Tuesday, November 25, 2008

Recording Wear and Corrosion on Ancient Coin Finds

In searching coin reports from excavations one can find analyses and lists of highly varied quality. Much of this, of course, can depend on the academic tradition from which the numismatist comes. In worst case scenarios, a "coin report" may not even have a list of the coins and may only provide a vague summary of types and rulers represented. In addition to a detailed analysis of the overall series from a site, and also of coins in stratified contexts, the most useful coin reports will present the numismatic data in a complete and detailed manner through a list or catalogue. These lists should include reference numbers or a description of the coin if an exact reference cannot be provided, dates/date range, die axes, diameter, weight, etc. One aspect of coin finds that is often overlooked, however, is wear and corrosion.

Recently, Kris Lockyear commented on the need for more standardization in the publication of coin lists from archaeological sites, citing the basic elements that should be included in an average coin list (K. Lockyear, "Where Do We Go Frome Here? Recording and Analysing Roman Coins from Archaeological Excavations. Britannia 38 (2007), 211-224). One of his proposals was to begin an international database for the cataloguing of coin finds which would necessitate certain types of information to be entered. Some regional and national databases and corpus-making projects have been underway for sometime. Apart from the basic information listed above, one of the desirable things on a coin list he mentioned is a measure of the wear and corrosion present on individual coin finds. Some of the things that wear may inform field numismatists about is the individual lives and circulation of coins.

In 1976 Alan Walker commented on the value of studying wear and corrosion on coin finds (A. Walker, "Worn and Corroded Coins: Their Importance for the Archaeologist," Journal of Field Archaeology 3.3 (1976), 329-334). He detailed the need to pay close attention to corrosion on a coin. Although a coin may be nearly unidentifiable because of corrosion, one may still be able to glean some information from it; even coins that are heavily corroded could exhibit few signs of wear. He then surveyed the value of studying wear on coins, which is useful for establishing the chronological sequences of coins that are not dated - an important aspect of Greek coin hoard studies. Wear patterns on coins also provide insight into the duration of time in which they were in circulation and can also be used to examine external issues such as coinage reforms, demonitization, etc. He lists some examples where the study of wear on coin finds proves essential for the understanding of site or the coins themselves. He concludes:
"In order to obtain all the information that numismatic material can supply, archaeologists must be aware of the ramifications of wear and corrosion. The examples cited above show that even exceedingly worn or corroded coins may have a real significance: they can either give a precise date for their contextual assembly or illuminate the economic and political circumstances of their period of deposition."
Certainly, as Lockyear, Walker, and others have pointed out, wear and corrosion deserve more attention in contextual studies and the analyses in excavation reports.

The market already has a general method of "grading" coins in terms of general "eye-appeal" which may be affected by wear and corrosion. The grading of ancient coins is naturally much more arbitrary than grading modern coinage, for which there are entire firms are devoted grading, certifying, and "slabbing" coins in plastic containers. In recent years there has been some demand for slabbed and professionally graded ancient coins and a new venture devoted to the slabbing of ancient coins has recently been a hot topic in online ancient coin collecting boards. Nevertheless, the relatively loose grading systems used in the ancient coin market cannot be applied well to excavation finds where one is more interested in technical aspects than aesthetic concerns. Additionally, market grading systems do not differentiate well between measuring wear and corrosion.

Recognizing the need for a system for coin finds, one was devised and published in 1995 (S. Frey-Kupper, O.F. Dubuis, and H. Brem, "Usure et corrosion. Tables de référence pour la détermination de trouvailles monétaires / Abnutzung und Korrosion. Bestimmungstafeln zur Bearbeitung von Fundmünzen," Inventar der Fundmünzen der Schweiz 2, supplement (1995) (I could only find the French text available online). The text of the guide is in both French and German.

The guide provides a system for gauging corrosion and wear on both sides of a coin using a numbering scale from 0-5:

Wear (W). In German, Abnutzung (A); in French, Usure (U)
W 0 uncertain
W 1 very slightly worn
W 2 lightly worn
W 3 worn
W 4 very worn
W 5 extremely worn to completely worn (flat)

Corrosion (C). In German, Korrosion (K), in French, Corrosion (C)
C 0 uncertain
C 1 very slightly corroded
C 2 lightly corroded
C 3 corroded
C 4 very corroded
C 5 extremely to completely corroded

"0" is used if for some reason one aspect is completely indeterminate, e.g. a coin is so badly corroded (C 5) that one cannot determine how worn it is (W 0). A coin find report using this system might thus include a set of numbers such as W 1/2, C 3/3 in order to relay information about wear and corrosion. Numbers on both side of the slash refer to obverse and reverse respectively. In French or German the abbreviations would of course be different. I first became aware of this system when I began working with Fundmünzen der Antike in Frankfurt, which has implemented the system on new entries to its finds database. Although I did not use the system when I began identifying and analyzing the coin finds from Yotvata, I did use it for the last couple of seasons of coins after I learned about it.

It seems to me that the use of this system in coin lists is rather efficient and exacting, but still compact enough to incorporate into a coin list. The system has been accepted in a number of find publications on the continent, but I am not aware of its use in the English/American scholarship or coin find publications. If any readers are familiar with it, have comments on it, or have incorporated it into their work, I would enjoy hearing from you here or privately.

Certainly we are in need of preserving as much detailed information about archaeologically recovered coins as possible and we ought to make as much of that information available as possible for future comparative research. The study of wear patterns on ancient coin finds ought to provide further insight in future studies and is an area of Fundnumismatik with little detailed exploration at present. In my opinion, these guidelines seem to be a good way of enhancing our corpus. Feedback welcome.

Updated 28 November 2008: After discovering the French version of the corrosion/wear guide was available online, a link was inserted above.



This is very interesting. I will track down Walker's paper.

Do Frey-Kupper, Dubiis and Brum provide a quantitative definition of their categories? I have wished for better ways to describe corrosion on coins, especially for the purpose of detecting forgeries. I have Caley's book _Analysis of ancient metals_ but lack the chemistry background and equipment to follow it. Having just five categories seems limiting. I feel like the particular corrosion chemical products could tell a lot about the soil conditions (which might be especially handy for market coins whose findspot isn't known) and the texture of corrosion can tell about the quality of metal at striking.

Caley and others made very nice pictures by cutting coins in half, polishing them, then treating them with acid to see the metal grain. It might be possible to non-destructively do such things today. Can those pictures be summarized in a useful numerical way?

I'd also like to see a quantitative standard for measuring wear. Perhaps in the future when 3D scanning is possible something can be calculated automatically without taste entering the picture? Such a technique could also look for odd/faked wear patterns.

What is the current opinion of destructive testing of coins? I remember talking with Peter van Alfen at the ANS about William Campbell's book from 1933, _Greek and Roman Plated Coins_. (Cambell cut fourees to look at the inside.) Peter frowned at the idea of trying that today. I know some countries like Israel prohibit destroying antiquities. I find these attitudes surprising; as a technology person it seems ideal to take things apart to see how they work.


Dear Ed,

I'll email you a copy of Walker's paper.

I'll summarize their more detailed descriptions, which they break in 4 categories: Republic/Imperial, 3rd century and 4th century, Early Medieval, Medieval and Early Modern.

On wear they describe 1 as all details visible, and the highest points of relief are only lightly worn. 2: highest areas of relief are flattened; hair, letters, and other such details are not sharp. 3: Most details are blurred. Lines are no longer sharp. 4: The coin image is recognizable only in the form of a silhouette, most of the legend is unrecognizable/illegible 5: essentially worn flat.

They have three pages showing these descriptions with Roman Republican/Imperial coins, 3rd and 4th Century coins, and Medieval coins. Generally, the classification is more lenient with later Roman and Medieval coins since fabric and striking pressure are more likely responsible for some of the loss of details in later coins. The illustrations are meant to serve as a guide for using the system.


C1: The surface of the coin is intact and no remarkable corrosion is visible. 2: The surface is only partly corroded or only lightly covered in pits. 3: The surface is mostly corroded or the whole surface is covered in middle/large pits or encurustation. 4: The whole surface is heavily corroded and altered through pitting or encrustation; thin coins are often broken or eaten through. 5: The surface is completely corroded or encrusted; the form of any image is indiscernable; thin coins are eaten through and fragmented.

A colleague has also informed me that the guide is available online and so I have put a hyperlink in the main post now.

Regarding destructive testing, it does still take place though it is not as destructive a before. Standard archaeometric tests on coin finds and museum coins are now conducted by micro-drilling from the rim into the core of the coin. The drill hole is only 1/2 mm in diameter. Interior metal is necessary since the surface of coins have been altered through their exposure to soil and air over time, but the use of a small drill preserves the coin and is hardly noticeable. While one might expect museums to be less willing to subject their objects to such tests than archaeologists would be, I heard a story about the British Museum which partnered with a German university to do some testing on BM coins. The university was only doing thermoluminesence testing on the surface and the BM said they would pull out of the project unless they drilled the coins because studying the surface would provided no valuable information. It does seem be standard practice to employ these less invasive and less destructive techniques in such studies.



Thanks for the reference to the wear and corrosion-grading guidelines. The tendency to lump worn, corroded and misstruck or otherwise blank coins under the heading "illegible" has led to serious scholarly misinterpretations, especially among the late Roman/early Byzantine "minimi" - I try to account for this in the publication of a hoard of them from Sardis, in Revue Numismatique 163 (2007). In this case, the fact that the coins were neither much worn nor too corroded, but still illegible is key to the mint practices of the late fifth century. I didn't know about the guidelines when I wrote the article, but I will be sure and use them in the future, so thanks for publicizing them.


Thank you for the comment. Certainly regarding minimi and other later coinages, close attention to wear and corrosion can be especially useful. I recall seeing Sam Moorehead give a paper on attributing minimi a couple of years ago at RAC/TRAC in London and he talked about some of the issues you raise regarding wear and striking. As I'm sure you are aware, it is unfortunate that so many minimi fail to be recognized or are ignored by numismatists working with coin finds.

Thank you also for the citation to your publication in RN, I'll take a look. By the way, while I have the opportunity to say so, I've found your book Neokoroi: Greek Cities and Roman Emperors very useful for some of my dissertation research.

All the best,