Building on the VASLE project, we should now be able to use its methods with the far more extensive data of today’s PAS database to understand ‘productive sites’ better. VASLE analysed the relative proportions of artefacts and coins at different sites to produce graphs that could allow identification of different site types, or could suggest whether differences between sites should be explained in chronological, cultural, or functional terms. Local small-scale studies (e.g. those in a single eastern district, or a south-western county) could now re-run these analyses, and eventually combine to compare different areas in a consistent way.
Although Iron Age and early-medieval studies are strongly regionalised, Roman Britain tends to be seen as a continuum. Is this a real picture, or could we find evidence of regional variation to link the Iron Age to the early Anglo-Saxon period? The regional nature of brooches is well known, and Eckardt and Crummy (2003) have explored regionality in nail-cleaners, but other artefacts to do with personal adornment might be expected to show regional variation, and these might include bracelets, finger rings, cosmetic sets and mirrors. Undergraduate dissertations could choose a single artefact type and compare the dataset in two counties.
Purses with metal bars, loops and frames were first collected and classified by Ward Perkins in 1940. Before his untimely death, David Williams was working on a new classification and chronology, published in a FRG Datasheet (Williams 2018). This work now needs to be taken on by another researcher. Questions might include whether particular types cluster in particular geographical areas, and whether some places start or stop using purses at particular times. Linking PAS data to excavated examples, both in Britain and on the Continent, might allow us to refine the chronology and to raise questions of the use of these objects. Were they all really purses (for carrying money), or might some have been used for other purposes such as carrying tinder, or small trinkets such as hair fasteners or dice? In addition, many Class A2 bars bear inscriptions and/or images. Is there patterning within these, perhaps indicating workshop groups or regional fashions?
Keith Parfitt (Canterbury Archaeological Trust) and Andrew Richardson (former Kent FLO) noted many years ago that the proportions of brooch types from grave assemblages in Kent are different to those from metal-detected assemblages. There are various possible reasons for this (see McLean and Richardson in Worrell et al (ed) 2010), but one might be that the excavated evidence comes from graves – the costume of the dead – and that PAS data includes a substantial element of casual loss from the costume of the living. McLean and Richardson 2010 looked at brooch types from southern counties of England, but there is scope at undergraduate or master's level to look at eastern and midland counties too. An undergraduate dissertation could look at one county, a master's at two or more. To establish differences between funerary and living costume would revolutionise early Anglo-Saxon studies, which are traditionally heavily costume-based.
John Hines drew attention to a lead model of a great square-headed brooch found in Geneva many years ago (Hines 1997). Since then, the PAS has recorded 26 early Anglo-Saxon lead brooches, in addition to three lead wrist-clasps and a lead girdle-hanger (as of March 2019). Kevin Leahy (Finds Adviser) has pointed out that lead brooches do not appear to have been found in graves. Is their distribution (Anglian areas only at present) meaningful? Do they contain enough information to work out if they are cheap copies of brooches intended as jewellery, or just models for casting copper-alloy versions? Or can they be both?
Industrial workshops are not obvious on early Anglo-Saxon settlements, and there has been little work on how craft production was organised. A first step might be to look at the punchmarks on copper-alloy objects and arrange them into groups. A later objective would be to find die-links between the same objects in different places, or different objects in the same place.
It has long been thought that Anglo-Saxon and Scandinavian brooches differ in the orientation of their pin fixings. This idea now needs testing, as casual observation suggests it isn’t that simple. This study would define a set of pin and catch attributes and match them with different brooch types, sizes, shapes, art styles and so on.
A recent PhD on excavated urban buckle assemblages (Cassels 2013) has emphasised how different these are from our rural, metal-detected finds. A useful extension of Cassels's study would analyse the buckles from medieval rural settlement excavations and compare these to PAS data. At the very least, this would establish date-ranges and allow the three groups to be compared. At present, even the most complex and peculiar medieval buckles on the PAS database can rarely be dated to within a couple of centuries, and this is hampering our understanding of the entire period. Undergraduate dissertations could pick a single county for analysis; a PhD could look at the whole country. A related task would be to systematically research the pictorial representations of buckles on the (usually very well dated) medieval effigies and brasses.
We now have over 500 Boy Bishop tokens on the PAS database, over 400 of them from Suffolk. Yet the popular use of Boy Bishop tokens is little understood. Preliminary work on their landscape contexts appears to show that they cluster around churches and manorial sites. What factors underlie this pattern? Can the distribution help us understand what the tokens were used for? Another project using Boy Bishop tokens would be to look in detail at the non-Suffolk finds. Are these local issues? Or are there any patterns in the data that might suggest how (and when) these objects moved as far as Somerset and North Yorkshire?
Single bullets or musket balls usually cannot be dated, but the overall characteristics of large groups can usually suggest a date. Checking the calibres of shot found and recorded over years within single fields could reveal unrecorded Civil War skirmish sites, or later target-practice sites. This would perhaps best be done at parish level to begin with; a useful comparative case study would be Easton Maudit, in Foard 2008.
In theory, the discovery of graphite deposits in Cumbria in the late 16th century rendered the lead pencil obsolete. But the PAS database has recorded hundreds of mysterious items, apparently writing leads or pencils, that look late post-medieval. They are not apparently found on archaeological sites in the UK, but they do turn up in North America. This project offers a workout for all the skills of the historical archaeologist – the archaeology of American colonial sites, together with research into sales catalogues and stationers’ or building merchants’ inventories.
There are an increasing number of Iron Age objects on the PAS database in the shape of, or decorated with, human heads or faces. They include harness fittings, sword pommels and vessel mounts. It would be worth collecting them up and asking several questions. What kinds of objects attract this decoration? Can the faces be characterised, to allow a distinction between an Iron Age and an early Anglo-Saxon way of depicting the human head (building on Lisa Brundle’s PhD for the Anglo-Saxon examples)? Do they vary across the regions of Britain? And what is their relationship to depictions of the human head in Roman art? A simple catalogue raisonné would make a good undergraduate dissertation, but extended to include the wider picture of head cults in the Iron Age (following, for example, Armit 2012) it could make a master's thesis.
The uses of papal bullae are known in theory (from the viewpoint of the documentary historian) but the process of their loss and deposition is mysterious. Can we learn anything from analysis of their findspots on a detailed landscape scale?
Links between flint objects and early metal objects could profitably be explored using metal skeuomorphs of flint objects. There are several arrowheads with noticeably ‘lithic' tendencies on the PAS database, and daggers could be brought into the argument too.
There are an immense number of projects that could be done on these objects, which are only just starting to be used for research on medieval literacy (there is a current Leicester PhD scoping out the potential). The PAS database currently (April 2019) has close to 6000 examples, and adds about 300 per year. PAS seal matrices are also different to those that have survived in museum collections, or the seals on manuscripts; the vast majority are very ordinary everyday personal non-heraldic seal matrices. Women’s use of seals is an obvious topic, but there are plenty of other ideas, particularly for an interdisciplinary historical archaeology or medieval studies MA. Pointed-oval seals are more equally divided between men and women than the more common circular seals; is there something interesting about the men (e.g. are they clerics)? What links are there between central motif and the identity of the owner (the answer could be ‘none’)? Do seals show any form of regionalism – for example, do the East Anglian seals show differences which could be interpreted as the result of the region having a larger number of free landholders? Can we use seal matrices to trace the development of various types of byname and surname? For seals with mottoes, we need a scoping survey; which are the commonest, how do the mottoes relate to central motif, and (for historians) how were they used?
Using GIS, coastal finds can be studied on their own. Is there a difference in artefacts between coastal and inland 'productive' sites? Is trading happening at beachmarkets, or is there evidence of control of trade, with the goods being transported to royal palaces or major monasteries? This subject could include small-scale (county) studies at masters level, as well as a larger (perhaps) nationwide overview for a PhD.
Until a recent PhD (Weetch 2013), ansate brooches were thought to be Frankish imports. Weetch 2013 has now given us the ability to classify ansate brooches, and we are slowly adding Weetch types to the records. A student with good languages should be able to compare French and German sources to PAS examples to work out the relationship between Continental and English ansates. At first sight, they appear to differ both in date and in style. Are the links and differences related to trade, or politics, or both?
John Hines’s monumental corpus of these early Anglo-Saxon brooches was published in 1997. Since then, the PAS has recorded fragments of over two hundred more examples. How do Hines’s conclusions work when tested against the PAS examples?
Some years ago it was noticed (by Karen Hoilund Nielsen) that early Anglo-Saxon art styles were rare on some objects at some times. More recently, Jane Kershaw has also pointed out that brooches and strap-ends differ a great deal in the art styles that are chosen for decoration (Winchester style, for example, is rare on brooches but very common on strap-ends). This work is still very much in its early stages, and we need to establish the broad outlines of which art style is found on what objects before going on to look at which factors might underlie the choices. This is really two pieces of work, one for the early Anglo-Saxon and one for the later Anglo-Saxon and Anglo-Scandinavian material.
Spiral finger-rings are found in the Bronze Age, Iron Age, Roman and Anglo-Saxon periods, apparently with very little change in their form over this period. Most case studies of archaeological objects focus on why designs and uses change; this object type appears to have remained unchanged over three thousand years. Are we correct in thinking this, and why should it be so?
We have an odd group of perhaps two dozen equal-arm brooches in south-east England which seem to date from c. 450-550. Their only formal parallel is Thӧrle’s Type III, found in later contexts (c. 560-640) and a long way away (southern Switzerland, northern Italy and the Istrian peninsula). These English brooches rarely appear in graves, and so have attracted little notice hitherto. The big questions for this research are, firstly are they related to the Continental Thӧrle Type III, and if so, which way did the influences travel and how? How do they fit in to the rest of the early Anglo-Saxon brooch repertoire; what motivations went into choosing them as a brooch? Working with people studying excavated vs. detected brooch types in specific counties (see Costume in Life and Death), we could review the statistics of brooch use to work out whether the long equal arms really are under-represented in graves, or whether the sheer volume of finds being recorded by the PAS means that eventually even types as rare as these would be noticed. It is also be essential to review the method of fixing and fastening the pin, as this varies from one area of Europe to another and changes over time.
Above it was suggested that an undergraduate dissertation might compare types of Roman object found in two counties which are notably different in their Iron Age and/or early-medieval material culture, to see if the apparent gap in regionality can be bridged. A masters thesis might look at several artefact types, or a larger geographical area.
Mary Chester-Kadwell's 2009 PhD established a method for studying early Anglo-Saxon cemetery locations in the landscape – how they are sited in the existing (Romano-British) landscape, how they relate to natural features, contemporary settlements, and later land-use. Now that the method has been established, it should be reasonably simple to apply it to other counties.
A comparison of the spatial distribution of objects with late Anglo-Saxon and Scandinavian art styles on them could discover how much the Danelaw boundary was reflected in material culture. A larger international study could profitably add in Danish, Swedish and Norwegian finds to establish just how Scandinavian the Danelaw was – or how English. Relating this to lingustics, for example the study of place-names, would add depth.
Philippa Walton’s PhD (Walton 2011) analysed PAS data on Roman coins by both date and denomination. She carried out case studies which found regional differences in both these aspects, raising interesting implications for the organisation, monetisation and Romanisation of different areas of Britain. As her data collection ended in 2008, the vastly enlarged dataset could now be used for several masters' dissertations. Firstly, Walton suggests that parishes with more than 500 Roman coins should be studied in detail to reveal their status in the Roman period, and offers a hypothesis that there may be a connection with religious or votive sites; this could now be tested with the far greater number of parishes that pass this threshold. Secondly, her regional analyses could be extended, particularly to East Yorkshire where the dataset has more than doubled since 2008. Thirdly, her period-specific case studies could be augmented by adding in second-century coins.
Around 3,000 lead tokens have been recorded on the PAS database as being of medieval date. An informal typology for these exists (Powell 2010). Do PAS finds fit well into this typology? Can they improve it? It has been suggested (for example here) that these tokens may have some link with pilgrimage, or monasteries; do the findspots support this hypothesis? How do the landscape contexts compare with other finds, such as pilgrim badges and ampullae? A larger study might combine the medieval and post-medieval lead tokens, given the difficulty in dating many of them.
Nearly 60 moustache-shaped objects have now been recorded on the PAS database. They are dated from similar objects found in Bronze Age hoards, but are sometimes thought on stylistic grounds to be Iron Age. Until more are found in context (e.g. in hoards or on excavated sites) it will be difficult to assign a definite date and function, but there is now a large enough PAS dataset to investigate some patterns, including different types and geographic patterns.
Working from Plouviez 2008 (in Londinium and Beyond), can we use PAS data to establish a 'normal' pattern of loss over time, in the same way that Richard Reece and Sam Moorhead have established for Roman coins? Due to the amount of data, and the historic lack of an accepted nomenclature for Roman brooches (only partly resolved by Mackreth 2011) this topic may be a challenge.
Papal bullae are the seals from papal documents; there are currently (March 2019) just over 500 on the PAS database. Some will have been attached to land grants, but many will be from indulgences. The doctrinal history of indulgences is well understood, but their actual use within popular religion has been hard to study. Recent small-scale research (Gilchrist and Sloane 2005) has concluded that most papal bullae in England date from the mid to late 14th century, coinciding with the Black Death. We now need a bigger study to establish the relationship between the deposition of papal bullae and the details of both institutional and popular religion? In common with many other medieval topics, this is highly interdisciplinary, with links to theology and history.
Despite many records, PAS lithics data has not yet been used for research to the same extent as metalwork and coins. Is this to do with the nature of the evidence, or is it due to the way in which the PAS records flint and stone objects? Looking at a sample of PAS lithics records could not only inform and improve the way in which the Scheme records lithics, but could bring these records to a wider audience and suggest new research topics.
How do PAS records of late Roman and very early Anglo-Saxon artefact types (e.g. clipped siliquae and Quoit Brooch Style objects, late Roman buckles and strap-ends, very early Anglo-Saxon brooches) relate geographically and chronologically to sites known from excavation to have been used in the late Roman and early Anglo-Saxon periods? In addition, can PAS data help to explain what is happening in the fifth century in the west of Britain? Due to the differences between the east and the west of the country, this topic is probably best explored on a regional rather than a national scale.
The PAS database has recorded nearly 40 examples of a specific form of radiate-headed brooch which appears to be confined to England. Like the long equal-arm brooches, they are rarely found in graves, and their relationship to Continental brooches is little understood. Radiate-headed brooches are not easy to study, as recent research is scattered and in a variety of languages, but they are crucial to questions of the relationship between England and Europe in the 5th and 6th centuries AD.
Mapping the findspots of Bronze Age hoards using GIS to compare them with landscape features, e.g. slope, elevation, soil type, watercourse proximity, previous landscape actively etc would help to shed light on hoarding behaviour. This has been done for north-east England (Poyer 2015, PhD from University of Sheffield) but could profitably be tested in other areas or nationally.
It has been noticed by several studies recently that, in some areas, later Roman coins are rare or even absent. This has led to suggestions that the intensity of Roman rural settlement might decline in some areas well before the end of the Roman period. The phenomenon has been noticed by Phil Harding in Leicestershire, Jude Plouviez in coastal Suffolk, and by the Roman Rural Settlement project. Philippa Walton has suggested that it is a widespread rural phenomenon, representing a contraction of coin use to urban centres. These hypotheses all need testing. What do the changes in coin use mean for our ideas of how Roman Britain became Anglo-Saxon England?
Recent work by Carenza Lewis and Paul Blinkhorn has been able to show the effects of the Black Death on pottery concentrations in England, thanks to the excellent chronology of 14th-century English pottery. It is possible that English metalwork may show the same effect; the PAS database has some material that we can date to the 13th century, then a massive amount of material that can be dated to the 14th century, then much less that is datable to the 15th century. If we had better resolution within 14th-century metalwork, could we see the Black Death in metalwork as well as pottery? This would be an excellent opportunity to integrate numismatic data with non-numismatic objects.
In Lincolnshire, the monasteries of the Witham valley seem to be just the final link in a chain of ritual significance (evidenced by the deposition of objects) that reaches back to the Bronze Age. Work is now starting at Strata Florida in mid Wales to see if this pattern is replicated elsewhere. Is the pattern of prehistoric object loss different in the vicinity of medieval monastic landscapes? What is the relationship between monastery sites and earlier sacred landscapes?
It is often suggested that lost roads trackways could be re-discovered by meticulously plotting linear scatters of finds. Roman roads are thought to be particularly amenable to this approach, due to the large amounts of easily recognisable material culture from this period, particularly coins. But little structured research has actually taken place to confirm the validity of this approach. An undergraduate dissertation could take a known (but disused) Roman road and look at the finds signature along its length; a PhD could look at the entire concept of movement around the countryside and how this might be visible through portable antiquities. Active researchers in the field include Martin Bell at Reading and several at the University of East Anglia, for example James Albone and Sarah Harrison.
Over 6,000 lead tokens of post-medieval date have been recorded on the PAS database. What patterns can be found within this enormous quantity, and can they help us narrow down date and function? There have been plenty of suggestions as to use, but all of these need testing, and PAS data is ideal for this. A large study might combine the medieval and post-medieval lead tokens, given the difficulty in dating many of them.
Now that we have a usable typology for late Iron Age and Roman brooches (Mackreth 2011), we can combine this with PAS data to work out the geographical distributions of different brooch types. Do the distributions of the earliest brooches coincide with Iron Age tribal boundaries, and if not why not? Do the later brooches have any meaningful geographical variation, and if so how does this relate to administrative boundaries? And lastly, do the latest brooches show any relationship to the later Anglo-Saxon cultural divisions? It should be noted that because the PAS has not yet produced specific guidance on how to record brooches, the dataset will need a great deal of cleaning.