General Questions

How do I find my way around in the Celtic Coin Index Online?

See the navigation page.

How do I use the Celtic Coin Index Online in my research?

See the using the Index page.

What do the abbreviations in the text fields mean?

See the abbreviations page.


Can I use wildcards in my searches?

At the moment, no. The current search technology does not support wildcards. Wildcards are planned for future updates, however.

Why am I having trouble finding results for my search terms?

Look for the simplest answers first. Look through the following checklist:

  1. Have you typed the search term in correctly? Check for typos and spelling mistakes. Misspelled words in searches are common, so it is worth taking care: you will find no "minim" if you search for "minum."
  2. Are you searching for something that should be in the Celtic Coin Index Online? If you are looking for non-Celtic coins, you will not find them here. Currently, only the British Celtic tribes are online, so you will not yet be able to find Gaulish or "Continental" issues in your search. Check the main page for the latest news on the Celtic Coin Index Online's currently available records.
  3. Is there a different term you could be using? You may be searching for a specific term that is not used in the Celtic Coin Index: try using alternate terms to see if you can find what you are looking for. For example, potins are listed as bronzes; the Coritani are more correctly referred to as the Corieltauvi; and, when the Gaulish issues are online, the Curiosolites or Coriosolitae will be referred to as the Coriosolites, but the search term would be "Coriosolite".
  4. Is this a coin that has not been reported to the Celtic Coin Index? Obviously, then, it will not be found in the database. If you are aware of a coin that you do not believe has been reported, kindly contact the Keeper of the Index, and arrange to provide the information. Private collectors and metal detectorists can be assured that their identities and exact find spots can be kept confidential.
  5. Can you find the item by navigation without using a search? Try this by clicking on the "coin records" button, and following the links through the coin regions, tribes, and Van Arsdell numbers, to see if the information you want is there.
  6. Are you trying to find a record you accessed previously? You may be finding it difficult to remember the exact terms used, so try experimenting with a less restrictive search may enable you to quickly scan through the thumbnails and descriptions to find the specimen again.
  7. Have you found a record or a publication that does not show up in the search? If so, please email the web site maintainer with the exact details of which search was used, and the terms searched for, and we will determine if this is a problem with the search engine, or whether some other factor is at play, and make any necessary adjustments.
  8. Have you tried variations that may be remnants of the original Oxford data formats? Van Arsdell numbers are often recorded in a decimal form, so that 1205-1 might appear as 1205.01 (while we have made every reasonable attempt to correct major occurrances to Van Arsdell's original form, some notes fields will still contain the decimal form). Also, the original CCI numbers had a decimal after the year, like 61.0001, which is now simply 610001; however, cross-references to various other coins in the notes may still include the decimal. As well, various abbreviations are used in the notes, and while some of these can be easily understood (such as BRM for "British Museum"), others may not (such as "R'borough" for "Richborough"). You can look these up on the abbreviations page, so you can use the abbreviation in your search, to see if there are records you may have missed in your first try.

Why do I have problems getting results with multiple searches?

There may be no coins with exactly all the parameters you have chosen. Eliminate the most restrictive terms first, and see what results you get. Metal analysis is the most restrictive, as only 2% of all coins have metallurgical data; present location can also be restrictive, particularly for smaller institutions. Experiment with less restrictive searches to get a sense of how many coins meet the criteria you are interested in. If the coins you are looking for can be found by Celtic Tribe and Van Arsdell number, consider navigating through the Index to see whether the other criteria exist in the records. If you find at least one record that should have shown up in the search, contact the site maintainer with complete details of all the criteria you chose, and the coin number(s) of the records that should have shown up, and we will look into the matter.

Why can't I find any potin coins in my searches?

Unfortunately, potin, a high-tin bronze alloy, is not a term used in the original Celtic Coin Index records at Oxford. Potins have been recorded simply as cast bronze, much as Northover does. Theoretically, you could either 1) search for issues you know to be potins, or 2) look for cast bronzes where a metal analysis has been done, to see which ones are potin. However, the second choice would be unfruitful: while only about 2% of the specimen records in the Oxford CCI data show details of metal analysis, none of them are cast bronzes. Van Arsdell mentions the high-tin bronzes found in Kent, thus catalogued as Cantiian (Van Arsdell, Celtic Coinage of Britain, p. 7), and many Trinovantian/Catuvellaunian cast bronzes are high in tin. Northover has examples of high-tin cast bronzes among the Thurrock, Snettisham, Takely, Stansted, and Kelvedon types, as well as among some Durotrigan coins.

As to when an alloy should be classed as potin, we have not found any conclusive, authoritative definition. According to Marc Breitsprecher, potins can contain as much as 25% tin, "and quite often an appreciable percentage of lead as well". Greater amounts of tin lower the melting point of the alloy, making it easier to cast. The Liberty Bell Memorial Museum mentions that the properties of bronze change appreciably above 10% tin, becoming harder, stronger, and more resistant to corrosion. A site on Russian Bells mentions that the strongest bronze alloys contain tin and a small amount of lead, adding that such alloys are "preferred for intricate ornamental castings." Considering the properties that would be ideal for casting coins, it seems reasonable to consider potin to be a bronze alloy containing approximately 10%-25% tin.

Why can't I find any Coritani coins?

The coins of this tribe are now called "Corieltauvi", after new, more complete inscriptions came to light. You can find any of the coins you would be looking for under the section on the Corieltauvi.

Why can't I find any Gallo-Belgic coins?

Currently, only the British issues are present in the Celtic Coin Index Online. The Gaulish issues, including the so-called "Gallo-Belgic" coins, will be processed separately. Numismatically, Gallo-Belgic coins are simply Belgic coins, and cannot logically be separated from other Belgic issues. All of the Gaulish tribes will be organized differently from the British issues, according not only to region, but also by the definitive study or studies done about each.

Reference Data in the Index

Why are are the grid references given in the Celtic Coin Index Online incomplete?

Many of the coin records have been provided by metal detectorists who, quite understandably, want the exact find spot kept confidential. In other cases, coins have been found on private land, and the owners are reluctant to have their property trespassed upon by individuals who have not received permission.

However, many grid references are from published excavations and other finds where confidentiality is not a concern. The provenance data will usually indicate where this is the case. If the provenance publication is given, the full grid reference can be found there. If the information is not available, and it is necessary for you to obtain this data, contact the Keeper of the Index, and if the exact reference is not confidential, he will be more than happy to provide further detail.

Why are some Van Arsdell Reference Numbers not given fully in the Celtic Coin Index?

There is no information available as to why some coins are fully catalogued by Van Arsdell number, and some have only the root number.

We would suggest you look at each specimen, and make your own judgment regarding whether there are difficulties with condition or viewability of vital design elements, or other issues at play. As the Celtic Coin Index Online is meant to encourage further inquiry and research, the data, including the cataloguing, should not be seen as final, but as part of the ongoing refinement of the study of the subject.

Why are some Van Arsdell Reference Numbers given in the CCI not found in Van Arsdell's "Celtic Coinage in Britain"?

There is no direct information from the cataloguers regarding this matter; however, some are numbers Van Arsdell reserved for coins assumed to exist but not yet known when "Celtic Coinage in Britain" was published in 1989, and others are numbers Van Arsdell has given in later publications.

Why are the British Museum catalogue numbers in the Celtic Coin Index different from those in the published catalogue?

The short answer? These are Derek Allen's catalogue numbers, not Richard Hobbs's.

Most of the entries for the British Museum (BM) in the Celtic Coin Index (CCI) card catalogue were added in 1968 and 1969, well before the publication of Richard Hobbs's catalogue in 1996.

For this reason, most of the BM numbers in the CCI tend to be the old catalogue numbers started by Allen, rather than Hobbs's.

The Keeper of the Index has expressed his intentions to add Hobbs's catalogue numbers to the Provenance data, over time.

What is the difference between "Catalogue References" data and the Classification System References, such as Van Arsdell, Mack, and Allen numbers?

Essentially, the Classification System References refer to particular coin "types" (though each system groups things according to different criteria, and at different levels in a hierarchy), whereas the "Catalogue References" refer to unique numbers given to an individual specimen at various times in its history.

This is where any individual specimen number is recorded: whether a unique number when excavated, a unique (not group) accession number when taken into a public collection, or a museum or collector's catalogue number, or a unique specimen number when published in a study. Sometimes, references in literature are to such numbers: we think it is helpful to track these unique numbers, as it helps to avoid confusion among similar specimens, and is an aid to building a more accurate history of the specimen.

What is "Unknown Die Study A"?

Die numbers are recorded in the Celtic Coin Index in the original die number fields, but the study they are based on was not recorded, and knowledge of its identity has been lost. "Unknown Die Study A" is given to differentiate these numbers from other die numbers based on known studies. Die numbers are also found in some notes fields: these are also unidentified, and would be recorded as "Unknown Die Study B." Some of these notes may refer to more than one study. When the die study sources become known, their names will be duly recorded. If you have conclusively found the source of these die studies, please contact the maintainer of this site and include all relevant references and the evidence and reasons for your attribution(s).

Structure of the Celtic Coin Index Online

Why are some images missing from the Celtic Coin Index Online?

Notably few images are missing from the Celtic Coin Index Online. With some minor exceptions, we have all the images for the coins catalogued from 1961-2001.

Coins that have no image on file at Oxford display the "No Image Available" graphic, so that you will know not to expect that coin to have an image. To provide images as rapidly as possible, after the images of the first few years, we have processed the images to standard dimensions and file size, and put them on the site. Any further image enhancement will need to await further funding.

Why do the pages showing Van Arsdell numbers for each tribe have so many images missing?

On the pages for each tribe, showing the various Van Arsdell numbers available in the Celtic Coin Index, we have an "exemplar image" for quick visual reference to each VA number. Since we currently have only images for the coins catalogued in the earlier years, we often do not have images for the VA numbers in question. Even where we have one or more images, many of these are of specimens too worn to be of use. Previously, we were awaiting the rest of the images, so we might choose the clearest and most typical exemplar image.

Now that we have received the rest of the images, we are in the process of selecting the remainder of the Van Arsdell "exemplar images". This process will take more time than simply uploading images, as it requires careful thought and judgment.

What does the "No Image Available" graphic mean?

It means there was no image of this specimen on file at the Celtic Coin Index in Oxford when this online index was created. While it is possible that, in the future, an image may become available, without a comparison image, determining whether a coin is the same specimen is problematical. Even reported weights can vary, so without conclusive provenance information for that individual specimen, it is likely to remain without an image.

Why are there two levels of quality in the coin images?

One of the difficulties with the original Oxford Celtic Coin Index images are the large number that are of poor quality. While there are many fine photos that were taken by Mr. Bob Wilkins of Oxford's Institute of Archaeology, and excellent photos are also provided by some dealers, it is a sad fact that many coin photos have various problems. Often, photos have been crudely cut out and attached to index cards, making the background complicated and distracting; in other cases, the only available photos are halftone reproductions from auction catalogues and dealers' lists. Poor scanning and image conversion has left blurriness and many digital artifacts which make it difficult to see the details of the specimen.

John Hooker is in the process of perceptually enhancing the digital images, so the design on the coins can be seen more clearly for research purposes, without distraction. Improving the quality of the coin images is a major addition to the usefulness of the visual data in the Index, and is an ongoing part of the project. Enhanced images can be clearly identified, not only by their quality, but by the fact that the coin image is presented on a perfectly black background.

Why are there no Gaulish coins currently in the Celtic Coin Index Online?

We have put the bulk of the Celtic Coin Index online, which consists of the British issues. The Gaulish issues, which some refer to as "Continental," present special problems of categorization, as they cannot all be organized by Van Arsdell number, but must be sorted according to the definitive studies for each area. These records will be our next priority, should funding be available for this further step. The Gaulish coins will be presented in a parallel system to the British, with its own area maps, and breakdowns by study and types -- and this includes the "Gallo-Belgic" coins, which, numismatically, belong with the other Gaulish issues.

Why are the Van Arsdell pages not simply in order by Celtic Coin Index number?

Because the Celtic Coin Index number is an arbitrary number, assigned according to the year and order in which coins were catalogued, we did not feel that it was the most important factor in organizing the coin records. Considering that many tribes have numerous records without Van Arsdell numbers, we felt that a secondary sort by Allen number would group similar coins together within those pages.

The sort order is actually first by Van Arsdell number, second by Allen number, and finally, by Celtic Coin Index number. This means that within a tribe, each category, either "No Van Arsdell Number" or specific Van Arsdell numbers, begins with those coins which have not been assigned Allen references. Within those first records, the coins are in CCI number order. When and if coins have been assigned Allen references, the CCI order begins again.

While this may appear confusing at first, it groups coins together by meaningful similarities, rather than by the random order of CCI no. If you know the CCI number, and want to find the coin record, the simplest method is to look for it using the search page.

Why are the "reliability" fields no longer used in the full coin record pages for individual Celtic coin specimens, as they were in the Epaticcus prototype?

In the Epaticcus prototype, and the first phase of the Celtic Coin Index Online proper, the "reliability" fields were retained from the Oxford database.

The reliability relates to the quality of the data, in the opinion of the cataloguer. There is no standard system for assessing the reliability of reported information. We have observed that, over time, various cataloguers have assessed reliability in different ways: some appear to have used a 3 point scale, while others have used a 5 point scale; furthermore, comments regarding reliability often contradict the reliability ratings themselves.

Since these fields are based on subjective judgments that are not consistent from one cataloguer to the next, over a span of forty years, and considering that these fields do not represent any hard data, we have thought they were more confusing than helpful. We believe that future researchers cannot do better than to assess the totality of the data based on the criteria they think most pertinent.

Why is there so little data in the new data tables: Die Study, Typological Analysis, Publication Details, Catalogue References, Specimen History, and Metrology?

There are two basic reasons for this:

  1. The data is recorded, but is mixed in with other data. Much of the information specific to these new tables is found in various places, particularly in notes fields. In many cases, the information already in the Index is quite good, but is not to be found in one consistent place. Given funding for data cleanup and enhancement, the current information could be organized more systematically into these new tables. At the moment, a particular notes field may contain information on die numbers, condition, metrology, classification, specimen history, publications, or typological details. Breaking out this information into its separate topics would make it easier to use, and simpler to find on the page. Funding for such organization of the data has not yet been found.
  2. The data is not available, or has not been recorded. This kind of information may not have been systematically recorded in the past because there was no standard place for it on the index cards or in the Oxford database. Specific, detailed data fields explicitly encourage the recording of particular information. Since the Celtic Coin Index Online is based on Arethusa, Hooker & Perron's research database, it is built on the idea that data about the specimens is not static, but will continue to be added to. The Celtic Coin Index at Oxford already does this, to some extent, through the tracking of each specimen's history, though much available metallurgical, archaeological, and typological data is not currently captured in the database. Adding all available relevant information to the Celtic Coin Index Online would be another potential project, but there is currently no funding to do so.

How should the Typological Analysis fields be understood and used?

The best way to understand how these fields work is to see them in practice. To show how such work would appear, we have used the example of Derek Allen's excellent study on "Cunobelin's Gold" using the specimens available in the Celtic Coin Index. In this example, the "Common Name" and the "General Type" Primary and Subsidiary motif information are from Van Arsdell (which may or may not be the preferred choice for other series), the "Specific Die" motif information is from Allen, and the "Specimen" motifs and "Miscellaneous Notes" are from the Celtic Coin Index data itself. While there are some typological studies of Celtic coins of the required detail and specificity, much of this is study yet to be done: we hope access to the data will inspire scholars to do so.

More detailed explanations on the use of these fields will be provided in the "Help" files of Arethusa, when it is released.

History, Development, and News about the Celtic Coin Index

What is the History and Purpose of the Celtic Coin Index?

While the Celtic Coin Index (CCI) has seen many changes since it was founded in 1960 by Derek Allen and Professor Sheppard Frere, it has not changed from its original conception of a comprehensive index of all Celtic Coins found in Britain.

In the early 1980s, Professor Barry Cunliffe took over responsibility for the Index, and throughout this period, his research assistants continued to add previously unrecorded coins from various sources. In 1992, Dr. Philip de Jersey began the computerization of the Index; he has continued to keep and expand the Index, not only based on coin records contributed by the public and institutions, but by noting Celtic coins from dealer's lists, auction catalogues, and other documentation. In this way, not only are new coins recorded, but previously-recorded coins are followed, and their history built up over time.

Hooker & Perron have created this "Celtic Coin Index Online". The database-web interface was designed by Carin Perron, based on the ongoing development of her Arethusa database. John Hooker designed the banner and other graphic design elements, and has standardized the size of the original coin images and made thumbnails of them. He is also working, understandably more slowly, on the more ambitious project of perceptually enhancing and optimizing all the original coin images. Image enhancement enables details of the coins to be more clearly seen and studied, minimizing the difficulties caused by blurriness, poor contrast, halftoning, and digital artifacts, as well as eliminating distracting details from the background the coin is on (See "Why are there two levels of quality in the coin images?").

Lexi.NET took over maintaining the web site in 2008 and has put together an updated web interface using a relational database to store the data instead of static html pages. This will allow for easier updates of the online index as information becomes available. Lexi.NET are not experts on celtic coins, however, and, as such, cannot maintain the data itself.

Subsequently, Lost Wizard Enterprises (LWE) has taken over maintaining the web site. As with Lexi.NET, LWE has no expertise in the study of celtic coins and, thus, cannot maintain the data itself.

How has the online Celtic Coin Index project developed to date, and what are the plans for the future?

At this point, we have completed all the coin records from 1961 to the middle of 2001 for the British issues, which is more than 28,000 records. We are now awaiting up-to-date coin record data, which will take a bit more time to process.

Regarding the history of The Celtic Coin Index Online, we began this project with a prototype containing all 320 CCI records of the coins of Epaticcus, the Atrebatian king, which was up in January of 2001.

Unlike the Epaticcus prototype, which was organized by CCI number, we organized the new structure by coin region, tribe, and also by Van Arsdell Number. We continued with the plan of adding complete records for each year, working tribe by tribe. We began with the huge first year of the Index, 1961, which alone contains more than 1,600 records, starting with the Atrebates, and continued in this manner until the following tribes were up for 1961: the Atrebates, Cantii, Trinovantes/Catuvellauni, the Durotriges, and some unknown types, so at that point, we had only three tribes left to complete for 1961.

We saw that, 1961's large size aside, working year-by-year was proving to be too slow, so we reassessed our approach to speed up the process. While we worked on further development, we decided to make as many records available on-line for use as quickly as possible, so we put up shorter listings of the coins before the full coin record pages became available.

The next phase of development incorporated two broad improvements: the elaboration of fields on the coin record pages, and a speedier method of making the coin records available. Our refinement of the design of the individual pages is based on proprietary Arethusa data structures: in particular, we have added the ability to do a more thorough typological analysis.

Future plans include, as funding and time allows, adding the 4,000+ Gaulish (Continental) coin records: there are greater difficulties in breaking these coins up in a logical and useful way both geographically and numismatically than there were with the British issues, as there are different authoritative texts and reference number systems for different locales.

We may also add later enhancements, like distribution maps, and searchability for more complex queries than those currently available. We hope you find this site useful; if you have things you would like to see, please feel free to contact us.

How does the Celtic Coin Index Online relate to the original computerization project funded by the Leverhulme Trust?

Hardly at all. According to the Oxford site, the Leverhulme Trust funding ended in 1998, a year before Hooker & Perron offered to help the Celtic Coin Index move forward onto the WWW. Initial work was begun on the prototype, and funding applied for, in late 1999 and early 2000; when the funding was not forthcoming, Hooker & Perron agreed to put up a prototype, consisting of 320 coins of Epaticcus, to help Oxford obtain funding for the online project: the prototype went up at the end of January, 2001. After this, we have continued into the full project, adding coins catalogued in 1961, working on deferment in anticipation of future funding.

The Celtic Coin Index Online project has, of course, used the existing digital data and images rather than work with hard copy. The raw data was entered from the original index cards in digital form during the initial 1992 computerization pilot program funded by Oxford, and during the following five years of the Leverhulme Trust grants, the last two years of which were devoted to digitizing the images by scanning old photos and catalogue images. The original project to computerize the Celtic Coin Index was accomplished internally at Oxford, and enabled the Keeper of the Index to process the increasingly large numbers of finds reported each year, and enter specimen data into an Ingres database built by Dr. Gary Lock, based on consultation with Dr. Philip de Jersey and a number of other individuals. This database does not contain the images, so does not display them alongside the data itself.

The Celtic Coin Index Online is based on the proprietary relational database, Arethusa, an outgrowth from earlier original Hooker & Perron research databases, applied to the Internet Coin Project in 1998, and in further development by Hooker & Perron since then. Arethusa does display the images with the data, but this is not the only difference. The original Oxford CCI data files have been indexed, normalized, and reorganized into the Arethusa structure, to create more flexibility in future recording of such things as provenance, design elements, associated finds, and more complex metallurgical analyses. The current web pages do not fully demonstrate this potential for two reasons: 1) much of this data has not yet been recorded in the Index; and 2) Arethusa is not primarily meant as a static recording system, but as a research tool, so that as information is analyzed, findings can be recorded as part of the research.

The original computerization of the Celtic Coin Index has succeeded in expediting the management of the Index from within the walls of Oxford. The Celtic Coin Index Online Project is further extending this function, in the original democratic spirit of Derek Allen and Sheppard Frere, to the entire Internet community, providing free access to the information, and the ability to search and study it in different ways, to everyone on the World Wide Web.