Holes in History, for what?

On a collectors’ forum yesterday, I got involved in a discussion of an ancient coin (a hemidrachm from Cherronesos in Thrace) which had been turned into a piece of wearable jewellery. Making it into a ring really does not seem to me to add anything to any numismatic scholarship or anyone’s knowledge or appreciation of the past (neither their own or anybody else’s). In my opinion, it is an ignoble end to what could have been used as archaeological evidence somewhere before it was hoiked out of the ground to end up decorating a party-goer's finger. The seller justifies the treatment of this object by claiming “there is nothing historically significant about this coin” and then goes on to say what he understands by a “historically significant coin”. (“It isn't a gold multiple. It isn't an Athenian dekadrachm. It isn't an Eid Mar denarius. It's a common hemidrachm”).

There seems to be a common type of misunderstanding among collectors that there are “major antiquities” like an early copy of Polykleitos' Doryphoros and the 'Euphronios Krater'. These they see for some reason as somehow different from “minor antiquities”, the ones they think its OK for someone like them to collect without bothering about provenance or documentation of legal export. This ignores the real reason why archaeologists oppose the current form of collection of portable antiquities. While being notably eager to criticize archaeology and archaeologists, collectors using this argument seem actually to have very little idea what archaeology is about, what it does and how. In connection with this approach, I cannot fail to recall two “coins I have known” from my digging days (please note: I have not got the report readily to hand so am writing here from memory).

The site concerned in the Baths Basilica site in the middle of Roman Wroxeter, England (excavations directed for English Heritage by Phil Barker, a wonderful man, who passed away in 2001). Its quite a well-known site and excavation, so I will not describe it in any detail. Suffice to say that it is a site with thin and incredibly sensitive vertical and horizontal stratigraphy, most of which relates to the period at the end of Roman Britain, and the various uses the huge public building and its site was put to in the fourth and fifth centuries AD. The excavation was slow, painstaking and very costly. There were thin trampled floors, rubble spreads, ephemeral (sometimes extremely ephemeral) traces of complex timber structures. There were also thousands of finds mixed in. Some of them were obviously redeposited (“residual”), first century samian and fragments of equipment from the underlying military occupation, but the main bulk was either undatable in itself (like loads of animal bone fragments telling us about urban diets if only we can date them by association) or of typical generic Late Roman type, including tonnes of pottery sherds. All this was carefully collected and segregated according to archaeological contexts (“layers”). Datable finds were carefully plotted. Among them of course were the coins.

Most of the coins from Wroxeter were the typical stuff metal detectorists and fieldwalkers get from any Late Roman site in southern Britain, various “radiates” and especially small corroded “Constantinian bronzes”. These are 'two-a-penny' to most dealers and collectors in Britain and beyond. Metal detectorists refer to them contemptuously as “Roman grots” and without much thought toss most of them into the bag destined to end up as an eBay “one kilo of unsorted uncleaned Roman” lot for spotty teenagers in Tucson to “zap”. So not much of any dealers’ idea of “coins of historic significance” then.

Their use as archaeological evidence however is not reliant on how “common/saleable” an artefact is. Although somehow its an uphill battle persuading collectors of this (they “know better”), in archaeology this comes of course from where and how it is found.

At Wroxeter at the end of the 1980s I was supervising the east end of the South Aisle (just in front of the Old Work for those that know the site). Something really odd had happened there after the basilica was not being used as a building, but an open space after it had lost its roof. A huge rectangular hole had been dug through the floor across the whole of that end of the South Aisle, and then backfilled with a thick sloping layer of rubble mixed with black earth (in the background of the photo here by Roger White). This infill contained many finds, plaster, glass, roof tile, nails and other metal objects, and as we slowly and carefully removed it several corroded 'Constantinian' small bronze coins came up from one spot. We started detailed three-dimensional plotting of these items as they came up over the next few days. This was potentially useful dating material for the activity we were studying. If their distribution in the rubble showed they had been hidden there as a group, then their deposition would be a date (a so-called Terminus post quem) for the rubble layer itself. These coins changed quite rapidly and if the find had been a hidden hoard, we’d have got quite a close date for the whole deposit and what is more, those stratigraphically related to it. Also we’d have to think about why a hoard was deposited precisely here right next to a major thoroughfare, but in an area of known ceremonial activity at the time. As it happened when we’d finished all that work, it became relatively clear that the distribution of the coins relative to the internal structure of the rubble dump was such that they had been dumped (probably unnoticed) together with the rubble. Presumably they had been dug up somewhere outside the excavated area in a clod of earth and only scattered when the earth was tipped into the hole with the rubble. As it happens our hopes to use this particular assemblage of coins were futile, but had the question not been posed and the three dimensional distribution of finds not plotted in detail, we would never have known, and indeed could well have drawn quite the wrong conclusions.

Now let us think what would have happened if this shallow deposit had been located and dug out by a metal detector using artefact hunter. There is absolutely no way that in a narrow hole they could have recorded the distribution of those items with the same detail as over a week of detailed open area stratigraphic excavation. In fact, I suspect they would not have located all the coins of that group. This is information about that part of the site that would have been completely lost had that deposit merely been dug out to go on sale as “grots” on eBay. Even if the PAS had been involved, my opinion is that the character of the evidence was such that true nature of that group of coins and its relationship to the basilica would never have been established.

The second coin was also a Late Roman bronze of the same general type, a single coin, but a crucial piece of evidence. After several years digging, recording, planning, finds plotting our way through very complex stratigraphy of several major phases compressed into a few tens of centimeters, an oven was found, securely sealed by all the deposits above it. At the bottom of the oven, and absolutely certainly deposited in it after it had gone out of use, was a single coin. The point is that the entire sequence of stratigraphy and a floating seriation of the pottery evidence laboriously constructed over the preceding years was at last given a fixed starting point (almost) by that one securely sealed coin. That single coin gave the rest of the work over most of the site the fixed beginnings of a chronological framework that had been lacking (or rather the subject of informed surmise) before that. It turned out that beginning of the formation of the series of deposits we had been examining [and which we'd already worked out clearly had been the product of a long period of development and redevelopment of the site] had been much later than we had previously concluded on the basis of other evidence. What we had thought were layers of the early fifth century higher up in the sequence were in fact in most likelihood much later and all the finds in them were redeposited. Explaining that away, when Britain was supposed to be in terminal economic, social and cultural collapse is a connundrum a decade after the excavations were published.

Again, if a metal detector user had gone over that field before the excavation and pulled out that “Roman grot”, the whole sequence of this part of the Wroxeter site would have been irreparably damaged, and that information and all the consequences following on from it, would have been irretrievably lost.

The portable antiquity collector really should be more aware of this kind of situation and how their hobby does affect the archaeological record in ways they cannot even imagine (though they should try). The bottom line is that the importance of individual pieces of evidence in archaeology only emerges when they are seen cumulatively and in the full richness of its context. Taking coins or any other atrtefact, no matter of how “common” a type, out of its context produces very little information in itself, but potentially (and in ways the artefact hunter is unable to predict or guard against) simply trashes a whole portion of a site, making it impossible to interpret.