Thursday, October 30, 2008

The Controversial "Excavation" of a Coin Hoard

Paul Barford and David Gill have both commented on a news report (L. Hannam, "Treasure hunters set to coin it with Roman haul," MK News) about English metal detectorists who dug out a meter of soil to recover a large Roman coin hoard, which was associated with other ancient remains. To their credit, the detectorists did report the hoard - formally declared a treasure - to the responsible authorities.

Barford and Gill, however, both question whether or not situations like this are what the PAS was designed for. Many think of the PAS as recording primarily surface finds from ploughed fields, which were already "decontextualized" (in addition to the two original posts, see also Barford's "What would the PAS Say?"). The concern in this instance is the amount of earth removed to get to the coins and the fact that the hoard was not an isolated find. Important contextual information, which could provide greater insight into the circumstances surrounding the deposition of this hoard or conversely the associated remains have been destroyed.

Peter Tompa, a former president of the Ancient Coin Collectors Guild (ACCG), has responded forcefully to Barford and Gill's respective concerns. While he is correct in that it seems the detectorists did what UK law requires of them, his attempts at assuaging ethical concerns from an archaeological perspective are inadequate.

For example, he stated that since the find was made on ploughed land, the context was already disturbed. This is an argumentative claim since it is impossible to know how far the ploughed topsoil affects the archaeological remains without a proper archaeological investigation. Since one meter of earth had to be removed to get at the coins, it would appear the plough might not have reached this depth as the coins would have been scattered and pulled out to the surface had this been the case (on the depth of ploughing, see also Barford's "The Washington Lawyer and the Metal Detectorists").

Mr. Tompa also claims that the broken vessel which contains the coins is an indication the plough broke the vessel. Again, this is another assumption. It may be a possibility, but it impossible to tell without the context (now lost). Coin hoards are frequently recovered in archaeological contexts in broken or damaged vessels - this can be an effect of the geology or weather or can be a result of other post-depositional processes in which something may have fallen on the container or the container itself fell, etc.

Mr. Tompa cites Roger Bland and the PAS in his response, as so many of the dealers and collectors at the ACCG often do. It appears, however, they frequently take his work and views out of context. In a review of Cuno's recent book, Bland reacted strongly against "US cultural imperialism" of the sort the ACCG subscribes to (see Gill's, "'An Example of US Cultural Imperialism at its Worst'").

The archaeological inaccuracies in Mr. Tompa's reply are perhaps natural since he is not an archaeologist but rather a collector and attorney. I am sure if I were to attempt to discuss legal issues in some detail I would make some errors as well.

The concern of archaeologists and many numismatists is that information and history is destroyed in the search for curios and/or profit. The value of context and the threat of the indiscriminate trade has been highlighted in my article "A Survey of the Material and Intellectual Consequences of Trading in Undocumented Ancient Coins..." Frankfurter elektronische Rundschau zur Altertumskunde 7, 2008, 1-13. A collection of 11 essays - 10 of them in English -discussing the value of archaeological and contextual methods in relation to coin finds from excavations, hoards, etc., is about to be made available: H.-M von Kaenel and F. Kemmers (eds.). 2008-forthcoming. Coins in Context I: New Approaches in Interpreting Coin Finds (provisional title). Studien zu Fundmünzen der Antike 23 (Mainz: von Zabern). It is the goal of these essays to make developing contextual methods, which are already at the fore of research on numismatics and archaeology, more widely known and they build upon methods and theories that have been forming over the past few decades.

As Barford and Gill point out it is a false assertion made by some ancient coin dealers (which does not qualify them as being archaeologists) that most hoards are "isolated," i.e. found in the middle of nowhere with no associated remains and thus archaeologically "insignificant." In the above-mentioned article, I examined this assumption and pointed out that hoards can compose substantial percentages of all the coin finds recovered at archaeological sites. One of the largest hoards ever recovered (perhaps the largest), the Reka-Devnia Hoard, contained 350 kg of Roman silver coins and was found within a structure in the ancient city of Marcianopolis - not in an empty field devoid of associated remains.

While the finders of the controversial treasure acted legally and are to be commended for reporting the find as mandated by the law, we ought to consider the ethics of disturbing archaeologically significant sites in such a way as removing large amounts of earth and disassociating objects with their broader contexts (associated remains) and stratigraphy: these are the building blocks for writing histories for which we can only use material evidence (cf. Barford's forceful, but on point, "Give and Take of Obeying the Law"). Indeed, in most source countries this type of destruction is illegal, much to the chagrin of ancient coin and antiquities dealers who often trade in such material despite laws in source countries. There is a difference between picking up decontextualized surface finds and disturbing contexts deep in the earth.

(Image: A selection of coins from the hoard in question from L. Hannam, "Treasure hunters set to coin it with Roman haul," MK News)



Since I first posted this discussion here, there have been some additional posts and facts emerging, see:

"The Newport Pagnell Coin Hoard: Further Thoughts"

"The Newport Pagnell Coin Hoard: Update"

and Barford's
"English Detecting Smoke and Mirrors"



So as not to belabor all this, I just wanted to correct a statement in your post. I am no longer President of the Ancient Coin Collectors Guild, though I remain on its Board.

As to Roger Bland's statements on Jim Cuno's books, I'm not sure how I see they necessarily apply here. The link in my post was a summary of Dr. Bland's talk in Washington. It also includes a link to his slides. They speak for themselves. I should also note Brian Rose of the AIA was there as well as representatives of the State Department and the Turkish and Cypriot Embassies. The program was not designed to pump up coin collectors, but rather to let Dr. Bland explain the UK program and discuss whether he thought it might have any application outside of the UK.

Incidentally, you are correct. I am obviously not an archaeologist, and though I might falter on occasion, when I speak to the issues, I do try to stay within what archaeoogists who are more friendly to coin collectors than some have told me (or I have read).


Peter Tompa


Dear Peter,

Thank you for your comments. I knew Bill Puetz was listed as president-elect, but I did not know when his term would officially begin. Thank you for letting me know this. I have made the change in the above post.

While I did reference your citation of Bland, I was thinking more generally about the way dealers, Sayles and Welsh come to mind, have used Bland's work and research as a means for supporting their ideological agenda. Perhaps I was not as clear as I should have been on this point. What I was getting at is that the PAS and laws in Britain are often skewed or presented from a skewed perspective by some individuals.

In this case, the fact that the hoard was recovered on ploughed land, for example, does not mean that cultural layers were not disturbed given the depth of the digging.

All the best,


For the loss of context see Looting Matters ("A stratigraphical relationship could not be established").


I examined this assumption and pointed out that hoards can compose substantial percentages of all the coin finds recovered at archaeological sites.

That is interesting work... could you give the percentage that you have seen of hoards found within archaeological sites to hoards found outwith. To me of course a hoard is a site, (as a action has taken place) there is also the difference from a votive hoard - and a 'bury it to return later' hoard. Context is indeed important, so is discovery. For example, a JCB hits a stone coffin.. it is damaged, however, you a) now know about a coffin you never knew before.. and b) may have located a wider site. If you take the hoard for example, the psots make it sound like the entire 'site' (and I note that nobody has made much of the lack of what sort of site has been destroyed) has been wiped out... that somehow this unverified 1 metre deep hole has removed the whole site... if this is a Roman farm, this hole could have actually allowed the careful investigation of teh other 99.9% of this fictional (currently) site. People can get hot under the collar and cut text to fit.

My question is
a) would you rather the finders had not gone to the FLO the next day?
b) would you rather the find had never been made?
c) would you prefer a new site to be found at the cost of a hoard context?

I have said to them that I would have prefered that they had not kept digging... but then, they did stop, and the did go to the FLO, which shows that although they may not be archaeologists (who are trained in this sort of thing) they did know enough to go for archaeological help.




Thanks for the comment. Regarding percentages of hoard finds at sites, I used the example of the Magdalensberg where approximately 38% of all coin finds are classified as hoards. One of my colleagues at Frankfurt University is a specialist in coin hoards and says that coin hoard can often make up substantial numbers of finds at sites. I am the numismatist at the Yotvata excavations in Israel, which is a very small Roman auxiliary fort. The excavations - the fort was by no means completely excavated - produced about 200-250 late Roman bronzes. One hoard was recovered containing about 35-40 coins. So even here a hoard comprises about 15-20% of the total finds. I'm sure there are plenty of published reports where one could examine numbers in detail.

I think the PAS is a great program which allows for metal detectorists to find (and keep in most cases) material which is often in disturbed layers in the topsoil. By reporting finds, of course, archaeologists can track where things are being found and send out teams to investigate as necessary to conduct further investigation should a certain type of material start appearing in one area or material that indicates a noteworthy site might exist.

The problem in this case - an ethical one I think - is the way in which the coins were unprofessionally and hastily removed and at such depths. I agree with your opinion as to how it should have been handled. Ideally, once the finders had found this important hoard, it would have been best to report it so that the archaeologists could have preserved as much information as possible. As the finders, they still would have had rights to the find, but they would have been able to preserve some further knowledge as well.


The story of the two guys laying on their bellies pulling hundreds of coins out of the ground is very dramatic. They did the right thing by informing the authorities.

Perhaps you could write a post clarifying the ethical problem you have with these guys so that the ethical lapse is understandable to the non-archaeologist?

Your point is about *how* the digging was done, correct? I read the article but don't understand why you regret that lack of a report.

1. Is the problem that these guys took a risk that could have lost data -- that if there had been a Roman Chariot or Viking helmet nearby we wouldn't know if it was above or below the coins? Or is the stratigraphy always of value?

2. Mostly what I know about finds comes from books like Margaret Thompson's _Inventory of Greek Coin Hoards_ which just lists types and findspots. Thus I've never seen a stratigraphic report. Are there any good ones, written for the educated lay person, that I could read and see what was lost? Because I'm a "coin person" a stratigraphic report for a coin find would be best for me.

I've seen pictures from Israel of expeditions putting sand through grates but I don't understand what they are looking for and how the raw data becomes statistics and then narrative.

3. If a full stratigraphic report was done, would it be put online in the PAS database? Elsewhere?


Dear Ed,

Thank you for your thoughtful comments. There is a lot here and so I'll try to go through them in order. I don't think it is necessary to make another post about this as I indicated in the present one the problem from an archaeological perspective is an ethical concern the loss of information in the manner of digging. Had it been unconvered and then allowed to have been excavated properly, more information would have survived and we could know more about the find through its associated contexts.

Perhaps many 'lay people' do not fully understand the value of a stratigraphic context.

Have you seen the AIA's educational video on archaeological shoebox simulations in the post I made previous to this one? I understand the video and simulation is designed for grade-level classrooms and so I do not mean to be insulting, but it is actually a good and straight forward summary, in plain English, on the use of stratigraphy and finds in stratigraphic contexts.

Archaeological sites are composed of layers, which we call stratigraphy, and typically it follows that the layers closest to the surface are the most modern and the deepest the oldest. These layers can only be detected by careful digging and watching for changes in the texture, composition, and/or color of the soil. Usually the topsoil is often disturbed by modern activity and surface finds often provide the indication of an archaeological site and cultural layers below. Objects found within these layers can be associated with one another and with historical events.

The hurried and non-scientific digging destroyed the stratigraphy around the find and that cannot be reconstructed. If the stratigraphic context had been preserved, we might know something about the deposition of the hoard, e.g. was it buried under a floor, did it fall out of a wall etc. The stratigraphic relationship and other associated finds might also allow one to determine when and what historical event led to the deposition of the hoard - often times the coins themselves are not enough to do this precisely.

Inventories such as the one you mention (Thompson) can be useful for certain studies, but are limited in the lack of "closer contexts." In fact for many of them the general find spot is not even known, but rather "reported to be from XYZ." If you are interested in the value of close contexts and coin finds and hoards, I would recommend the forthcoming collection of essays I cited in the present post. One essay in this book discusses the essential importance of examining coins within their stratigraphic features and in conjunction with other finds, and even uses the example of certain well-excavated hoards. There are other essays examining hoard evidence more specifically, one of them in German. Another examines Medieval hoards finds from various contexts.

In addition to examining coins in stratified contexts, it is also necessary to examine them more generally from a site - i.e. the whole series from a site. Therefore, with coins it is necessary to both examine the stratified context and the entire coin series from a site. Numismatically, the stratified context provides a sample of the coins in circulation in a particular region in a distinct period of time among other things of archaeological and numismatic importance.

Incorporating both approaches is essential and both answer different types of questions. One may argue that close context is even more important with hoard finds as this gives insight into the deposition of the hoard itself and events affecting its composition, and in many cases, the motivation behind a particular hoarding event.

I hate to keep referring you to this collection of essays but, it would take me along time to explain everything here. If you are interested in previous studies and the information gained from studies of coin finds (and hoards) in stratified contexts, two very good and thorough treatments are M. Peter, Untersuchungen zu den Fundmünzen aus Augst und Kaiseraugst (Mainz, 2001) and a very good one in English F. Kemmers, Coins for a legion:
an analysis of the coin finds from Augustan legionary fortress and Flavian canabae legionis at Nijmegen (Mainz, 2006). Kemmers' book is a thorough treatment of all coins from the fortress at Nijmegen as well as an examination of the coins in stratified contexts, which she then uses to reconstruct a picture of the coin supply to Nijmegen and the logistic behind it. Towards the end of the book, she shows how attention to the finds in context indicates how at least in the Flavian period the Roman state was supplying the soldiers there with coins bearing designs relevant to the military - something which we could never have discovered without the context. Kemmers' book breaks methodological ground on many fronts.

1) I think I might have addressed this already, but in case its not clear I'll say stratigraphy is always of value since this allows us to associate specific finds with other finds. It seems many people seem to think of archaeology as simply going out to get the goodies to bring back to the museum. This is an antiquarian perception. That was the case in the 1800's, but not anymore. Archaeology uses the scientific method and in several ways the methodology could be compared to the examination of a crime scene: using "contextualized" material evidence to recreate past events.

Stratigraphic excavation allows us to examine the detailed history of a specific area. For example, stratigraphy could allow us to know that a certain building was first built in wood in about the 20's BC, then we might have several layers showing raising of floor levels, we can date these by the finds in the stratigraphic layers. Stratigraphy may also tell us when this wooden building burned, and often why it burned and in what year or general time frame - often times these are associated with specific historical events and the site itself can then be incorporated into a larger regional or world history. We can also see how the building changes over time, is modified, rebuilt, etc. The finds themselves and their relationships with each others tell us about the people who used the structure and the objects themselves. The utility of coins, whether or hoards or single finds, in all of this I think is quite obvious. Again, the methodological essays I mentioned earlier discuss such things in greater detail and with several specific examples.

2)I think I've touched on this. I can't think of anything specifically geared towards the lay person, but I'm sure the forthcoming methodological essays would be easy enough for you to get into. As the contributors prepared their chapters, they did so with the intent of making the approaches better known to a wider audience, which admittedly is mostly an archaeological and academic numismatic one. While you may not be a trained archaeologist though I think you could follow the argumentation of the contributions. Most of them are also in English, which is significant since, the study of coins in close contexts is more prominent in Germanic scholarly literature.

I would also recommend the two books I mentioned - definitely go with Kemmers' if you don't read German. It can be tedious in some areas, but it is well-researched and has copious citations and an entire section explaining methodology, so it should also provide some indications. I believe the ANS library, which I think you have access to, should have Kemmers' book and no doubt will get the collection of essays when it comes out.

3) The PAS only catalogues finds (many if not most of them surface finds) and so the PAS database is a useful tool for regional comparisons and certain quantitative studies, but not for the study of objects in close stratified contexts. As I stated earlier both methods have their benefits and answer different questions and so it is necessary to do a bit of both in site-specific studies. I do not know how the process works in the UK, perhaps Barford could provide more information, but a very small excavation like the one that officials did after the discovery of the hoard may be published as a short note in an archaeological periodical. Find reports from larger excavations generally have an entire section devoted to the stratigraphy and their relationship to finds which are then analyzed subsequently in the chapters of the book discussing specific finds, structures, site history, etc.

Sorry for the long response, but you raised some very good and vital questions and I still can't respond to each one of these completely without killing my Sunday! I hope it helps anyway.

All the best,


Adding a tiny post to this one... which I agree with .. is the point I always plug... rather than saying people are unethical (per ce) - one must take the time and effort to explain ... I too have looked at the AIA vids.. and teachers downloads.. I like them so much I will be using them myself to teach strat... and recently at a metal detecting meeting I spent 2 hours explaining and showing how recording works, etc etc... in a non-judgemental way. I am after all an archaeologist, its what I do, non archaeologists just need to be given the reasons and remove the misconceptions. In the end its all about training, talking and time. Money well spent to gain the vast amount of info that could be available. Railing against people in my experience never works, working with those who are willing to listen does. Some people have jumped on this as an example of looting or destruction… I see it as people who will (if somebody takes the time to explain) be more aware the next time. After all, they did get an archaeologist on board within 24 hours… (They did not have to.. ! But they did.) Lets look to a positive move forward, rather than a negative put-down..

Many thanks for your well considered and thoughtful post.



Thanks for this. Certainly this is the point, we need to inform the public better about why context is important.

The PAS scheme is meant to preserve context instead of licensing disturbances as some perceive it. We are concerned about the loss of information, but if metal detectorists in Britain know why context is useful maybe, as you say, next time will be different.




The comments for this post are very illuminating regarding context. Your explainations are very clear.

Could an archaeologist go to the site where these coins were found and begin a dig? Or are there rules protecting the property. If someone could go to the site, how much information is lost from the digging of the one meter (3 foot) deep hole? Could you use the earth around the perimeter of the hole to make assumptions about the earth that used to fill the hole?

I have really been enjoying picking through your blog! Very educational!



Hi Bill,

I'm not sure what the laws are for excavating in Britain, but I assume that the State or County archaeologists could go and excavate further, and in this case they did and determined too much damage had been done to tell anything useful.

Typically excavation permits are granted to archaeologists by a state agency and local archaeologists, employed by the state, can do small survey excavations like this as needed and then get permits for larger projects if required.

One meter of depth is well below the plow line of most plows and so the damage can be pretty extensive. As the archaeologists who investigated afterwards indicated, they could not really say anything about the context of this hoard because of the damage done to the stratigraphy.

It is difficult to speculate on what damage was lost because the close context is so crucial. In theory, one could expect the stratigraphy to be similar in adjacent areas, but you really can't "reconstruct" it as such, because its not always straight like a layer cake and there are often intrusions. For example, if the stratigraphy had been preserved and excavation was done properly, we might have seen that the hoard was part of an intrusion (i.e. someone dug later into the lower level to bury it), but that is just speculation since it is lost.

Stratigraphy is not just about the dirt layers either; it also includes finds that are associated with a specific layer and layers can also be quite fine and thin. Therefore, the archaeologists may not - and in this case they weren't - even able to associate it with a stratigraphic layer. I hope this is clear.



Yes, thanks, that is very interesting. I hadn't thought of intrusions before when considering the stratigraphy. I feel like I am taking a college course from you sans tuition. Thanks!

With this case at least the depth of the hoard is roughly known. Having information is more helpful than if it was never found or not reported. At least some knowledge was gained, right?

Incidently, and on a personal/anecdotal note: My father used to dig up our back yard (in Massachusetts - USA), to find old blown glass bottles. Apparently the yard was a garbage dump over 120 years ago. He found some really interesting and beautiful bottles. I don't think you would consider his activity unethical, but where does the line get drawn? Older then 200 years becomes a archaeological site? Just curious.

Thanks, I really enjoy reading this blog + your responses to comments.


Hi Bill,

Yes, at least some information was preserved, i.e. we know the general find spot of the hoard. However, how that hoard functioned within the rest of the site, where exactly it was placed, and all of its associated finds (as would be defined via the stratigraphy) cannot be defined.

Questions like "why was it deposited" can never be answered because of the destruction and there are great limits to general geographical studies, especially when the function of the hoard cannot be defined.

When the PAS was instituted I don't think they had in mind people violating archaeological contexts at this level; instead it was a scheme meant to promote the reporting of decontextualized finds in the plow levels.

If you can read German, there is one monograph on a hoard from Neftenbach, which demonstrates how much information can be recovered from a hoard when it is context is preserved: Der Münzhort aus dem Gutshof in Neftenbach (1993). Not only was the context preserved, which shed light on the history of the villa in which it was found, but also the excavators took it to a lab still undisturbed in the vessel. The "micro-context" of the hoard was thus also preserved and the coins were slowly "excavated" in the vessel with the aid of an endoscope. The results of this micro-excavation were striking and indicated the hoard was deposited in the vessel over a period of time (a savings hoard) and even traces of organic material were preserved indicating how it was stored and sealed.