Tuesday, 11 November 2014

PAS Finally Try to Estimate Scale of Non-Reported MD Hoiking (1), but Can't Even Manage to Present that Properly

Readers of this blog will be well aware that I consider that the PAS are really pretty incapable of giving a transparent and nuanced account of the actual relationship of what they are "recording" to what artefact hunters are taking. I am among those who've been asking the PAS and its supporters about this for over a decade, getting fobbed-off and the question ignored all the time. How nice it was to actually see, after nearly two decades PAS-ing about, that they actually address this problem in their (rather superficial) Guide for Researchers produced as part of the Leverhulme Trust funded project 'The Portable. Antiquities Scheme as a tool for archaeological research'.  

To cut a long story short, in this document they give figures which more or less correspond with the Heritage Action Artefct Erosion Counter (dullard doubting detectorists and know-it-all-Florida-arts-graduates in denial please take note).

The only trouble is that when you try to check it out you come across more than a few problems for your seventeen million quid (plus a Leverhulme grant). So the project's Bland, Robbins and Pett start off with trying to say "how many detectorists' there might be (I am quoting below without the footnote numbers)...  
"Thomas suggests there are 12,500 detectorists; Barford believes there are around 9000 detectorists; whilst Heritage Action uses an estimate of 8000 detectorists. Based on all these figures there are (sic) ~9500 metal detector users across England and Wales. Of these only ~7125 detectorists are likely to recover finds that could be recorded by the PAS as it is thought that a quarter neverfind any 'recordable' artefacts" (Robbins 2014, 14)
So Robbins takes her ~7125 and multiplies it by a coefficient she determined from a questionnaire (where of course we can be sure that every respondent told the truth, the whole truth and nothing but the truth irrespective of how they thought the results were going to be used) and comes to a figure of 260,000 objects hoiked a year. That's pretty close to the HAAEC's figures.   

That simple short passage however reveals that the PAS in all its Bloomsburian arrogance and carefree avoidance of the open debate of the issues simply does not give a tinkers. I'd say something like this needs copy-editing, and I'd say this has not been done. Just take a look at this:

1) The first figure Suzie Thomas's "12,500" is referenced by the document's author "Thomas 2010" with no page number. The bibliography tells us that we are being referred to Thomas, S. (2010), ‘The Relationships between Archaeologists and Metal-detector Users in England and Wales: Impact of the Past and Implications for the Future’, Ph.D. thesis (Newcastle University).... except the actual year on the title page of that work is 2009. Wrong reference, lack of page number -go back and check it. It gets more complex, searching the whole of Thomas's thesis (available online) does not reveal anything like the figures Robbins quotes in the garbled footnote 1 on page 14, where "England and Wles" and "the UK" are irretrievably mixed. The number of metal detecting clubs has gone up since 2009 (besides which 200 x 50 is not "10100"). Fail for PAS.

2) "Barford believes there are around 9000 detectorists [in England and Wales]", well no, no I do not. She gives a reference here to a blog post on this blog (Barford 2011) which the bibliography says is ... wait for it... the number of detectorists in AUSTRIA (see the difference?) and the number 9000 does not occur in it. This is the second WRONG REFERENCE in this passage. In fact this is not even a mistaken substitution of one interesting Barford blog post for another, I believe I am right in saying that nowhere in this whole blog (or I'll wager anything else I've written) does it say that I "believe" there are "9000 detectorists in England and Wales". Even if I said it, it's not what I now think. Looking at the way PAS reports of numbers of clubs and memberships are rising (and other factors), I believe the number of active metal detectorists  in England and Wales has increased over the past decade or so since I originally estimated the "8000" which is what I've since written here a number of times, but before Robbins put her text together.

3)  "Heritage Action uses an estimate of 8000 detectorists", but why Robbins (fn 3) gives a reference to "Barford 2008" here is beyond me. I am not "Heritage Action", I am not even a member of Heritage Action. She really ought to give a reference to Heritage Action's own material. The third bibliographical faux pas in the same sentence... not looking good. Nevertheless it should be pointed out that this "8000" used by HA is my "8000", I gave it  (and not 9000) to them. The figure comes from metal detector dealers and was lower than other estimates being used at the time.

4) This "quarter who never find anything" comes from Roger Bland. Whether that is the result of a proper survey or just a figure from the top of the Keeper's head remains unclear.

But let us see what eliminating the false "9000" does to Robbins' figures. In her text of the three estimates quoted, one is a phantom. This leaves us with two, the old 8000 and Suzy Thomas's (roughly contemporary 12500). The average is therefore ten thousand detectorists for England and Wales, not "9500". Getting rid of an (arbitrary?) 25% means we are discussing the effects of 7500 artefact hunters.

On present evidence, Robbins' figures, which are the best the PAS can come up with after eighteen years of outreach and seventeen million quid, would therefore indicate that 273750 artefacts are being hoiked by artefact hunters (750 daily). That works out as 4,452,750 non-Treasure finds hoiked since the beginning of the PAS.

While this is lower than the HAAEC estimate, it is the same order of magnitude. Let us compare that with the non-spun figures for the voluntary recording of artefacts by artefact hunters from the quantity they hoik for their collections or sale. The PAS claims "over a million finds recorded" but that number is boosting for publicity purposes with multi-thousand coin hoards (which are Treasure objects which it is mandatory to report - so have no relevance to the question the HAAEC is designed to address). If we are to assess the success, or lack of it, of the current voluntary recording scheme we need to remove these figures from the total. I did this a month and a half ago and we get Wednesday, 10 September 2014's total of 840,870 objects within 622, 356 records. Although the PAS has recorded a number of non-Treasure and Treasure finds since then, the general picture is not going to look much different from that represented here (see vignette above), with the Robbins total takings figure of 4,452,750 non-Treasure finds hoiked since the beginning of the PAS compared with the 840,870 objects actually recorded (though readers should know I consider the number of records to be more significant).

In other words, since the PAS was set up and started calling itself a "success" and all artefact hunters "responsible", its own figures reveal that less than one in four hoiked finds are being recorded, and an estimated total of  3 611 880 recordable finds have vanished into artefact hunters pockets and some ended up by now in skips without any kind of proper record.

If anyone wants to visualise that, using the 'chalk-line from Marble Arch up Edgeware road' model I adopted earlier for our own figures, the PAS estimate - at one centimetre per unrecorded find - would reach a kilometre beyond the site of the Boxmoor Roman villa on the northern outskirts of Hemel Hempstead.

Now whether Doubting Dullard Detectorists in Denial and others prefer the PAS's own figures or Heritage Action's is immaterial, there obviously is a very serious question to be addressed here. Let me make it easy for you, by how much would the PAS' own figures given in the 2014 research guidelines have to be wrong to make this situation in any way acceptable

No comments:

Creative Commons License
Ten utwór jest dostępny na licencji Creative Commons Uznanie autorstwa-Bez utworów zależnych 3.0 Unported.