However much cartoon specialists might deplore the fact, the principal academic use of cartoons originally published in newspapers and magazines is as supporting illustrative material for primarily text-based enterprises. Think only of the many books on modern British history which lack entirely a visual component except for the Punch cartoon adorning the cover (mea culpa: I published just such a book myself this year). Cartoons lend themselves to such uses because they touch on so many aspects of human life, including some that are not otherwise much illustrated (foreign policy, for example), and with a humour and visual flair that are attractive and, sometimes, revealing of psychological layers to a subject that textual evidence yields only with difficulty. On the other hand, cartoons are difficult to use. They are damnably allusive, relying even more than other matter from the daily press on subtle contemporary nuance that is hard to decipher. Their artists are still relatively under-studied – the Oxford Dictionary of National Biography (1) does not include, for example, one of the best-loved cartoonists of the 20th century, the Daily Express’s Sidney Strube - and their relationship to the editorial policy of the publications in which they appear is not easy to pin down. Above all cartoons are tricky to locate, requiring laborious paging through fragile newsprint or scrolling through dimly-lit microfilm. To the rising generation of digital searchers, they might become invisible altogether – optical character recognition (OCR) can track any text string, but many cartoons lack captions, where captions exist they are often handwritten and thus opaque to present-day OCR, and of course visual cues are completely unsearchable.
To the rescue rides the Cartoon Database  of the Centre for Cartoon and Caricature at the University of Kent. I say ‘Cartoon Database’ but it is not at all clear that this is what it is, or what it is called. The Centre itself has existed since 1973, and holds in the University library a collection of newspaper cuttings, illustrated magazines and original cartoon artwork. It began to catalogue them in earnest in 1996. Shortly thereafter it gained a grant from the Research Libraries Support Programme (RSLP) to digitize 43,000 images from its own and other major cartoon collections and make them available on the web. This project was known as CartoonHub, and has been live since at least 2002. Subsequently the Centre has extended the database from its own collection. At different places on its website, it refers to the online resource as ‘CartoonHub’, ‘the database’, and ‘the on-line catalogue’; at different places, it attributes to it 35,000, 85,000, 90,000 and 120,000 images. Such are the vagaries, no doubt, of a chequered funding (and administrative) history – but they do leave the user in some uncertainty.
Whatever we call it, and however large it is, the database is a wonderful resource, and the search facility resolves many of the minor and all of the greater irritations traditionally involved in using cartoons. Each cartoon is presented in the form of a low-resolution or watermarked scan that is usually perfectly legible and indeed satisfactory for any conceivable private use, and the Centre can assist with copyright clearance and provide high-resolution images suitable for publication. One can of course search for particular artists or publications, and specify spans of dates; one can search captions and all ‘embedded text’ (handwritten text has been transcribed), including the most casual labels. But the truly miraculous element, one that has the potential to unlock hitherto hidden visual content, is the tagging of each cartoon with a range of subject terms: characters (real people but also symbolic figures such as John Bull and Uncle Sam) and subject keywords (arranged in a hierarchical thesaurus so that one can relatively easily identify appropriate terms that the database will recognize).
Clearly the tagging can be no better than the tagger, and not every reference or context will have been accurately identified. Neither is all the embedded text perfectly transcribed. Nevertheless quite significant results have been achieved. For example, a search for ‘Livingstone, David’ produces 13 hits, including a few for which ‘Livingstone’ appears nowhere, but which have been associated with Livingstone because the words ‘I presume’ or even ‘Stanley’ appear, and an ‘implied text’ thus alludes to Livingstone. (More than one puns between the two Livingstones, David and Ken.) In each case the (sometimes quite obscure) political figures being portrayed in jungle meetings are also identified. The same cartoons could be found by searching for terms such as ‘big cats’ or ‘elephants’, ‘Zimbabwe’ or ‘Rhodesia’, ‘weapons, spears’, ‘expeditions’, ‘proverbs and sayings’, even ‘flies’ or ‘gnats’. A search for ‘tapestry’ yields 18 hits, all but two of which are parodies of the Bayeux Tapestry; the other two refer to Graham Sutherland’s tapestry for Coventry Cathedral. Again, all of the trade unionists, European politicians and Soviet apparatchiks portrayed in the 16 Bayeux tapestries have been identified and indexed. One could also have found the 16 Bayeux tapestries on their own by following the hierarchical thesaurus from ‘needlework’ or ‘tapestry’ down to ‘Bayeux tapestry’ and entering the more precise term in the first place.
Since May 2006 the database and its search engine have been overhauled and more complex searches are now possible. The menus are somewhat over-complicated and some search windows do not close after they have completed their task, leaving the search results hidden beneath until the user closes them manually. It is not easy (in some cases, it is impossible) to browse search terms; some buttons inviting you to ‘browse’ are really only designed to help you search, though a search for a term high up the hierarchical thesaurus will at least reveal narrower terms. It is not easy (again, sometimes impossible) to get back to the Centre home pages from the database. A regular user can quickly master these and other wrinkles, but first-time users may be perplexed, especially given the confusing variety of information about the database in the home pages.
One risky but potentially brilliant innovation is that the database has been ‘wikified’, that is, it can be edited by registered users. There is, of course, a great deal of controversy about the accuracy of wikified resources, but given the vast array of discrete knowledge necessary properly to contextualize cartoons, this database is probably as good a candidate for wikification as any. No doubt the Centre will be monitoring closely the work of its volunteer editors. It has already mounted an online forum on which researchers can post queries and discuss their editing; at present the forum has attracted only a small handful of users, but it is early days yet.
Like many such resources – and again, typically of a resource that expands in irregular phases - the Cartoon Database is better at telling you what it is than what it is not. It represents only a portion even of the Centre’s own holdings, but it is not in the Centre’s interest to say much about the holdings that are not yet digitized, still less about cartoons it does not hold. One page, evidently an outdated one, tells us slyly, ‘A complete list of cartoonists whose work is held is available from the Centre’. There is in fact an online index of all cartoonists whose work is held by the Centre, but the main link to it was broken when I tried it in August 2006. When revealed by another route, this index proves to contain very helpful thumbnail biographies and also some indication of the Centre’s holdings, but it is not easy to tell from this index which portions of the holdings are in the database and which not.
Last year, the Centre acquired the enormous collection of Carl Giles – ‘Giles’ of the Daily Express. It will take some time to catalogue; it is unclear from the website whether the Centre has the right to digitize it. Copyright of course poses unfathomable problems for online providers of content like this. From that point of view it is curious that so much of the database content dates from the 20th century. Only about 500 cartoons date from before 1900, most of them by ‘HB’ (John Doyle), from the Tabley Collection at the University of Manchester, included as part of the original RSLP programme. There are no Punch cartoons dating from before 1923. Naturally the Centre started its digitization programme with its own collections, many of which have been donated by living or recently deceased cartoonists; but one would have thought that RSLP (or AHRC or JISC) funds might have been sought and applied to the bulk digitization of Victorian cartoons which are mostly out of copyright.
In short, the Centre’s holdings represent only an uncertain fraction of the British cartoon universe, and the Centre’s digitized holdings represent only an uncertain fraction of that uncertain fraction, augmented by the anomalous ‘orphaned’ images from other collections that formed a part (how big a part?) of the RSLP programme. Some of this uncertainty arises from the variety of forms in which cartoons can be held now – as original artwork, as artwork embedded in pages of newsprint, as cuttings, as scans – as well as from the chequerboard of copyright restrictions, so that it is not always easy to say either what the Centre ‘holds’ or what it can provide to users from its holdings. Yet something called ‘CartoonHub’ – or even ‘the Cartoon Database’ – which is the centre for the study of cartoons in this country has a responsibility to its users to be as comprehensive and as candid as possible in its presentation both of what it can itself offer and of the larger universe of resources within which it sits. When the resource is updated, traces of the earlier versions need to be erased as thoroughly as possible (preferably with a historical record lodged on a dedicated page). The extent of the updated resource needs to be specified as clearly as possible – with reference to what has been added (for the convenience of returning users) and to what still remains outside its remit (for the education of novice users).
It is, of course, ungrateful to complain about the limitations of an enterprise such as this which represents such an enormous step forward in making accessible a hitherto hidden resource – especially when that frequently misused cliché, ‘hidden resource’, has rarely been so aptly applied as it is here. Some of these limitations derive understandably from stop-and-go funding, from the limited administrative capacity dedicated to web resources (especially those intended to be permanent – which all too often means neglected), from narrow institutional interests of universities and copyright-holders, and from the ever-rising expectations of users. Nevertheless, a reminder of these limitations might be a useful prod to finding new ways to cope with them. Above all, if we can establish scholarly norms for electronic resources as rigorous and conventional as the norms we have for print resources, then we might have some hope that funding, administrative and institutional considerations get accommodated to our norms rather than vice-versa.
- Oxford Dictionary of National Biography (Back to (1)