Last man standing

Screen shot of our 2001 CCI online announcement

Yesterday, Bob Van Arsdell emailed me to say that the British Museum has finally pulled its version of the Celtic Coin Index (CCI) off the Internet. Although the records still exist within the Portable Antiquity Scheme, the British Museum Celtic Coin Index no longer has a separate identity. Their site advises that new finds of Celtic coins be handled by Finds Liaison Officers (FLO).

The Celtic Coin Index as a card file was started at Oxford in 1961 by Sheppard Frere and Derek Allen. More than just providing find spots of Celtic coins, the CCI also included coins from public and private collections and coins noted in the literature (mostly sales lists and auctions). The British Museum seems now only interested in new finds from the ground.

After reading Bob's email, I did a Google search for Celtic Coin Index, and sure enough, the British Museum page records had vanished and my original version's main design (William Astle of Lexicom changed it into a database driven site) now dominates the search results. Even the PAS Celtic Coin Index url does not even have a redirect to the Iron Age coins guide and delivers only a "404 notice" of the missing file.

Almost from the start, producing the first CCI online was a lot of work. My wife spent the last three years of her life building the site and the database that generated the pages while fighting terminal breast cancer. The Institute of Archaeology at Oxford went back on their agreement to fund our work from a grant and spent some of the money on their staff salary and the rest went into general coffers. We decided that the CCI online should exist, anyway even though we decided to include only the British issues and not the imports, and it became my property (according to the Oxford lawyers) because we had received no payment whatsoever and thus it was not "work for hire". I might even now arrange to update the records at some time.

As Oxford had left me with a debt of more than a thousand dollars for their last year's bandwidth charges, I thought that Lexicom, its hosting ISP, would take it down so I went to see Lexicom's president, Michael Rae to give him the bad news. His response was a most pleasant surprise: he said that he had always believed in Carrie and myself and the importance of the work. He would thus write off the debt, and would continue to host the CCI online on its dedicated server and our old personal site (now archived) at no cost to me at all. Before I left their office (in a daze!) he mentioned that he owned a condo on Regent's Canal in London and if I was visiting London anytime he was not there I would be welcome to borrow it. The day turned out to be far better than I imagined.

The unannounced limitation of Frere and Allen's original vision is a serious matter. While keeping a record of find spots is important, by including all coins regardless of whether there is a recorded find spot, it make the CCI a valuable research tool for typological, iconographic, metallurgical and epigraphical information. I had designed the site so that it could be easily browsed by Colin Haselgrove's regional system as well as by Bob Van Arsdell's tribal listings of coins. This was done to make identification of coins easier for those who only had a coin but little or no data.

Another very important feature of the complete CCI online was that the data could be searched to see how find spot data was handled by public collections, private collections and the trade. The PAS database apparently has no such function as it's main theme is to record find spots and if all you have is a hammer, every problem starts to look like a nail.

I'll give you the source data for the CCI records of coins from 1961 to 2001:

Total number: 32,348

Public collections: 9,770
Private collections: 2,319
Literature: 20,259 (this is mainly coins published in the trade)

Of these sources, find spot data comes with 69% of public collection coins, 59.3% from private collections coins and 57.7% from the literature. The span thus is only 11.3 % and the average is 61.2%. Public collections contain a percentage of known site finds, and private collectors reporting coins to the CCI are more likely to have considered the find spot data more important than in all of the literature where, in the case of sales, some of the data comes with the coins but is not published in the catalogue (The Mossop sale, for example, lists find spots and/or weights but does not always associate each with a specific
coin in a lot of several, so we sometimes get a list of weights and a list of find spots but we don't clearly know which belong together from just the catalogue The coins, themselves, came with their tickets with all of the data). This, and the fact that a few would come from general catalogues where no specifics of each illustrated specimen are given (such as Seaby catalogues) would explain the small differences in the literature and private collection data.

We can see from this, that findspot data is mostly recorded or retained by dealers where it is known. This is just common sense, there is every reason to include such data and supply it with the coin, although with the animosity from the anti-collecting lobby, many dealers do not seem to be including as much of this data recently in their sales catalogues. Since the CCI was taken over by the PAS, they also report less to it as they probably realize that there is not anyone working there that is a specialist who can provide further information. This is just speculation on my part, but I have noticed a trend of fewer find spots in the literature over the last few years.

The importance of Celtic coins in the trade is revealed by the fact that the percentage of coins held in public collections is only just over 30% of the total. The differences in the numbers of reported find spots compared to those where this is not recorded is far less than might be imagined.