Anglo-Saxons on the PAS: 2013 in Review (Part I)

As 2013 breathed its last, I did a bit of data gathering on the Portable Antiquities Scheme database.  According to my reckoning nearly 45,000 items were logged on the publicly available database over the course of 2013.  Some of them were spectacular; most far less so.  But all of them taken together are gradually (though perhaps not that gradually) transforming our view of the past.

With so many finds constantly streaming into the database, it’s difficult to get a grip on its cumulative contribution.  An annual basic exploration of a narrowly defined period would therefore have some value.  And that’s exactly what I want to provide here: a summary of what kinds of objects the PAS recorded in 2013 from the early Anglo-Saxon period (here defined as most of the 5th and 6th centuries, also known as the Migration Period).  My hope is that this dataset should save other commentators some time, permitting some very general observations without the need to dredge through the raw data.  I’m only going to present the first part of this survey here, but will return to the dataset in future blogs, looking at some of these finds in more detail.

Excellent summaries are of course provided elsewhere.  The Portable Antiquities Scheme publishes annual treasure reports and Medieval Archaeology also publishes an annual summary of important PAS finds.  However, both these publications focus on the most impressive or unusual items, ignoring the towering mass of mundane finds.

Retrieving this kind of data from the PAS database is not as straightforward as might be imagined.  Because the scheme does not record finds with the specificity required to compile such a dataset, a search for finds from the “early” + “Early Medieval Period” from 2013 does yield anything close to a comprehensive list (nor, for that matter does searching for finds ascribed “Anglo-Saxon”).  Hence, the dataset you see here was compiled by combing through every find recorded as “Early Medieval”, using my own knowledge to extract those that I knew could be assigned more specific dates.  I’ve tried to be comprehensive, but doubtlessly a handful of finds escaped my attempts altogether.

Hence, there are caveats to this dataset.  Primarily, it is limited to what was publicly available on the 31st December 2013, without any privileged login to the database.  The dataset does not therefore represent everything that was recorded in 2013 as considerable numbers of finds are not yet published.  This should not be taken as an authoritative or official total because the actual figures seem to increase day by day.  Rather, regard this as a snapshot of the database from that point in time.

Without further ado then, my sample of finds recorded in 2013 included 44,690 objects, of which 2,017 were early medieval (AD 410-1066). Of these early medieval finds, only 319 belonged to the early Anglo-Saxon period (AD c.410-570).

Figure 1 Frequencies of 44,690 finds from all periods in the wider sample of finds recorded by the PAS in 2013.

Figure 1 Frequencies of 44,690 finds from all periods in the wider sample of finds recorded by the PAS in 2013.

In the grand scheme of things 319 objects is not very many, though it is by no means inconsiderable.  Nevertheless, these are illuminating statistics, showing the relatively impoverished quantities of metalwork from the whole early medieval period.  The 5th and 6th centuries especially, with their relatively paltry 319 finds, show more similarity to the Iron Age in terms of frequencies of deposition and recovery than any other period (there were 697 Iron Age finds, compared to over 15,000 Roman finds and nearly 13,000 Medieval finds).

The early Anglo-Saxon finds can be grouped into the 13 categories given in Figure 2.¹ As is abundantly clear, brooches and other dress accessories (e.g. clasps, belt equipment, girdle items) make up the lion’s share of the sample, comprising about 85% of it.  There’s nothing unexpected about that, but if we pause to think about it for a moment, it’s quite an important thing.  The vast majority of items recorded by the PAS from this period is not just brooches, but worn items of all varieties.  There’s a good reason for this – most of this material comes from disturbed graves. What we’re mostly looking at here are ploughed-out mortuary assemblages minus nearly all the ironwork and ceramics, obviously not recovered by metal detectorists looking for decorative metalwork.

Figure 2 The 319 5th - and 6th-century  finds divided into broad categories.

Figure 2 The 319 early Anglo-Saxon finds divided into broad categories.

With the basic data outlined, we can look at those categories in a little more detail.  Below is the breakdown of brooch types.

Figure 3 Frequencies of specific types among the 193 brooches.

Figure 3 Frequencies of specific types among the 193 brooches.

Among these, bow brooches are easily dominant, occupying the largest four categories and more than 80% of the entire sample.  Among these, cruciform brooches are the most common.  I’ve explored the potential bias toward these items in a previous blog.  The sheer size of the gap between cruciform brooches and most other types, however, is still quite remarkable.  They easily dwarf even the relatively common disc and saucer brooches.  Because disc and saucer brooches are common in the west and rare in the east, this may well have something to with an easterly bias in metal-detecting activity.  But I’ll leave that for someone with more knowledge of such matters to comment on.

Figure 4

Figure 4 The distribution of the 319 5th- and 6th-century finds recorded by the PAS in 2013.

Looking at the overall distribution of these finds, there is an overall easterly bias, which should be expected given that this zone quite comprehensively covers the area of 5th- and 6th-century furnished burial.  The westerly outlier in Shropshire is a notable exception. Again, there may be a bias toward Norfolk in this sample, again perhaps down to metal-detecting activity, and it seems the Isle of Wight has seen far more finds than might usually be expected.  There’s also been a small number of significant finds from the northeast, where this kind of material is usually quite rare and even a couple of finds from London.

With there basic parameters established, I’m going to leave this survey here for now just as the bare statistics. but I’ll be returning to some of the significant finds and groups of finds in future blogs.


¹ Some more detail on those categories. “Brooches” includes all varieties, but not pins, which have their own category.  “Clasps”, comprising wrist-clasps and their accoutrements, form their own class.  “Belt equipment” includes buckles, buckle tongues, strap-ends, belt mounts and other such items.  This has been kept separate from “Girdle equipment”, which includes just girdle-hangers and chatelaines, but could theoretically also include items like keys and purses.  “Toilet items” includes just tweezers and ear-scoops.  “Weaponry” includes not just ironwork such as swords, spears and shield bosses, but also their decorative fittings.  “Knives” have a separate category.  “Vessels” here includes just fittings for copper-alloy bound stave buckets.  There is a single coin, a Visigothic tremissis, in the sample too.  “Beads”  applies only to glass beads.  “Mounts” is a largely miscellaneous category comprising unspecific but generally decorated plates of unclear use.  The “miscellaneous” category is composed of unidentifiable fragments and assemblages of related objects.


2 Comments on “Anglo-Saxons on the PAS: 2013 in Review (Part I)”

  1. Morgana Krinsley says:

    Wow, thank you so much for doing this! What a great resource!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s