need to be fastened with brooches in London and less activity on this site in the Late Roman period. The relative proportions of 1st century types, however, are also interesting and particularly the atypically high amount of continental hinged bow brooches and early continental plate brooches. The site then has clear evidence for continental connections amongst its brooches, indeed more so than almost any other site in Britain with the exception of some early forts. As previously mentioned London was a new city founded by incomers and the likelihood is that this prevalence of continental brooches is a reflection of this.

However, as we shall see later, these brooches do not seem to have been equally popular amongst all sections of society and after the initial influx the popularity of brooches such as the Hod Hill type begins to wane. Although some will have continued to travel to London from continental Europe either as personal possessions or for sale it seems that local products increasingly dominated the market. As with the Celtic objects described above this may reflect both changes in the taste of the initial immigrants and a secondary influx of Britons from the surrounding countryside coming into the new city. The ease of access to British styles may have played a big part in consumer choice when a broken brooch was being replaced but, if there was a market for them, there is no reason that continental styles of brooch could not have been made in London. Instead by the beginning of the 2nd century AD London has largely diverged from continental brooch fashions. Instead the typical form of bow brooches are those made in this country with a loop at the head reflecting the widespread and distinctive British practice of wearing brooches in pairs with a chain strung between them. This can be seen clearly in the case of a pair of Headstud brooches found together with their chain at 1 Poultry (Wardle in Hill and Rowsome 2011).

Changing hairpins, changing hairstyles?

Hairpins are a type of dress accessory specifically associated with Roman women and were used alongside various other implements to arrange and fix female hairstyles. They are amongst the most common classes of finds from Roman London and provide evidence for the widespread adoption of this hairdressing technology. Well over 1000 examples have recently been digitised by a fantastic team of volunteers working at the Archaeological Archive and you can read about the project and view the results yourself on Museum of London Collections Online. There are a number of quite simple types but as you can see here they are really quite varied and some are highly elaborate. Like brooches some types are regional and there are also trends in material, shape and size which may indicate the wealth of the owner or functional differences. This last point has been explored by Dr Hilary Cool (1991) who demonstrated that metal hairpins get shorter over the course of the Roman period and argued that this reflects changing hairstyles as evidence for example in sculpture and portraits on coins. Similar trends have recently been observed in bone pins from London as part of a UCL MA thesis by Carolina de Rangel Lima (2014). The social messages bound up in elaborate and variable hairstyles are likely to have been complex. We know that Roman women dressed their hair for special occasions and elaborate styles might also have economic implications because of the time and expert help required to achieve them. In many cultural contexts there are prescriptions or taboos about what women should or should not do with their hair.

I do not have time to go into that in great detail today instead I am going to simplify things and focus on this idea of chronological change and consider how broad shifts in the use of different types of hairpins might be able to tell us about the uptake in these hairstyles and how they changed over time. Because we have lots of hairpins from modern excavations we can normally provide a fairly accurate date for when they were discarded on the basis of the context they come from and what datable objects such as coins and pottery were found with them. By combining lots of these dates we can define and compare changes in the popularity of different forms and perhaps different associated fashions or other social practices.

So here are some morphological groups of pins for which we have quite a lot of dates from London. The graph here is based on about 200 pins. Firstly here are the two major forms of bone pin; slender tapering forms in blue and examples with knob heads and swelling shafts in red. I hope you can see a broad pattern in which the blue line declines as the red one rises. This pattern has previously been identified by scholars such as Stephen Greep (1983) and Nina Crummy (1979) but this is a really nice demonstration of it and to me suggests that here one form is essentially displacing the other. This is particularly evident in the years around 200 AD. Perhaps even more interesting is what is going on even earlier than this change. Bone hairpins are rather scarce in London during the 1st century AD and then seem to rocket in popularity in the 2nd century. This seems to be a genuine pattern as we have extensively excavated in 1st century contexts in London and while the availability of datable contexts for each period may have an impact the same peak does not appear for other classes of objects and we have excavated enough early sites to be confident about the trend.

So what does it actually mean? Well that is still a work in progress. Cemetery evidence suggests that there was some demographic imbalance towards men in early Roman London but it seemed unlikely that there were no women. Instead I think it is more likely that if reflects a lack of women with those particular Roman hairstyles that required pins. Hairpins are absent in the pre-Roman Iron Age and so it may be that many of the women were of local background or else female continental immigrants that did not feel the need to use hairpins either because they wore simpler styles or used an alternative form of hairdressing technology such as stitching hair together with thread (Stephens 2008). The 2nd century peak might simply indicate a shift in hair dressing techniques or in hairstyles, perhaps to those requiring larger numbers of pins or those more likely to shed pins leading to more being lost.

I was a little surprised when we added the evidence for metal pins to the picture. Overall these are much rarer than the bone pins but here we see not only the 2nd century rise but a prominent earlier peak not visible in bone. This seems to indicate that hairpins were being used in the first century but perhaps by a smaller group of people and using examples made of a more expensive material. We could hypothesise that in the 1st century these sorts of hairstyles were worn by a smaller group of Roman women, perhaps the wealthy, and that the explosion of cheaper bone versions indicates uptake amongst a much wider section of the population from the beginning of the 2nd century. Frustratingly there are relatively few inhumation burials of this period and so we do not have thein situ hairpins to really confirm or disprove this idea. Regardless there are some clear trends in how these hairdressing tools were being used over time and it will be interesting to see how closely this corresponds with the evidence from other Roman sites. Experimental work and close comparisons of the shifts in dated iconographic depictions of Roman female hair may help us to develop these ideas further.

Bracelets for girls and armlets for boys?

My next case study deals particularly with bracelets and asks the question who was wearing them and how that may have changed over time. I am a firm believer in that fact that within artefact studies, size really does matter, as it is a variable that can impact on how objects function and how people could interact with them. Where a bracelet is made of a rigid substance and is either slipped on or has a closed fastening it can also tell us something about the size of the bodies of those involved.

While there are certainly exceptions to the rule, bracelets are conventionally considered to be a female dress accessory and to have been most popular in the Late Roman period. This is borne out by the large number of bracelets of this date, many of which can be associated with female inhumation burials in the Eastern Cemetery of Roman London (Barber and Bowsher 2000).

What is slightly more unusual is the relatively large number of bracelets that have been recovered from early Roman contexts in the city in recent years. These included some types rarely seem elsewhere such as iron wire bracelets and twisted polychrome bracelets which made use of both iron and copper-alloy. Their survival here is in part due to the presence of waterlogged deposits but their date and character suggest they represent a slightly different phenomenon from the predominantly late bracelets found elsewhere.

Also common are lathe turned bracelets made out of black or grey shale. Interestingly if we compare the size ranges of these bracelets with those found in the Late Roman cemeteries we can see that they tend to be larger and curiously there is a gap in the size distribution where bracelets worn on the writs of adult females are normally found. This perhaps indicates either that they were worn by adult men, or particularly that they were being worn differently, perhaps on the ankle or the upper arm. A recent survey of bracelets made from shale and similar materials on the military sites of early Roman Scotland noted a similar pattern with a predominance of large examples and a similar trend amongst early shale bracelets from the legionary fortress of Caerleon in Wales (Hunter 2014).

Also amongst the metal bracelets are a type which Nina Crummy (2005) has convincingly argued are armillae, military battle honours known through literary and iconographic sources, awarded to soldiers who fought in the campaign of conquest in Britain. Some of the metal wire bracelets are small sizes that must have been worn by children. Few seem to be as large as the stone examples but it is also perfectly possible that some of the fragmentary examples would have been sizes appropriate for adult men or women to wear on the wrist.

Large examples from Roman London have been found at the fort and in the Middle Walbrook valley where they are associated with militaria. As such a picture seems to be emerging in which large stone bracelets, perhaps worn by men on their upper arms, were popular in early military communities and in some cases may have been worn alongside other metal styles on the wrist.

These three examples have shown that there was considerable diversity in the way that people dressed in Roman London across time and between individuals and that there are all sorts of connections to be chased. These were dynamic shifts in the ways that both men and women expressed their identities and given the lack of iconographic evidence and the scarcity of inhumation burials close study of the dress accessories themselves are crucial if we are to find a way to address these issues.

Artefacts in context

Sometimes fate preserves some of this important contextual information for us, allowing us to get some sense of where and how specific objects were used, which other objects they were used with and who used them.

Building assemblages

One of my favourite examples from the city is this substantial building on Gresham Street which was destroyed by fire (Casson et al 2014). In these photographs you can see an elaborate mosaic floor probably from a reception room and there is a more modest room through the wall. Here a group of pottery can clearly be seen amongst an area of burning. It is substantially complete and seems to form a group. All of it probably originally sat on a shelf in what appears to have been a storage room or perhaps a corner of a kitchen containing ingredients. Careful work by the excavators preserved these associations and can allow us to start thinking about pottery vessels as functional sets and in terms of how they fit into the Roman home.

Even where we lack such deposits, which almost freeze moments in time, it is sometimes possible to associate specific rubbish pits or midden heaps with individual properties. Taking such approaches across the whole range of domestic material culture in Roman London allows us to get a sense of how people of different backgrounds and means might have lived and that is reflected very well in the wonderful displays here in the Museum of London that use real material to reconstruct a range of different rooms and different households from modest dwellings to luxurious townhouses.

Handled jars in wells

While the context in which an object was found can be a product of how they were used or casually discarded as rubbish there is also some good evidence for situations in which objects were deliberately placed into the ground. The burial of objects as grave goods with the deceased is one example that I have already mentioned but there are other possibilities such as the concealment of objects for safe keeping or even the placement of votive deposits, burying objects as a way of making gifts of them to the gods.

Distinguishing and interpreting such deposits is not always easy. For example a series of complete pots belonging to the form seen on the left have now been found at the bottom of several wells in Roman London. They are not broken and as such are probably not rubbish in the conventional sense. Furthermore enough unusual things have been found down well in Roman Britain, from hoards of valuables to human remains, to make us suspect that these types of features were favoured places for making votive deposits perhaps because of the literal importance of a clean water supply which could be guaranteed by the gods or because symbolically they were seen as shafts that crossed down into another world.

However, wells are unusual environments in other ways. If you accidentally drop something down a well the water at the bottom may cushion the fall and stop if from breaking but the depth and damp also make climbing down the well after the pot a fairly uninviting prospect. In fact these handled jars have been found in wells elsewhere and bear at least a passing resemblance to those held by water carriers in iconographic depictions (Croom 2011). Or course it is perfectly possible that these are water carrying jars AND were votively deposited. Perhaps returning a jar of water to the well was an appropriate way to ‘give something back’ and thank the local spirits for providing years of clean water.

Foundation deposits

Perhaps more overt is a tradition of ‘foundation deposits’ placed under floors or within walls during episodes of building and conventionally interpreted as offerings to the genius loci to bring good luck to the building and its occupants (Merrifield 1987). Here is an example of a beakers buried under a floor at Bloomberg London with a lid on top and a brooch inside. London has produced many examples and while they vary in character they often include a complete pottery vessel, perhaps containing libations, foodstuffs and occasionally personal objects. Sometimes this material is deliberately broken or burnt and there are hints of quite complex rituals and beliefs associated with these offerings.

One of the most interesting is an example from the site of the Walbrook Building between Cannon Street and Bank stations excavated some years ago. Here archaeologists found a small pottery vessel buried within the wall of a 1st century building of mixed domestic and industrial function. It was carefully excavated and inside was found a burnt fruit, a snapped military style belt buckle, a small fragment of a broken amber object, perhaps another amulet, and an uncut chalcedony intaglio of the sort normally inscribed with a design for a Roman seal ring. Also within the pot was a bone die and a broken iron key was placed in the open mouth (Marshall in Blair in prep).

Something quite complex has clearly gone on here. A selection of different objects, some quite valuable, has been very deliberately brought together and treated in highly specific ways according to some kind of scheme that made sense to those involved but which is difficult for us to fully understand. The deliberate breaking and burning of objects may have been a way to put them beyond the world of man and dedicate them to the gods and we might imagine that these different stages were accompanied by some ceremony.

We can only speculate at the significance of the individual objects but we can at least make educated guesses. Perhaps the amber amulet was designed to provide some protective magic, the buckle and the uncut gem might make reference to the professions of those involved after all we know that metalworking took place within the building so perhaps one of its occupants was a jeweller. Personally I like to think the die was rolled into the mouth of the pot as a symbolic gamble to bring good fortune upon the new building and perhaps the start of a bold new enterprise. The iron key was then used to seal the offering in place.

In fact more recent work since has made me more confident in this interpretation. Underneath the wall of a contemporary building elsewhere in Roman London archaeologists have recently found a phallic bone amulet suspended on a string of glass melon beads along with small lead alloy miniature model of a butcher’s cleaver and another bone die. To me this seems very much like we are seeing similar elements brought together (a symbol of the individual’s profession, an object with amuletic properties to provide protection and the roll of a die to bring good fortune) and most likely reflecting similar beliefs. In fact a possible third variation on this theme of similar date can also be identified. Some years ago Dr Martin Henig published a group of inscribed intaglios buried together at Eastcheap and on stylistic criteria he suggested they might be a craftsperson’s stock. These were found alongside near complete pottery vessels and more recent work in the archive (Wallace 2014; Davis et al in prep) has identified the fact that this same deposit included another bone die. This seems very much like a pattern developing which is providing us with a surprisingly clear picture of the superstitions of some Roman Londoners during the 1st century AD.

Small finds, big city

In the last part of my talk I would like to zoom out a little, beyond a focus on individual objects or contexts, to consider ways in which the study of artefacts can help us to understand how an urban centre like London functioned. Decades of developer funded excavation by MOLA and other archaeological units have provided us with a wealth of Roman material culture from hundreds of sites across the city. It is only in recent years that the potential of this evidence is beginning to be fully exploited and the availability of more sophisticated databases and geographical information systems have helped immensely.

Looking at this scale the things that really catch our attention are the big public buildings like the forum and the amphitheatre, differences in the layout of streets and different styles of architecture. These might all reflect variation in function and status. Indeed some have gone as far as to argue for quite specific zoning of different parts of the city and suggested that different communities whether that is the Roman army, groups of immigrants from different areas of the Empire or local Britons. For the earliest periods these ideas have been most clearly expressed by Dr Lacey Wallace who envisages an initial settlement of continental immigrants east of the Walbrook around the forum with suburban areas to the west and in Southwark accommodate different communities and functions. It is notable for example that roundhouses, an indigenous Iron Age style of architecture, are found in these suburban areas but not east of the Walbrook.

Comparing the distribution of brooches and pots

The next few slides are plots produced using a statistical technique known as Correspondence Analysis to examine pottery and brooch assemblages from these areas (Cool and Baxter 2002). You do not need to understand the mathematics to interpret these just that they are a way of exploring the variation in a table of data after you’ve counted up how many pots or brooches of a give type you have from each site. Sites will plot closely together if they produce similar types of pots or brooches and further apart if there is little overlap. Similarly brooch or pot types will appear close to one another if they appear more frequently together on the same sites and further apart if they are rarely found together.

This first plot looks at the makeup of pottery assemblages in broad terms based on a functional interpretation of the vessels. This is a fairly simplistic form of analysis but demonstrates the high potential of this approach. What is quite remarkable is how clear the differences between the different areas of the city really are with only a little overlap between the red dots and orange dots which represent urban core assemblages from east of the Walbrook stream around the forum and Fenchurch Street, the light blue dots which represent assemblages from the western area, including those sites with roundhouses and the purple dots which represent sites in the Middle Walbrook valley. It is clear that there are significant differences in how pottery is being used in these different areas which may relate to different types of activities particularly variation in preferred cuisine or dinning traditions.

The eastern area is characterised by the high amphora content. In part this may reflect a greater consumption of imported foods and drinks such as dates, wine, olives and fish sauce all of which were transported in amphorae. However, the main factor is probably the proximity of the busy waterfront and the forum market where many commodities brought to London in amphorae will have been sold. The lesser amount of amphora found elsewhere might indicate that the contents of these transport vessels were decanted into smaller containers before being taken to areas further west within the city. These might be flagons and jars similar to we saw in the burnt store room at Gresham Street.

The Middle Walbrook sites produced rather less amphorae and are instead dominated by imported tablewares such as glossy red samian. If amphorae are excluded from the analysis, however, these two areas would look quite similar and the range of pottery tablewares in particular are also quite closely comparable to those found in contemporary contexts in Gaul and Germany. The range of imported food stuffs and pottery therefore supports the idea that these early communities were eating in a continental style.

Qualitatively the western part of the city is the most distinctive during this early period. It still receives some of the imported food stuffs and pottery found further east but overall the assemblages are quite different with more dependency on locally made coarse tempered wares and dark coloured reduced wares. The range of forms is also quite different and characterised by the proportion of jar forms especially on those sites with roundhouses. The overall impression is of pottery with some similarities to that which was in use in the pre-Roman Iron Age and later on in the Romano-British countryside. There is less emphasis on complicated sets of tablewares and more on cooking and eating from communal jar form vessels. More detailed work on the same sites is currently underway and more patterns are already emerging.

This analysis supports the kind of model put forward by Wallace and other scholars. There may have been quite significant differences between these areas and the communities who lived in them may have had quite different cultural backgrounds and lifestyles.

We can come at this problem from another angle by returning to the issue of brooches. We have data for fewer of these so I have lumped site assemblages together to form four broad areas. These correspond to those already discussed but also include Southwark. Here we notice that the key native tradition brooch types in use in south-east Britain during the mid 1st century cluster together in dark blue in the top left corner while continental forms form a broad red swathe across the right side. It should be remember that as with the pottery these are not binary divisions and that most types can be found in most areas of the city but the results show a clear trend, broadly suggesting suggest that ‘British types’ are more popular in the western and southern suburbs while continental types are best represented around the Forum and Walbrook areas.

We must remain cautious about oversimplifying things here. Objects were not always used where we find them they may have been dumped as rubbish from elsewhere. These techniques are also quite sensitive to chronological variation as London expands and areas are likely to have changed in character over time. This next distribution map is a work in progress but shows a rather different picture for the later Roman period when brooches are much rarer and seem to have been associated to a greater or lesser degree with military communities or with official modes of dress (see also Eckardt 2005). 2nd – 3rd century knee brooches are absent from the Forum area and instead appear in Southwark and in the west where there was a 2nd century fort.

Fourth century crossbow brooches and elaborate belt sets worn by officers and officials seem to be concentrated north of the Thames but are quite widely distributed crossing the Walbrook. The late Roman archaeology of London is sometimes ephemeral and has been badly truncated in places and so the distribution here, focussing on extra-mural burials and on deep untruncated areas next to the Walbrook and Thames, is unlikely to be completely representative of where these objects were used. The exclusion of Southwark is interesting however.

Investigating industrial activity

In these final slides I want to talk about economic activities and long term shifts in industry within Roman London. In this first slide I have used MOLA data to divide finds of crucibles used for melting metal into those found in early and late Roman deposits. Hopefully, you will agree that the distribution is quite different. In the early period there are quite a few large assemblages perhaps suggesting a widely distributed model with many workshops. For the late period there seems to be a shift towards centralisation around north-west Southwark and a major monograph on industry there has been published by MOLA (Hammer 2003). This pattern is even stronger if we remember that some of the smaller dots during this period match those in the first map and thus could be redeposited traces of early Roman industry. What does this shift represent? Is it a deliberate attempt to centralise in order to control or improve efficiency of the industry perhaps organised by the craftsmen themselves or an external force such as the Roman army or administration? Is it a sign that dirty industrial activity is being expelled from a gentrifying settlement on the north bank of the Thames?

A much more sophisticated analysis of the glass working industry by John Shepherd and Angela Wardle (2009 and in prep) examines the distribution of glassworking waste such as glass moils which form on the end of iron blowing irons and cullet dumps, essentially Roman bottle banks, intended for recycling. They have demonstrated clear fine grained chronological patterns and argue that glassworkers, perhaps partially peripatetic, were always living on the margins of the city periodically having to move on and establish themselves in new areas. They discuss a number of hypotheses which I cannot explore in detail here but amongst which are the need to move find more appropriate premises that had convenient access to supplies, transport etc orley markets or circumstances in which previous premises became unavailable because of price increases or because others had laid claim to the area. Glassworking should not be considered in isolation and in some periods this industry seems to have deliberately been organised in a complementary fashion with the pottery industry.

This final slide relates to a different class of object, transport equipment, and plots the distribution of iron temporary horse shoes and cart rein rings. There are obvious problems with trying to interpret the static distribution of mobile objects and while these classes of evidence seem to somewhat relate to the major roads and areas that we might expect to be transport hubs such as the forum and port the patterns are not particularly clear within the city. What is clear, however, is a major concentration on the northern edge of the city where about 50% of the iron horseshoes (hipposandals) from the city have been found. A concentration here was already evident from 19th century finds in the Museum of London collection but has been further reinforced by modern excavations (Harward et al 2015) and they continue to appear in recent work in the area.

The significance of this concentration is less clear but might have something to do with the road which skirts the city here. Could this be a Roman ring road where traffic sought to avoid passing through the busy city centre? These slip on horse shoes might be more easily lost where this road passes through the muddy and marshy area of the Upper Walbrook valley or it might be that better preserved finds from the area have been more consistently recognised by archaeologists. However, this does not seem sufficient to explain the pattern and the area has also produced a concentration of articulated horse burials which might also suggest a wider focus of equine activity. Was there some kind of major staging or stabling post in this area where animals stopped short and goods and people continued into the city? Much of this evidence relates to the 2nd century precisely when the major fort nearby was active and when the pottery, glass and leather industries in this part of the valley were at their height.

These city wide issues cannot be studied in isolation and this has been a fairly superficial examination based on data repurposed or hurriedly brought together for this talk. However, I think that the patterns are robust and that there is significant potential for detailed examination of how Roman London functioned as a city based upon its artefacts. I have deliberately shied away from trying to put London into a broader provincial or Empire wide context. This is principally due to a lack of time and this is something that can and should be attempted. Work such as the Roman Essex project which compared London with Colchester and the rural hinterland show that this is a potentially fruitful area for study.

Conclusions and acknowledgements

It is not possible to wrap up a lecture like this into some kind of neat grand narrative but I hope I have convinced you that we can learn a lot about life in Roman London through its objects. Human lives are complicated as is the interpretation of archaeological evidence and most of the issues I have touched upon are much richer, more complex and more interesting than my brief treatment of them here has allowed for. You can find out more here in the Museum of London and in the many publications that deal with the Roman city some of which are available from the shop or from the MOLA website It is a very exciting time to be working on finds in London with major new excavations as well as some great student and volunteer projects working on the archives and collections. Those pieces of Roman lives are here for us to collect and to try and piece together and the picture if incomplete is becoming clearer day by day.

I am grateful to all my colleagues at MOLA, Museum of London and the archaeological archive (LAARC) for sharing all their enthusiasm and energy with me and for putting up with mine. None of the ideas I have mentioned here today are mine alone and some of them are not mine at all but any flaws in them or in how they are presented should be laid at my door. Almost all of the projects mentioned here are very much collaborative affairs in which everyone including the developers who fund the work, the field team who excavate the sites and those who analyse and archive the results all take part. In particular I would like to thank Bloomberg and Mitsui-Fudosan who are supporting ongoing work on the major sites at Bloomberg London and 8 – 10 Moorgate, Frank Pemberton for permission to refer to his unpublished work on brooches, Caroline McDonald and the Museum of London for permission to use some of their wonderful images and Gresham College for the opportunity to speak to you today. Finally, thank you for listening.


Barber, B, and Bowsher, D, 2000 The Eastern Cemetery of Roman London: Excavations 1983-1990, MoLAS monograph 4

Bateman, N. Cowan, C and Wroe-Brown, R. 2007 London’s Roman Amphitheatre, Guildhall Yard, City of London MoLAS Mongr 37

Bateman, N. 2008 Death, women, and the afterlife: some thoughts on a burial in Southwark in Clark, J., Cotton, J., Hall, J., Sherris, R. and Swain, H. (eds) Londinium and Beyond: Essays for Harvey Sheldon. CBA.

Bayley, J and Butcher, S 2004 Roman Brooches in Britain: A Technological and Typological Study based on the Richborough Collection Reports of the Research Committee of the Society of Antiquaries of London 68, Society of Antiquaries of London

Casey, P.J. and Hoffmann, B. 1995 'Excavations on the Corbridge Bypass, 1974' Archaeologia Aeliana 5th series, 23, 17-45

Casson, L. Drummond-Murray, J. and Francis, A 2014 Romano-British round houses to medieval parish. Excavations at 10 Gresham Street, City of London, 1999–2002, MOLA Monogr 67

Calvi, M.C. 2005 Le ambre romane di AquileiaPubblicazioni dell'Associazione nazionale per Aquileia 10

Cool, H.E.M. 1991 'Roman metal hairpins from southern Britain', Archaeological Journal 147, 148–82

Cool, H.E.M. and Baxter, M. 2002 'Exploring Romano-British finds assemblages', Oxford Journal of Archaeology 21.4, 365-80

Cool, H.E.M. and Philo, C. 1998Roman Castleford: excavations 1974–85: Vol 1, The small finds (eds H E M Cool and C Philo), Yorkshire Archaeol 4, 140–194, Wakefield

Croom, A, 2011 Running the Roman home, Stroud

Crummy, N, 1979 A chronology of bone pins' in Britannia 10, 157-64

Crummy, N. 2005 From bracelets to battle-honours: military armillae from the Roman conquest of Britain in N. Crummy (ed.) Image, Craft and the Classical World Monographies instrumentum 29, 93-106

Eckardt, H. 2014, Objects and Identities in Roman Britain and the North-Western Provinces. Oxford: OUP

Feugère, M. Les fibules en Gaule méridionale, de la conquête à la fin du Ve siècle ap. J.-C. (suppl. 12 à la Rev. Arch. Narb.), Paris: CNRS

Greep, S. 1983. Objects of animal bone, antler, ivory and teeth from Roman Britain. Unpublished PhD thesis: University of Cardiff.

Hall, J. & Wardle, A. 2005. Dedicated followers of fashion? Decorative bone hairpins from Roman London. In: N. Crummy (ed.) Image, craft and the classical world: essays in honour of Donald Bailey and Catherine Johns. Montagnac: éditions monique mergoil. 173-178.

Hammer, F. 2003 Industry in North West Roman Southwark MoLAs Monogr 17

Harward, C. Powers, N. and Watson, S. The Upper Walbrook Valley Cemetery Of Roman London: Excavations At Finsbury Circus, City Of London, 1987-2007 MOLA Monogr 69

Henig, M. 1984 A cache of Roman intaglios from Eastcheap, City of London Trans LAMAS 35, 11 – 15

Hill, J. and Rowsome, P. 2011: Roman London and the Walbrook stream crossing: excavations at 1 Poultry and vicinity, City of London. MOLA Monogr 37

Holder, N. (2007) ‘Mapping the Roman inscriptions of London', Britannia 38, 13–34

Holder, N., and Jamieson, D. (2003) ‘The prehistory of the City of London: myths and methodologies', Archaeological Journal 160, 23–43

Hunter, F.2012 Frontier finds, frontier art – views of enamelled vessels.In: D Breeze (ed) The first souvenirs: enamelled vessels from Hadrian’s Wall. Cumberland and Westmorland Antiquarian and Archaeological Society, Kendal, 85-106

Hunter, F. 2014 Stories from black bangles: jewellery and other finds of jet-like materials in Roman Scotland. In R. Collins and F. McIntosh (eds) Life in the Limes: Studies of the people and objects of the Roman frontiers. Oxbow Books, Oxford, 152-165

Mackinder, A. 2000 A Romano-British cemetery on Watling Street : excavations at 165 Great Dover Street, London,MoLAS Archaeology Studies Ser 4

Mackreth, D. 2011 Brooches in Late Iron Age and Roman Britain Oxbow

Marshall, M. and Thorp, A. 2014 Three Roman vessels from Old Jewry / Cheapside, London, EC2 London Archaeologist 14.3

Merrifield, R. 1987 The Archaeology of Ritual and Magic London

Monteil, G. 2008 The distribution and use of samian inkwells. In Clark, J., Cotton, J., Hall, J., Sherris, R. and Swain, H. (eds) Londinium and Beyond: Essays for Harvey Sheldon. CBA.

Pearce, J. 2004 Archaeology, writing tablets and literacy in Roman Britain .Gallia 61, 43–51

Plouviez, J. (2008). Counting Brooches: Ways of Examining Roman BroochAssemblages in London and Beyond. In J. Clark, J. Cotton, J. Hall, R. Sherris, and H. Swain (eds.), Londinium and Beyond: Essays for Harvey Sheldon. CBA.

Rangel de Lima, C. 2014 A Girl's Best Friend: the role of hairpins in defining female identity in Roman LondonUnpublished MA dissertaion UCL

Riha, E. 1979 Die römischen Fibeln aus Augst und Kaiseraugst Forschungen in Augst 3

Shepherd, J. and Wardle, A. 2009 The glassworkers of Roman London MOLA

Stephens, J. 2008. Ancient Roman hairdressing: on (hair)pins and needles. Journal of Roman Archaeology 21. 111-132.

Tomlin, R. 2003 The girl in question: A new text from Roman London Britannia 34, 41 – 51

Wallace, L. 2014 The origins of Roman London Cambridge University Press

© Michael Marshall, 2015

This event was on Mon, 05 Oct 2015


Michael Marshall

Michael is Senior Finds Specialist (Prehistoric and Roman) at Museum of London Archaeology.

Find out more

Support Gresham

Gresham College has offered an outstanding education to the public free of charge for over 400 years. Today, Gresham plays an important role in fostering a love of learning and a greater understanding of ourselves and the world around us. Your donation will help to widen our reach and to broaden our audience, allowing more people to benefit from a high-quality education from some of the brightest minds.