Prosopography of Anglo-Saxon England

Anglo-Saxon historians are in an enviable position when it comes to electronic resources. We already have a host of helpful websites at our fingertips: the Electronic Sawyer (http://www.esawyer.org.uk/), Kemble (http://www.kemble.asnc.cam.ac.uk/), ASChart (http://www.aschart.kcl.ac.uk/) and Langscape (http://www.langscape.org.uk/) for charters; the Fontes Anglo-Saxonici Database (http://fontes.english.ox.ac.uk/) for the Latin sources of Anglo-Saxon literature (both Latin and vernacular); the Corpus of Early Medieval Coins for numismatics (http://www.fitzmuseum.cam.ac.uk/coins/emc/); and the Portable Antiquities Database for artefact finds (including coins) (http://finds.org.uk/database). It might, therefore, be doubted whether there is much room for a new resource to join these. Nevertheless, for a number of reasons the Prosopography of Anglo-Saxon England (affectionately known as PASE: http://www.pase.ac.uk/) is to be welcomed. It provides a powerful new tool which adds substantially to the previous resources available.

As the PASE database informs the reader, it is the product of two successive projects funded by the Arts and Humanities Research Board (and latterly Council): PASE 1, which ran from 2000 to 2004; and PASE 2, which ran from 2005 to 2008, with extensions up to 2010. The project team changed a number of times over the years, though two of its directors, Janet L. Nelson and Simon Keynes, remained constant, offering a degree of continuity. PASE 1 was launched first and the joint material from both projects was made available in 23 December 2009, though further resources were added later on 18 August 2010.

According to the site, the aim of PASE is ‘to record information about Anglo-Saxon persons, or persons important in the Anglo-Saxon world, as presented in contemporary source materials’ (http://www.pase.ac.uk/about/methodology/index.html). The resulting resource is designed to replace Searle’s Onomasticon Anglo-Saxonicum, which for generations had been the standard prosopographical tool, but whose shortcomings were many and well known.(1) Indeed, the advent of modern computer technology has been a great boon for projects such as this, offering the potential to create a more flexible and user-friendly tool than was possible in the age of print. Thus a book such as Searle’s Onomasticon, even if thoroughly revised and up-dated, can only be searched in a limited number of ways: the reader ultimately must know the name he or she is looking for and has little way of searching by geography (people attested in certain regions), gender, possession, and so on. However, all of these possibilities – and many more – are afforded by the PASE database and this is what truly makes it unique: the same data are accessible from numerous different angles. Moreover, it has been possible to aim for a degree of comprehensiveness previously unthinkable: without the limitations of printing costs there have been effectively no limits – save those imposed by time and manpower – on the amount of information included. A quick glance at the statistics accompanying the database is most impressive in this respect: over 19,000 individuals have been recorded, with over 180,000 different ‘factoids’ about them.

The database itself is set out in a rather chic-looking site with information about the project provided under ‘About PASE’. This includes not only background on the course and development of the project, but also an introduction to prosopography, an explanation of the project’s aims, and much more besides. Of particular interest is the section on ‘Data gathering’, where the methods employed by the team in collecting and rationalising data are set out. This is very informative on a technical level, explaining the forms used for data capture and the process of entering this material into the database, but somewhat less so when it comes to the approach taken by the researchers. Thus, for example, the job of ‘merging’ profiles – that is, of going through all the profiles and deciding whether two separately attested individuals bearing the same name are in fact same person – is described in rather lapidary fashion: ‘The merging of names was a process that could not be automated because it demanded a knowledge of the sources and scholarly judgment. During PASE 1, John Bradley provided a technical tool for doing the merging and the task of merging named persons was undertaken by Alex Burghart and Francesca Tinti during PASE 1, and by Ben Snook during PASE 2’. While this tells us who performed the task, it does not provide any information about how they went about it, or about the criteria employed in making such decisions. Indeed, given the fact that much – arguably most – prosopographical research in this period involves trying to answer the tricky question of whether two separately attested individuals are in fact the same person, one whould like to know rather here: has the team been bold, making as many identifications as possible, or has it been conservative, consciously leaving this task to the database’s users? My own impression – gained implicitly by exploring the database – is that the latter approach has generally prevailed, and there is much to commend this.

The database itself is at a first glance slightly intimidating: the user is faced by a series of search bars on the left-hand side, offering the option of searching by text, person, source, possession, location and date. One must also make an executive decision as to whether the search should include ‘coin data’ and ‘Domesday data’. However, while all of this may seem a bit daunting, particularly for the uninitiated, a detailed description of the search options is provided on the right-hand side. It is here that the rationale behind the layout and main search options available is explained, drawing attention to the fact that many of the individuals listed in coin and Domesday data are known only through one ‘factoid’ (generally that the figure minted coins or owned land). The advantage of being able to filter this material out is therefore clear: it might otherwise overwhelm a search, making it tougher to get at the data one really wants. There are ultimately two kinds of search, each of which can be pursued via further search options: one can search by ‘person’, in which case an individual associated with a specific source, possession etc. will be displayed; or one can search by ‘factoid’, bringing up details of the pieces of information that associate a specific source, possession etc. with a person (or persons) in the database.

The navigation of the site works well and, though it is perhaps not quite as user-friendly as one might desire, one rapidly becomes accustomed to it. Indeed, although I found myself confused at a few points – and it should be said that I already had a passing acquaintance with the database, as well as its PASE 1 predecessor –, I found that navigation became easier as I became more accustomed to how the database ‘works’. Indeed, though the interface is not always the most intuitive, its strength lies in the way that it guides the user through the process of searching, providing further information and explanations (visible on the right-hand side) throughout. This means that when something is unclear, guidance is always at hand. The actual mechanisms of searching work well and are largely self-explanatory: if one clicks on a specific search option the relevant data are then listed alphabetically and the user can single out the information he or she desires to view in greater detail. The ability to add or remove coin and Domesday data with a single click is very useful here and the default setting seems to be that successive searches are cumulative. Thus, for example, if one searches for the name ‘Edward’ and comes up with the relevant (50) entries for the name, any further searches will be run within this data set. This is a useful feature, especially when searching for people falling into specific categories (e.g. women attested in London). Moreover, at any point the user can quite literally press the reset button (or in this case the ‘reset constraints’ button), clearing all previous data.

Although it is difficult to compare the new interface with that of its predecessor, PASE 1, since this is no-longer available, the present version seems to be more clearly laid-out, better explained and – perhaps most importantly – more flexible. Indeed, a number of the search options, such as searching by location, are new, allowing users to access data from even more angles. This means, for example, that I was swiftly able to come up with a number of figures associated with the possession of weapons (ten individuals). However, not all searches work so seamlessly. When I tested the location search, by looking up Lydford, Devon, the database brought up six individuals associated with the site, ranging from ‘anonymi 2204’ to ‘William 1’ (i.e. William the Conqueror). Yet when I sought to discover what connected one individual, namely ‘Æthelred 32’ (i.e. King Æthelred ‘the Unready’), with this place, I was left baffled; all that clicking on ‘Æthelred 32’ did was bring up his entire (extremely detailed) profile, with no indication of where Lydford was to be found within it. Given that some 566 ‘events’ and even more overall ‘factoids’ are associated with King Æthelred, it is hardly practicable to scroll through all of this in order to find out where Lydford might be hidden amongst it. Of course, the canny user would at this point change from searching for ‘persons’ (as I had been) to ‘factoids’ and come up with the information that coins were struck in Æthelred’s name at Lydford, but this line of action was by no means self-evident and it took me a few moments to realise that this was the solution to my problem. Still, it would be wrong to make too much of such slight difficulties; these are the sorts of lesson the user need only learn once, and the database becomes easier to use the longer one works with it. Indeed, it is far more impressive that such details can be singled out in this fashion, than that they at times take a few minutes to isolate.

The overall accuracy of the database is naturally difficult to judge, since any single user can only scratch the surface of it. However, having completed a number of more or less random test searches, focusing on periods and individuals I am well acquainted with, I am satisfied that it is truly comprehensive and that omissions or mistakes – which doubtless will be present in a project of this scale – are few and far between. It was by performing such searches that I began to get a sense for the approach taken in merging profiles. As mentioned above, the policy seems to have been conservative: where two or more figures with the same name appear in documents of a similar era they are generally listed as separate individuals, unless there are good reasons for associating them. Nevertheless, helpful cross-references are often provided, pointing out where an individual may be the same as one or more others in the database. Thus, for example, in the entry for ‘Æthelgeard 2’, a mid 10th–century thegn who is known from charter attestations, the user is informed that he is likely the same individual as ‘Æthelgeard 3’, who appears in the New Minster Liber vitae. A different case, however, is posed by the ealdorman known as Æthelstan ‘Rota’. The career of this individual is hard to trace: an Æthelstan attests charters as ealdormen in Eadred’s reign, then disappears in 948, though in 955 an Æthelstan appears as an ealdorman at Eadwig’s court, carrying the by-name ‘Rota’. The difficulty here lies in deciding whether or not these two are the same individual and it is interesting to note is that despite the significant time-lag between their attestations these are all associated with one individual, ‘Æthelstan 31’, in PASE. This is in many ways a more daring identification than that of ‘Æthelgeard 2’ with ‘Æthelgeard 3’ and the impression, therefore, is that the policy followed at the stage of merging profiles may not have been as consistent as one might have desired. Indeed, in this case the choice to identify all of these attestations with the same figure is particularly surprising, since present scholarly orthodoxy holds that they in fact belong to two different ealdormen.(2) It is moments such as these which make the user want to know more about the rationale at work at the merging stage; it may be that PASE is correct here, but no justification or explanation is given. This also highlights a potential danger of the database: all of the identifications made within it are presented as fact and a user not already acquainted with the period would have little reason to query the database on a point such as this. However, this is but one case and it would be unfair to impugn the database on it alone. Indeed, in general a justifiably conservative policy seems to have prevailed – one which leaves more daring identifications up to the user.

Perhaps the most impressive part of the present PASE database – which was not available in earlier versions – is the Domesday material, which can be accessed by a tab at the top of the screen. This allows users to access prosopographical data from Domesday Book in tabular, cartographic or statistical form, making a wealth of information available in an extremely helpful manner. The ability to map Domesday data is to be particularly praised, as it allows one to envisage individuals and estates both on a map and in more statistical form. In fact, it might have been helpful to have the option of mapping relevant data within the main database (such as, for example, the estates mentioned in wills), though presumably both time and funding limited the team in this respect. In any case, this sophisticated tool will make PASE the first place to go for all future prosopographical work using Domesday Book.

Despite the occasional criticism, the ultimate verdict on PASE must be positive: it is a huge improvement on what was previously available and if at times it leaves the user wanting to know more, wanting to be able to search more material in more different ways, then it is to be all the more welcomed for this. It is designed as a research tool and as such does not mark the end of the study of Anglo-Saxon prosopography, but rather a bright new beginning. It is particularly important in this respect that PASE is open to the general public: anyone with an internet connection (or with access to one) can use the database and this is therefore certainly not an ‘Ivory Tower’ resource. Once again Anglo-Saxon studies has set a benchmark for its counterparts when it comes to electronic resources and it is to be sincerely hoped that similar projects will be devised to tackle the problems of Carolingian, Ottonian, Capetian and Anglo-Norman prosopography. PASE marks a milestone in the study of Anglo-Saxon England and will become the electronic vade mecum of a new generation of scholars.

Notes

The editors intend to respond collectively in due course.

1. W. G. Searle, Onomasticon Anglo-Saxonicum: A List of Anglo-Saxon Proper Names from the Time of Beda to that of King John (Cambridge, 1897).Back to (1)
2. See, for example, A. Williams, ‘Princeps Merciorum gentis: the family, career and connections of Ælfhere, Ealdorman of Mercia, 956–83’, Anglo-Saxon England, 10 (1982), 143–72, 146–7; and S. Keynes, ‘Edgar, rex admirabilis’, in Edgar, King of the English 959–75: New Interpretations, ed. D. Scragg (Woodbridge, 2008)Back to (2)