What Data About the English and Welsh Archaeological Record are in the PAS database? [UPDATED]

One of the things I used to write about when I was a finds specialist back in England was slag. I loved any kind of metalworking waste and wrote up reports for a wide variety of sites (most of which never actuially made it to the final monograph and were archived). In fact it was a doddle as most excavations produced one or two boxes of slag-like material at least, so it was one of the things that kept me in rent, beer money and fed my cats in the 1980s.

So, I figure, if the PAS is producing any kind of a record of the archaeological record of England and Wales, then because there are over a million metal items recorded there, it follows that there would be a lot of slag and metalworking waste on the sites that produced many of them. Wouldn't there? Each billet of iron that came out of the furnace would be accompanied by furnace and smithing slag. Many objects that were produced from it (depending on the quality of processing of the iron it was made of) potentially would produce more smilhing slag and hammer scale. Each non-ferrous cast object would produce some slag-like material (here it gets more complicated chemically). I have visions of boxes and boxes of slag. Big boxes.

And the PAS database? Pathetically, among the 1300000 objects in total, according to the database search engine, a paltry 158 records of slag (!). Now I know that this type of material is not everyone's cup of tea, and it is difficult to learn about, but some of those 158 descriptions by the FLOs and other recorders..... hmm. That's not me being snotty, it means that those descriptions cannot really be used (or at least relied on, as I can still guess from the pictures what some of them really are) but they are not a source of knowledge for any serious specialist study of the subject. What has happened is that artefact hunters, being collectors, have not been picking this unlovely sort of stuff up. In fact, looking at some of the shapes of the pieces, I suspect that a number of them were brought to the FLO incidentally with the question "wassat then?" or "is this sumfink?", as some of them for example have a slightly zoomorphioc form.

If we use the database mapping facility (top right) we get a picture of very uneven recording. So yes, there's quite a bit in the Weald, major iron-producing centre, but the Forest of Dean (Monmouth area)? Now, I've done fieldwork there and there are fields of the stuff. And it is interesting slag, as slag goes. Not a single dot on the map.

But there is something else, we know that PAS head Mike Lewis is doing some kind of a project on Medieval Productive Sites as potential market places (or something like that). Now, from what I know of the period (and not just in the UK) many of those those 'productive sites' would also have been production sites, especiallly if they really did function as exchange nodes or markets... and the lower map shows us Medieval+slag mapped from the PAS database... all six pieces of it. Now, does that mean that, by some quirk of fate, those 'market sites' (or whatever they are) that have now ended up in fields explored by metal detectorists are ALL sites where no iron was ever brought, smithed, or objects repaired? Or is it the case that the "partners" with their metal detectors are not actually engaged in a partnership producing archaeological data - but filling their own private collections with what they fancy, and not that which the archaeologist needs? What kind of a symbiosis is that?

What is the point of basing any serious archaeological research programme on data so flawed and so selectively biased because its gone through the hands of collectors with a different and private purpose in mind? What the PAS actually mainly documents is what collectors collect today and not very much more. That in itself might make an interesting thesis or two, but to spend millions of public pounds getting the information together for that....?

Update 31.10.2018

All MWD (1372 records)

Dan Pett in the comments suggests there in considering the topic of the appearance of the type of material that is less enthusiastically gathered by artefact collectors to put in their private antiquity collections, I could also have lookd at the category of metalworking debris in general, where I would find more information. Obviously if we are looking at the evidence of past activity through a 'metallic' prism (by looking at reported objects from the personal collections of "metal detectorists"), then evidence of the manufacture of those artefacts is also relevant to the pattern. Indeed there is information to take into account in this enquiry, the total number of results that are available is 1,372 records, so of the 1,300,000 objects there are somewhere around (see below) 158 + 1372 records of pieces of metalworking debris (MWD) of various kinds and slag. That means 1520 records of these fragments all together. Yesterday I was just making a general point in a short essay, rather than writing a thesis, so I did not want to get into a discussion of the more general topic, so I did not do a search for metalworking waste (the category will include furnace waste, tuyeres, crucibles, piece moulds, investment moulds, hammer scale, and various kinds of metalliferrous ashes and slags as well as metal offcuts, pre-forms and so on). But since a former member of PAS staff has brought it up, I'll take a look at this as well.

All copper alloy MWD (922)

The first thing is the map of all the finds (above) shows that there are FLOs who know what they are looking at and why its worth recording and those that do not. Those that have encouraged finders to recognize it and bring it in and those who have not. So I can see where Adam Daubney works from that map (and from his descriptions that he in particular knows what he's writing about), I can see the Surrey FLO (the late David Williams, I bet) and the Wiltshire FLO are reasonably active there - but some of these are metal bits from Bronze Age "founders' hoards"  which really are a separate category from the other stuff covered by that thesaurus term. I can also see that, for example, the Essex FLO who called the police on me and the mouthy guy from Durham who says this blog is falsehoods and 'fake news'  have not got much to show in this category.  So the coverage is uneven and this is due to the collecting habits of artefact hunters as well as the PAS staff interacting with them. Some possibly are just not recording slag and MWD at all, perhaps even there are some that do not see it as 'important' in their regional context as coins and heraldic pendants. There may also be others that are actively drawing artefact hunters' attention to this material and its importance, and encouraging them to report it for recording. And there are probably shades of grey between these two extremes. the point is of course, this is what the database is recording the effects of.

Lead and Lead alloy MWD

There are also issues about what should go into this category, so some slag is listed as MWD and some as just slag. Some of the desceriptions in this category are way off, written by archaeologists (or are these the PAStexplorer volunteer karaoke recorders in action? - as I think this one must be). Personally, I do not see how anyone can do any kind of work on archaeological finds without a knowledge that goes a little beyond merely 'basic' about how they are made, how metal is turned into the objects they are describing. Some of the MWD descriptions and the vocabulary used in them suggest to me that awareness of this is not general in the PAS-recording community. More in-house training needed instead of fluff ('aren't-we-doing-well?') conferences. Some proper validation of these public records would not go amiss, sorting out the terminological mess and infelicities (like "iron flag" in one record as just one example). It is symptomatic that of these the database says that 155 of these records have been checked (in the last twenty years) and of the total as many as 1,217 are still awaiting validation (just under 90%) (!). It seems there has been a sacrifice of quality for quantity here (not the only place in the 'database', I might add).

Medieval metalworking debris (14)

The next map (2) shows copper alloy metalworking debris (includes those BA hoards)  - not a lot different from the gebneral one, because over two thirds of this stuff is (what the FLOs have identified as) copper alloy metalworking waste. The third map (3) shows the 'lead' (264 records) and 'lead alloy' (34) plots. This suggests that the geographical extent is more indicative of different FLOs using different terms for the same type of material. I think also there is a great difference between lead offcuts and spills/driblets and (for example) iron furnace slag. Lead has a very low melting point and was used for soldering and plumbing, and (later on) casting all sorts of trinkets at home, so lead 'metalworking debris' is not readily classifiable as 'industrial' in the same way as ferrous and other non-ferrous production waste.

And, to come back to the point about the Medieval productive sites, map (4) shows them...  I do not think that adds an awful lot to the map in the orignal post above, and once again it seems that the map mainly shows the effects of the activities in the same three FLO areas

In conclusion, having looked at two related catgories of PAS recoirds, the conclusion remains that what is being recorded in the PAs database is the activities of modern collectors (and perhaps also the effects of the interactions between PAS staff and those collectors) to which the actual archaeological record from which the exploited material was derived, forms only a background, rather than being the subject that most supporters of the PAs imagine it to be. The PAS database is indeed showing the destruction of the archaeological record and only showing us in broad terms where it is being destroyed.