Winged Horses and Snail-shaped Dolphins: A Broader Perspective on Drawings after the Antique in the Census Database

Zur deutschen Version wechseln

The database of the Census of Antique Works of Art and Architecture Known in the Renaissance will have a new look this Spring: the blue and grey interface with which the Census went online in 2007 will be replaced by a new one. While preparations for the transition are in full swing, an update of the database’s visual resources is also underway. A photo campaign began in 2021 has allowed the Census to acquire, on a large scale, high-resolution digital images of antique drawings from the Renaissance. As student assistants at the Institute for Art and Visual History at the Humboldt University in Berlin, we have been continuously adding new images adapted to today’s viewing habits. Updating the black-and-white photographs from the time of the project’s earlier history to new, colour reproductions brings about not only an improvement in user experience. The latest digital images, which can be magnified until individual paper fibres are visible, call attention to the status of the drawings in the Census as independent works of art. Even the purely technical handling of these images — digital cropping, formatting and uploading of image files into the database — opens up a new perspective on the way that these ‘documents’ for the reception of antiquity are represented in the Census database.

Antique drawings as ‘documentary’ material

Drawings inspired by the antique appear with increasing frequency from the end of the 15th century, a time of flourishing interest in ancient art and architecture.¹ In traditional art-historical research, these drawings were primarily regarded as sources of information about their ancient models.² At the time the Census project began in the 1940s, this methodical approach to Renaissance pictorial ‘source material’, inherited from classical archaeology, was well established in art history. The data model of the Census emerged from this traditional approach and has largely remained intact to the present day. In the Census database, for example, written and pictorial sources from the Renaissance that refer to ancient buildings and works of art are grouped together in the category of ‘documents’. The sub-category ‘type of documentary material’ makes it possible to differentiate between the various genres of documents (drawings, prints, or sketchbooks), but also underscores the overarching ontological status of these objects within the Census. The core ‘documentary material’ in the database are sketchbooks by architects and artists of the 15th and 16th centuries, many of which were devoted to the study of ancient ruins. Among them is the author of the so-called Ripanda Sketchbook (1510-1530), who is thought to be a draughtsman from the circle of the Bolognese painter Jacopo Ripanda. Last year, the Census acquired new reproductions of the 63 folios that were still preserved from this sketchbook. Updating the corresponding database entries opened up a new perspective on these drawings, which have so far been unevenly studied.³ On closer inspection, it has raised the questions of whether the previous categorisation of such drawings as Census documents needs to be updated. 

The so-called Ripanda Sketchbook

The so-called Ripanda sketchbook, acquired in 1935 by the Ashmolean Museum,⁴ has also been known as the Oxford Sketchbook (‘Taccuino di Oxford’) and its draughtsman has been known as the Oxford Master (‘Maestro di Oxford’). Its approximately 170 individual drawings on 63 folios, mostly on both recto and verso—in graphite, black and brown ink and watercolour—are largely architectural sketches and studies of architectural ornament and reliefs. Presumably because they have been considered to be of mediocre quality, the drawings themselves have received scant attention.⁵ A study by Valeria Cafà, focussed on the architectural representations, identifies new relationships with antique works and is exceptional in its close consideration of individual drawings.⁶ In contrast to the usual description of the Oxford Sketchbook as a modelbook of antique forms for use in a workshop, Cafà emphasises the ‘autobiographical’ character of the drawings, which represent the author’s personal collection of ideas.⁷ Given the ordered arrangement of drawings on the folios, it can be assumed that the sketchbook was intended to show off the artistic originality of the Oxford master and would have served as a demonstration piece for potential clients.⁷ Some of the Oxford drawings seem to testify to the artist’s interest in ephemeral architecture, which suggests that designing antique theatre sets or similar might have been one of the draughtsman’s aims.⁸ However, the incomplete state of preservation of the sketchbook, which originally contained around 108 sheets, still leaves unanswered questions concerning the actual function of the drawings and the context in which they were created.⁹

These parameters have however had little relevance in the context of the Census database, since its focus has been on the documentation of ancient monuments, and thus on the informational content of such drawings and the identification of their subjects. Around half of the drawings in the so-called Ripanda Sketchbook have been linked with genuinely ancient monuments in the Census database, based on identifications made in particular by Karl T. Parker (former curator at the Ashmolean) and by Census luminaries Phyllis P. Bober and Ruth Rubinstein.¹⁰ However, a considerable proportion of the drawings evidently inspired by antiquity has not yet been convincingly matched with identifiable antique works. Until recently, the sketchbook was therefore only represented by selected sheets in the Census: those sheets which included representations of securely identifiable antiquities. High-resolution reproductions had been ordered of the many remaining drawings; with the awareness that these sheets would otherwise have remained invisible to the users of the database, and with the hope that further identifications of ancient works would be made in the future, the decision was made to add all missing folios and to represent the surviving version of the Oxford sketchbook as a whole in the database. The hope was to inspire new connections between the individual drawings and new ideas about the intentions of the draughtsman.

‘Copies, more or less free, from the Antique‘

The search for further unambiguous identifications of antique works in the Oxford sketchbook quickly became a challenge. The drawings that Parker described in 1956 as ‘Copies, more or less free, from the Antique’,¹¹ vary greatly in their degree of similarity to ancient originals. What is initially reminiscent of a genuine antique work reveals, upon closer inspection, clear differences or even modifications by the draughtsman. An intriguing example of this ambiguity between copy and invention is found already on the first folios. On folio 2 recto is a capital with two winged horses, one that is immediately reminiscent of the Pegasus capital of the Temple of Mars Ultor represented in various Renaissance drawing books. One of the most precise of these is a drawing by the architect Baldassare Peruzzi, which includes measurements of the original work (and which, incidentally, provided the model for the logo of the Census). A comparison between this drawing and the Ashmolean drawing suggests that the anonymous Oxford draughtsman must actually have known the ancient work. Both show the same basic composition of the Corinthian capital, with a lower row of acanthus leaves and winged, rearing horses on either side, their front hooves protruding over the outward-turning tips of the leaves. Yet, while their bodies merge into floral decorations in the original, as was copied meticulously by Peruzzi, the Oxford draughtsman drew the Pegasoi in three-dimensional form. Represented in a narrow pictorial space, the winged horses stand out voluminously and seem almost alive. The author of the Oxford sketchbook not only modified the original composition of the ancient capital, but also seems to have consciously detached it from its materiality as a work in stone. This becomes particularly notable when the drawing is compared with a surviving Pegasus capital of a pilaster from the same temple, which gives an idea of ​​the former appearance of the (now lost) column capital.

Something similar can be seen again on the verso of the same folio, where the imaginative, playful character of the draughtsman¹² is once again revealed. On folio 2 verso, in place of the stylised volutes of a capital, he draws shells, from which snails with long necks seem to peer out, extending their feelers. These two examples alone make it clear that the drawings in the Oxford sketchbook make no claim to archaeological accuracy. They appear less as precise documents and more as an image of the draughtsman’s creative response to ancient sources of inspiration.¹³  In drawings of a façade and ground plan on fols 7r and 25r, individual elements of authentic ancient buildings such as the Colosseum or the Arch of Constantine repeatedly appear,¹⁴ Some of these elements recur several times in different sketches¹⁵ or are combined with motifs from other ancient monuments in an architectural design. Overall, however, many of the drawings stray too far from the ancient originals to be classified as documentary evidence. If one tries to position these drawings within the data model of the Census, i.e. according to a semantic and logical system focused on documentation, one repeatedly comes across questions such as: how close to the original does an image have to be to included as a ‘document’ in the database? When is a Renaissance drawing actually relevant to the Census?

Ancient model or invention?

A capital on the left side of folio 2A recto raises these questions. It shows the motif of two dolphins whose tails form volutes. The search for matching antique capitals in the database resulted in several hits, all of which are categorised as ‘unidentified monuments’. These are monuments that are known solely through postclassical sources and are categorised as ancient on the basis of these. For example, an illustration in the Codex Escurialensis bears a close resemblance to the dolphins in the Oxford sketchbook in both drawings the dolphins are positioned over a central shell, with the points of their beaks almost touching. As if to show that they are diving down towards the shell from either side, their fins are raised to the sides and spiral inwards at the ends. In both cases, the shell fits into a row of acanthus leaves and at the transition to the abacus at the top, a pronounced ovolo is shown. In addition to the absence of the arched ornamentation connecting the row of leaves in the Codex Escurialensis drawing, upon closer inspection one notices that both the orientation of the shell and the rotation of the volutes are reversed in the two drawings. Such a fundamental difference only contradicts the theory of a common ancient model if one assumes that the draughtsman’s aim was documentary. The strong similarity of the compositions as a whole is indeed surprising. A similar observation can be made about two capitals in the Larger Talman Album (fol. 158 and fol. 159) which is also in the Ashmolean Museum. In both cases, the basic composition and the rotation of the shell and volutes correspond to the drawing in the so-called Ripanda sketchbook. But in addition to these similarities, there are also visible differences between the drawings.

These observations show the difficulty in considering whether drawings such as the dolphins capital in the Oxford Sketchbook can be classified as documents of a lost ancient original. How can one verify the status of the drawing as a document? And how can a work inspired by antiquity, but artistically modified by the draughtsman, be integrated into the Census database so that such connections can be researched? What do we miss if we look at drawings primarily in terms of their documentary value? The connection between the Oxford sketchbook and other sketchbooks, which can be seen in the dolphin capital, also points to the need to examine their relationships to one another more closely. In addition to this particular example, other drawings also strongly suggest that the three codices are related.¹⁶ The possibility that these drawings are related to one another cannot be ruled out,¹⁷ so that the question of the antique model is shifted to the question of the original drawing.

 Unanswered questions and a small surprise

The update of the Census database with new digital images of Renaissance drawings after the antique raises a number of questions that remain unanswered, but which are nonetheless fruitful for Census research. They touch on a fundamental, recurring problem common to all databases: the definition and further development of its ontology, the categories and semantics of its data model. A closer look at just one of the many codices in the Census reveals how much we are missing and the difficulties we encounter when examining these works for their documentary value alone. However, if we look at these drawings from a more specifically art-historical perspective, i.e. primarily as works of art, we come to a broader understanding of these drawings and their diverse functions.

Updating the Oxford Sketchbook database records also brought less abstract insights. When we reviewed the relevant material in the Census archive, we were rewarded with a surprising discovery, and a short journey back in time to the history of the Census. The digitisation of the database in the 1980s was preceded by an analogue system which combined index cards (where information on monuments and documents was compiled) and photographs. Phyllis Bober created two sets of index cards in New York, while employees of the Warburg Institute Photographic Collection compiled two sets of photographs in London; mailing the cards from New York to London and the photographs from London to New York allowed both institutes to maintain a complete copy of the Census. One set of the original cards and photographs is still kept at in the Warburg Institute, while the set of cards and photographs that were originally kept in New York are now in Berlin.

A comparison of the versos of photos of the Oxford sketchbook in the London and Berlin Census, which must have been created before 1972, brought a small discovery. What is different generally about the two sets of cards and photographs are the handwritten notes on the versos: while the set in Berlin includes only Phyllis Bober’s handwritten notes, the index cards and photographs in London include notes by several hands. The vast majority of the notes in the London Census are by Ruth Rubinstein, assistant of the Photographic Collection with special responsibility for the Census between 1957 and 1996, but other hands are also seen. In the case of the so-called Ripanda Sketchbook, both the photographs in Berlin and those in London have Bober’s notes on their versos. Yet there are also notes by another hand, which is not Ruth Rubinstein’s, on the back of the London version of these photos. Paul Taylor, Curator of the Photographic ­Collection at the Warburg Institute who kindly supported us in our research, identified the writing as that of Enriqueta Frankfort, Curator of the Photographic Collection at the Warburg Institute until 1970. Although Ruth Rubinstein took on a much larger role in building up the Census and was Bober’s most important longterm collaborator, Frankfort’s role should also not be forgotten. As J. B. Trapp emphasised in his short history of the Census, Frankfort was a key figure in establishing and sustaining the Census as a project of the Warburg Institute after Fritz Saxl’s death in 1948 and in the decades that followed.¹⁸ Our investigation into the Ripanda sketchbook brought a small glimpse into the ongoing collaboration between Bober and Frankfort, as well as Frankfort’s contribution to the Census, which deserves further study.

Translated by Kathleen Christian

Addendum: This text was written in the summer of 2022, when an update of the terminology of the Census database was already discussed internally. Finally, the decision was made to rename the main categories of “Monuments” and “Documents” to “Antique Monuments” and “Postclassical Works”. Accordingly, the property “Type of documentary material” was also changed to “Type of Work”. On 04.04.2023 the Census went online for the first time with a new interface and updated terminology.

1. Arnold Nesselrath: Der Zeichner und sein Buch. Die Darstellung der antiken Architektur im 15. und 16. Jahrhundert, Cyriacus, Vol. 4, Mainz/Ruhpolding 2014, passim. [↩]
2. Cammy Brothers: ‘What Drawings did in Renaissance Italy’, in: The Companion to the History of Architecture, Vol. I, Renaissance and Baroque Architecture, ed. by Alina Payne, 2017, p. 1. [↩]
3. In addition to the information in the catalogue of drawings in the Ashmolean Museum by K. T. Parker (1956), which is still relevant today, more focused studies of the Oxford sketchbook can be found above all in V. Farinella’s monograph on Ripanda (1992) and in an essay by V. Cafà (2002): Karl Theodore Parker: Catalogue of the Collection of Drawings in the Ashmolean Museum, II: Italian Schools, Oxford 1956, no. 668, pp. 357-360; Vincenzo Farinella: Archeologia e Pittura a Roma tra Quattrocento e Cinquecento. Il caso di Jacopo Ripanda, Turin 1992, pp. 160-191; Valeria Cafà: ‘I disegni di architettura del Taccuino KP668 all’Ashmolean Museum di Oxford’, in: Annali di Architettura. Rivista del Centro internazionale di Studi di Architettura Andrea Palladio di Vicenza, 14, 2002, pp. 129-161. [↩]
4. Farinella 1992, p. 160. [↩]
5. Cafà 2002, p. 129. [↩]
6. Cafà 2002, Appendice, passim. [↩]
7. Cafà 2002, pp. 132-133. [↩] [↩]
8. Cafà 2002, pp. 131-132. [↩]
9. Cafà 2002, p. 132. [↩]
10. Parker 1956, no. 668, pp. 357-360; Phyllis Pray Bober and Ruth Rubinstein: Renaissance Artists and Antique Sculpture. A Handbook of Sources, London/Oxford 1986, p. 466. [↩]
11. Parker 1956, no. 668, p. 357. [↩]
12. Valeria Cafà in particular has stressed the ‘fantastic’ nature of the Oxford Master’s drawings, Cafà 2002, passim. [↩]
13. Cafe 2002, pp. 131-133. [↩]
14. Cafà 2002, no. 15 (fol. 7r), or no. 17 (fol. 8 ) and no. 23 (fol. 11r). [↩]
15. For example the frieze of the Basilica Aemilia, see Cafà 2002, no. 27 (fol. 13r) and no. 37 (fol. 18r). [↩]
16. The lower half of fol. 6r shows two capitals which are probably not antique, each of which is also present in a half of the capital the upper right corner of fol. 153r of the Larger Talman Album; on the connection between the Codex Escurialensis and the Larger Talman Album see also Charlotte Schreiter: „Römische Schmuckbasen in Renaissancezeichnungen. Interpretationsmöglichkeiten aus archäologischer Sicht“, in: Pegasus. Berliner Beiträge zum Nachleben der Antike, 4, 2003, pp. 37-66, esp. p. 54. [↩]
17. Nesselrath 2014, p. 47ff. [↩]
18. Joseph B. Trapp: “The Census: its Past, its Present and its Future”, in: Pegasus. Berliner Beiträge zum Nachleben der Antike 1 (1999), pp. 11–21. [↩]

---

---