History of the Census

The Census began in 1946 as a colla­bo­ra­tion between the art histo­rian Richard Kraut­heimer (1897–1994) at Vassar College, Fritz Saxl (1890–1948) Director of the Warburg Insti­tute in London, and the archaeo­lo­gist Karl Lehmann (1894–1960) at New York Univer­sity. All three had fled the perse­cu­tion of Jewish acade­mics in Germany and had found posi­tions abroad. The initial rese­arch ques­tion had arisen in 1945, when Kraut­heimer and his wife Trude Kraut­heimer-Hess were writing a mono­graph on the Renais­sance sculptor Lorenzo Ghiberti. Finding it diffi­cult to locate specific infor­ma­tion about the antique monu­ments visible in fifte­enth-century Italy, Richard Kraut­heimer asked Fritz Saxl if they should orga­nise a ‘corpus of anti­qui­ties’ known to the Renais­sance. This exch­ange began the idea of crea­ting an instru­ment that would syste­ma­ti­cally record ancient works that were, or could have been, known to Renais­sance artists.   In the Spring of 1946, the Warburg Insti­tute and New York Univer­sity formed a part­ner­ship to develop a Census of visual sources (which were limited, at the time, to sculp­ture and figural works of art) and textual sources from the Renais­sance (prima­rily, 15th-century Italy) that respond to antique monu­ments. Their chosen method would be to gather infor­ma­tion on index cards and combine these when­ever possible with photographs.

In 1947, the American archaeo­lo­gist Phyllis Pray Bober began to realise the ideas of the foun­ders, giving the Census the shape which it still has today. Bober compiled hundreds of index cards orga­nised alpha­be­ti­cally; each card listed the Renais­sance texts and works of art which could be put in rela­tion to a parti­cular antique monu­ment. Photo­gra­phic repro­duc­tions of both ancient and modern works of art were acquired and cata­logued by the Photo­gra­phic Coll­ec­tion of the Warburg Insti­tute. At this early stage of the Census, the scope of the project was rest­ricted to antique sculp­ture and its Renais­sance recep­tion from 1400 to around 1530.

The Census had since 1949 been a formal project of the Warburg Insti­tute, with Phyllis Bober its director. In the early years Enri­quetta Frank­fort, Director of the Warburg Photo­gra­phic Coll­ec­tion, worked toge­ther with Bober to build the project’s image resources. Ruth Rubinstein’s appoint­ment in 1957 as assistant in the Photo­gra­phic Coll­ec­tion with special respon­si­bi­lity for the Census gave the project a further impetus, as well as another long­stan­ding protagonist.

Over the next decades Rubin­stein purchased photos of sketch­books and drawings after the antique for the Census, making the Photo­gra­phic Coll­ec­tion of the Warburg Insti­tute one of the most important centres for the study of Renais­sance drawings. The Warburg Insti­tute also supported the work of the Census by publi­shing nume­rous cata­lo­gues of Renais­sance sketch­books in its Studies of the Warburg Insti­tute series.

In the early 1980s, the Biblio­theca Hertziana under the direc­tor­ship of the archi­tec­tural histo­rians Matthias Winner and Chris­toph Luit­pold Frommel became a partner insti­tu­tion of the Census project. At the same time, the Census expanded to include ancient archi­tec­ture, a cate­gory which had previously been excluded.

The idea of ​​compu­te­ri­zing the index card system of the Census had already been proposed at the end of the 1970s, yet the expan­sion of the project to include archi­tec­ture offered another reason to move forward with the idea. The J. Paul Getty Trust had just begun its Art History Infor­ma­tion Program in order to test the possi­bi­li­ties of elec­tronic data proces­sing for the huma­ni­ties disciplines.

The Census won support from the Getty, and programming began. From 1981 onwards, under the direc­tion of Arnold Nessel­rath in Rome and in coope­ra­tion with the computer programmer Rick Holt, the data model was deve­loped. Programming was under­taken of an input mode, and later also a retrieval mode. In deve­lo­ping the data model, the complex infor­ma­tion written on Bober and Rubinstein’s index cards had to be given a clear struc­ture. An advan­tage of the chosen object-rela­tional data model became the ability to access the Census data from all sides: while the index card system could only be ‘entered’ via the names of the ancient monu­ments, the compu­te­rised data could be queried from all directions.

The project’s next change of loca­tion coin­cided with a tran­si­tion into a new era of tech­no­logy. Funding from the Biblio­theca Hertziana termi­nated in 1995 and Horst Brede­kamp, who had recently been appointed professor at the Humboldt Univer­sity, succeeded in secu­ring a perma­nent affi­lia­tion of the project with what is now the Insti­tute of Art and Visual History. Addi­tional funding and support was provided by the Bundes­mi­nis­te­rium für Bildung und Forschung (Federal Ministry of Educa­tion and Research).

In the first years in Berlin, the soft­ware of the data­base was reor­ga­nised. A migra­tion to the Dyabola system preserved the data model. While the consul­ta­tion of the data­base had previously only been possible only in London and Rome, now the data­base was published, first on CD-Rom, and later on DVD. In 2000, the first internet version of the Census data­base was made available.