Forum Archaeologiae - Zeitschrift für klassische Archäologie 74 / III / 2015


This paper is an attempt to re-think the ideals of one of the most impressive series of material publications in the fields of archaeology and art history: the Corpus Vasorum Antiquorum. Model for a range of other efforts to bring together material in a standard form of publication, but at the same time with only a specialist appeal. We can bring this project into the future by transforming it into an open internet platform, integrating with other archaeological projects and truly opening up the world of Greek painted vases to the specialist and non-specialist alike.

History and development
The Corpus Vasorum Antiquorum, set up in 1919 by the extraordinary effort of Edmond Pottier, had (and has) a very broad scope and practical aims. A standard of publication of all ancient vases would create a solid base for research, where before publications of ceramics were scattered and unsystematic. This reveals an underlying positivist attitude: it is possible to describe this section of the world in a systematic and comprehensive way and further understanding by presenting large amounts of data.
The series has developed, and standards have changed. Pottier’s complicated and intricate subdivisions were abandoned as were the loose plates, the stamp-sized photographs with two lines of text per object were replaced by full-sized images, profile drawings are now the standard, and the descriptions increasingly come with large amounts of parallels and interpretation. For larger collections it is possible to make thematic CVA’s, amounting to near-handbooks on specific shapes and techniques. While Pottier also wanted to include Near-Eastern pottery, the series is now limited to Greek and related painted ceramics; it is not to be a primary publication for excavation finds.
The most recent development is the CVA online [1]. The scanning of out of sale volumes and the integration of the data into the Beazley Archive Database (BADB), has made the series much easier to use for aggregated datasets. Where before it would take months to make a catalogue of all the chariots with a lyre player in the black-figure technique, it now is a matter of minutes to make a smart query (so probably including ‘Apollo’). But with this integration, it also shares the weaknesses of the BADB.
I do not want this paper to be a Beazley Archive bashing, because they did wonderful work pioneering the integrated publication of vases and photographs. Part of the problem, as always in our field, will no doubt be lack of funds, so it is unfair to criticize on that. We also must acknowledge, however, that the rigid organization and lack of openness stands in the way of the development of a database of ancient ceramics that is truly useful for specialist and non-specialist alike. The weaknesses are enhanced for the CVA online part, as only a few select fields are included in the database. I will come back to the shortcomings of the present-day digital representation of our material.

It is clear that the CVA series has its strengths. The standard format, the high standard of scholarship, the exquisite (albeit quite expensive) photographs and printing, the fact that it is written in accessible languages make it the best fulfilment of the museum imperative of studying and publishing the collection. It also offers the possibility to create specialist knowledge; it is almost impossible not to become a specialist when writing a CVA. The series also has serious shortcomings. The rather austere format and rigorous standard of scholarship makes the series useful for specialists only. It is difficult to aggregate data from the publications or use the photographs or the drawings for one’s own purposes. It is a hassle to compare material across the hundreds of paper volumes. It is an expensive project with a very limited diffusion (only a few hundred copies are printed of each CVA).
The question must then be: is the CVA an outdated medium? The little material available from regular excavations has been supplemented by huge amounts of pottery from a wide range of contexts. The chronological framework, still patchy and tentative in Pottier’s days, is robust, detailed and reliable nowadays. A database, preferably on the internet, is a much better way to make all sort of selections and grouping than the loose plates of the first CVA’s. In short, is the project a corpse rather than a corpus, as I so bluntly put it in my 1998 paper [2]? Seventeen years ago I answered in the negative, with the provision that the CVA reinvent itself in digital form. The present paper revisits the subject.

Crucial questions
Two questions are central in this paper:
How can we make the work on CVA’s useful for a much wider group?
And how can we make production of CVA’s affordable, more accessible and still maintain a high standard?
I will argue for publication as an integrated, open, universally accessible database with a high standard of information. The CVA needs to transform from project to platform.
Let us first turn to the wider audience. In the first place, the wider audience will be our fellow archaeologists. The excavators of sites around the Mediterranean often finds but few black- or red-figure fragments, but in spite of low numbers these are important for dating purposes, for linking the site in the wider Mediterranean network, for local appreciation of Greek iconography and ceramics. Likewise, the student of ancient history, religion, economy and iconography will all benefit from a more accessible corpus. For the general public, museum rooms full of vases are rarely popular, but in principle the very versatile world of vases is excellently suited for educational programs at schools and museums. In the Netherlands, there is a call for ‘valorization’, giving back to society some of the fruits of research funded for a large part by tax payer’s money, and this demand does not seem to be limited to Holland. In reaching out to a broader audience, the CVA project will respond to that call.

What can we do to make the work we are doing more useful?

  • Data integration
    Both the CVA online and the Beazley Archive Database stand in splendid isolation. They do not connect with for instance context information or other finds from the provenance area, with similar iconography on other media, with databases of fabric analysis. The Pelagios project shows that this does not need to be the case. In this interface, all available information (coins, terra cottas, objects from some public collections, literature) is linked to specific locations; in their own words, “its aim is to help introduce Linked Open Data goodness into online resources that refer to places in the historic past”. The original interfaces of the contributing databases are maintained, which offers a somewhat uneven impression, but this may be cured by further application of Open Data principles.
  • Look at the needs of potential users
    The CVA project generates large amounts of data. If these were well represented in a database, they would offer a diversity of information that would make its use on multiple levels possible. Here is where the shortcomings of the CVA online/BADB are most striking. The dating within 50 year brackets is applied unevenly; the broadness of the dating makes it unusable to all but the plainest of purposes, and the uneven application renders it unreliable. The profile drawings do not have a proper place, the provenance, if indicated at all, is very general (the location, not the type of context or e.g. the precise find spot). There is no check on the quality of the scholarship and no possibility of discussion; it is a closed system, not necessarily with the best possible information. The interface is made with the specialist in mind, and also for this group it could be much better and more useful.
    There are a few databases around which are more aware of the needs of a broader audience. The old Southampton Amphora Project for instance facilitates identification of amphora shapes. From the field of numismatics there is the Sylloge Nummorum Graecorum, which allows the user to select in a way that one does not have to be a numismatist to make sense of a coin. A more recent excellent example hails from Vienna: a database of amphora fabrics, FACEM. The user interface shows the advancement of internet technique; the fabrics are presented in a way that the researcher in the apothiki can readily identify the fabric of the sherd the students brought in from the field. We may not have the means to develop an interface that would match FACEM for our material, but we could make the data available in a way that others might construct useful guides.
  • Make it possible to define one’s own queries
    In most databases, the questions you may ask of the material is restricted those you can fit into pre-defined forms. This is understandable and for many applications even useful. Yet, one never knows what (combination of) questions may be asked of the material. The forms should then be as open as the data structure allows, and it would be ideal if the users were given the possibility to make their own queries. If open data standards are properly maintained, it may even be possible to extend queries among a number of datasets. As making queries is quite technical, the feasible way to do this is allowing websites to develop their own interface for specific purposes.
  • Make the images freely available
    Copyright serves to protect the rights of the artists. In our case, the creators of our pots and images have passed away more than 70 years ago (or so we may assume), the term for copyright protection in most countries. Museums do not have copyright on their holdings but on the photographs they make of it; they may choose to relinquish this right and make the images freely available. More and more museums are actually doing this (fig. 1). The Beazley Archive stamps the images with a watermark and strictly prohibits their use (fig. 1) [3]. To avoid these infringements and the disfiguring stamp, one resorts for lectures and lessons to sources like Flickr. This is hardly ever good publicity for the museum, so they are doing the sensible thing in making their own photos available. Vase painting is not a widely popular subject; it can only gain in popularity, and it may broaden the use, when the CVA would make the images freely available. With about 200 volumes sold per project, the loss of revenue will be negligible.
  • Expansion of scholarly reach
    The excellent ‘Beihefte’ of the German and, two years ago, the Austrian CVA committee bring the project beyond the descriptive. They offer reflection on the work done in the series, and particularly the Austrian volume introduces new technologies in the study of Greek vase painting. In so doing, it may even challenge the publishing conventions of the series. The first Austrian Beiheft showcased a number of interesting technologies: CT-scanning, 3d reconstruction, 3d and surface scanning. To these petrography, special photographic techniques to bring out specific features and doubtless many more may be added. The Austrian work is in a separate volume, but it should be integrated with the ‘normal’ data of the object. A database serving as a CVA platform could mean easy integration with all types of data.

Our own work and focus
All the above can be done with the data we normally generate within the framework of the CVA project. I finish this paper by addressing subjects which touch upon the actual work we do and the way we are used to doing it. It entails the development of the CVA project into a CVA platform.

  • Giving primacy to context
    The CVA is very much defined by the art historical archaeologist working from a museum. Context, or rather provenance, has its place in a CVA, but from a museum perspective. Since these collections have not been brought together with the archaeological context in mind, it hardly plays a role in the comment. In the CVA online, the provenance is not even included. Provenance is the natural connection between our work and that of the rest of archaeology, and also an increasingly important issue for scholars working on iconography, ancient economy and history. Provenance should have its proper place in the CVA. But perhaps we should go one step further: the energy and scholarship devoted to unprovenanced vases should be channelled into provenanced ones. There are enormous amounts of unpublished or poorly published provenanced vases. We should move out of the museums into the fields. Turning the CVA into a platform rather than a project would do away with the distinction between excavation reports and museum catalogues and with the counterproductive imperative that a CVA cannot be a primary publication of an excavation.
    Of course, there are practical problems such as the lack of accessibility of excavated material. Museum collections are often interesting cross-sections and offer other venues of research. Yet, with our scant means we would do well to channel our energies into provenanced material.
  • Inviting discussion and incorporating outside scholarship
    Writing a CVA is a solitary business, and the result is an authoritative work in two senses of the word: its standard of scholarship makes that it is taken seriously, and it does not allow for new input or dissenting opinions (save through the somewhat fleeting medium of the review). The CVA platform should be open and inviting to scholars wanting to contribute to the discussion of the artefact. If an historian of religion has an interesting observation about a scene of offering, it should be possible to integrate it into the comment on a particular vase. Maybe it goes too far to look at Wikipedia as a model, but a moderated version of it could be the way forward. Opening the possibility to add new material and serving as a platform for questions are other ways to enhance the CVA’s use. This calls for a more open and less authoritative attitude of the scholars working on CVA’s. Again, the FACEM project shows how this can be realized: inviting both material and input, but at the same time giving very clear guide-lines and a check through a review system.

To sum up: to reach a wider audience, we should publish digitally in an open database, maintain the standardization and high standard of scholarship, bear in mind the needs of potential users or at least allow others to do this. We should connect all publications within the CVA project in a digital platform and foster connections with datasets of other material. We should make the images and data freely available. We continue the expansion of the research associated with the CVA as exemplified by the German and Austrian Beihefte and try to integrate this into the CVA. Context should be a clearer focus, and our work should be open to comments and contributions.
Finally, a word about the costs and gains. The production of CVA’s is a costly business. The difficult format, high standard of the photographs and low print run make it expensive, but still this is only maybe 20-30% of the total cost of a CVA project. With making it freely available, one misses the revenue of the sale of the volumes, but this covers only a small part of the total cost of the project. The total balance could well be neutral, while the broader scope may well make the project more viable for grant funding.
The gains, on the other hand, are immense. It would integrate our field of study in the wider field of archaeology. The platform would greatly increase the usability of our work for many areas of research. It will contribute to more awareness of the relevance of our material. It would offer the building blocks to attract a wider audience to Greek vase painting. And last but not least, instead of filling shelves with impressive but also forbidding and somewhat antiquated-looking volumes, we can suddenly join in with buzzwords like ‘open access’, ‘big data’, ‘data mining’, ‘digital humanities’ and maybe even ‘networks’.

[1] T. Mannack – G. Parker, Das elektronische CVA – Stand 2010, in: E. Trinkl (Hrsg.), Interdiszipliäre Dokumentations- und Visualisierungsmethoden, CVA Österreich Beiheft 1 (Wien 2013) 17–23;
[2] W. van de Put, CVA, Corpus or Corpse?, in: R.F. Docter – E.M. Moorman (Hrsg.), Proceedings oft he XVth International Congress of Classical Archaeology, Amsterdam, July 12-17 1998 (Amsterdam 1999) 428–429.
[3] The Beazley Archive actually includes colour photographs from i.e. the Metropolitan Museum of Art, New York.

© Winfred van de Put

This article should be cited like this: W. van de Put, Corpus Vasorum Antiquorum: a future for ideals of the past?, Forum Archaeologiae 74/III/2015 (