The Archaeological Values of the PAS Database (XVIIIb): Smart Alec and the Crotal Bells (part two)

 

This follows on from The Archaeological Values of the PAS Database (XVIII): Smart Alec and the Crotal Bells. Apart from greed and self-centredness, one characteristic that may be associated with metal detectivitis is manic obsession. A UK metal detectorist hiding behind the pseudonym Alec for the purpose of posting flippant comments on this blog sided with fellow detectorist Paula Pennington in asking "how many crotal bells" we need recording if artefact hunters are removing them from the archaeological record and adding them to scattered and ephemeral personal collections. He suggested that their historical (sic) context was that they all: "fell off a cow in 15th c because the farmer couldn't attach his dingaling properly.. subsequently the cow wandered off and ended up in the medieval version of mcdonald's". When I pointed out that they were associated with horse harnesses, he noted that in "disputing that they were used for agricultural use" I was "implying that everyone at PAS are therefore incompetent". When challenged to prove that "everyone at PAS" said they were not related to harnesses, he first came up with a ten-year old mention on the website of a small local museum on the northern edge of Salisbury plain who'd acquired one of these bells from a metal detectorist and said it was a sheep bell (he could equally have referred to a page on Dartmoor folklore that has some common features with the museum's narrativisation). When I pointed out that the page he cited was not even "anyone" connected with the PAS, let alone "everyone", he then searched the PAS database for cow bell and came up with this:  https://finds.org.uk/database/artefacts/record/id/850742

This tickled me no end as (although anonymous, the time it was made, 'three years ago' tells us), this was the ex-Essex FLO Sophie Flynn, who is not on my personal list of liked and trusted FLOs. I have never met Ms Flynn, but she thought it appropriate professional behaviour to set the police on me for discussing one of her public Finds Days on this blog. As such, the opinion of Sophie Flynn is not rated very highly on this blog. I am not really convinced that if Ms Flynn describes something as, quote: "a cast copper-alloy Post Medieval (c. 1700-1850) rumbler or cow bell" that it means that "everyone in the PAS" thinks such bells were used as cow-bells [also note "rumbler OR cow bell"]. Smart Alec has failed to prove his point. 

The reason for this post is not, however, a comparable obsession with proving my own point, but that this exchange reveals an epic misunderstanding within the PAS database that might be worth exploring in the context of determining its reliability and archaeological values, and as a means of transmitting them to the public that pays for it.

In my previous post I mentioned the guide by Rod Blunt on the privately-run UK Detector Finds Database (UKDFD). While I am no fan of this 'resource', it should be recognised that there are some noteworthy features here, and some of the 'guides' to different types of finds are among them. Rod Blunt generally knows his stuff, and the UKDFD guide to Crotal Bells more or less contains everything any finds nerd might need to know on the topic (it's the one cited by the Market Lavington museum in preference to the PAS one). Notably, what that guide lacks is a discussion of the function of the cast spherical ones with an integral loop and internal pea, a basal slot and 'sound holes' in the upper body. it is this type that is colloquially known (among metal detectorists in particular, as 'crotal' bells - the name comes from the Greek crotalon/Latin crotalo, meaning rattle). 

If however we turn to the PAS website, we find a rather wishy-washy finds guide (by Helen Geake) that deepens the confusion. Exactly the same page is referred to when searching for "crotal bell" and "animal bell" and it may be this that confused Smart Alec. The problem is that the author Helen Geake includes on this page (titled 'Bells') a number of totally different types and dates of objects (at least twelve), and these cast spherical bells are just one of them, while hawking bells and open clappered cow bells are also included in this broad category. Indeed she says ("PAS object type(s) to be used"): "Always use BELL. The PAS database avoids ANIMAL BELL because in most cases we are unsure about the function of any particular bell. CROTAL is only used for a very rare Bronze Age object'. This is rather unfortunate, as the majority of recorders (FLOs and PUBLIC) totally ignore that. 

BELL/CROTAL plot
(Cf the plot in the previous post)

If we search for 'bell' in the PAS database, the search engine throws up 12,292 items of all twelve+ types, but if we look at the object categories, oddly: BELL (4,241), COIN (820), STUD (373), CROTAL (360), ANIMAL BELL (138), FURNITURE FITTING (131), VESSEL (97), HRNESS PENDANT (83), WEIGHT (71), UNIDENTIFIED OBJECT (64), TOKEN (63), BROOCH (58), MOUNT (55), TOGGLE (38) and lots more. there is obviously something wrong there. And look at the result for "crotal bell" under "bell" (3600, while a search for that term by itself produces ten times as many. What kind of a 'research database" is this? It can't be used for straightforward analysis by archaeologists (even if they had full access to the hidden fields), and how is a member of the public (like Smart Alec) to use it to learn anything at all?

Look what happens when you search within "bell" for "crotal". While there are outliers (in Buckinghamshire (5), Hampshire (4), Cornwall (3), Cambridgeshire (2), Suffolk (2), Herefordshire (1), Devon (1)) the map shows three clusters corresponding to modern counties (NMS (200), LIN (40), OXF (11)). Obviously these records are differentiated by means of local recording protocol.  

The same thing happens when you search for ANIMAL BELL within BELL. here they almost all cluster in and around Norfolk.  Is there contextual evidence for these (whatever type they are) bells being recovered from associations that support such a functional attribution? Or is this just a localised piece of guesswork?

Whatever the reason for this, it seems that there are serious problems in the "standardised terminology" of the PAS database that is supposed to make it easily searchable. This problem is likely to get worse as more and more people become involved in the recording.