Graphic devices in English incunabula

In this blog post, Mari-Liisa Varila recounts the results of our initial survey of graphic devices. Conducted on English incunabula, this survey served as the basis of the project’s data collection principles.

In the Early Modern Graphic Literacies project, we record the use of graphic devices, primarily images, diagrams, and tables, and classify them either according to this division – or where necessary – as ‘uncertain’, to create a record of the use of graphic devices in early modern England. Most of our data are collected from specific sample years at a range of 25 years. Our methods are described further in Ruokkeinen et al 2023. However, for incunabula, we examined all books in English available on EEBO from the 1470s to 1500. Investigating the incunabula was my responsibility, and this blog post discusses some of the observations I made during my data collection process. Although the results will probably not surprise anyone who has worked extensively on the earliest English printed books, this survey forms a backdrop for our findings from the later centuries.

There were altogether 185 incunabula included in our analysis; of the ca. 250 books on EEBO identified as incunabula in English, we excluded fragments, duplicates, and some items tagged ‘English’ that were in Latin. I went through the image sets of these books on EEBO, looking for illustrations, tables, and diagrams. Following the EModGraL approach, I did not take into account images that have a specific marketing or decorative function (such as images on title-pages, frontispieces, printers’ devices, or ornaments). This analysis also excludes paratextual devices such as tables of contents.

A majority of the incunabula have no images at all. Where images are found, they are typically illustrations with no captions or caption-like elements – text that would identify or describe the contents of the illustration and that would typically be set apart from the body text. However, there are also some illustrations that are equipped with captions. Some incunabula only contain a few illustrations, while others – for instance the Canterbury Tales (STC 5083) and the Legenda aurea (STC 24873) – make frequent use of them.

There are no images that clearly fulfil all of our present-day criteria for tables, but there are several instances of table-like elements or, as we call them in the project ‘unclear tables’. Some of these will be discussed in a separate study (Liira & Scase, in prep.). There are also a few other graphic devices that are difficult to classify, for instance the medallion-within-a-square designs in the ‘St Albans Chronicle’, STC 9995 (a later example of such a device is shown in Ruokkeinen et al., Figure 8).

Regarding the use of diagrams or diagram-like images, two works stand out from the rest. The first one of these is William Caxton’s Mirror of the World (STC 24762, 1481). This edition is exceptional in its use of not only images, but also many diagrammatic illustrations, missing from nearly every other incunable. Caxton’s illustrations draw on the French manuscript tradition. A digitized copy of the edition can be consulted for instance on the Library of Congress website here. In a copy of the later edition (1489) currently in Glasgow University Library (Sp Coll Hunterian Bv.2.30), a reader has contributed to the illustrations by annotating the world map.

Another intriguing book in terms of its image types is Wynkyn de Worde’s second edition of the compilation known as the Book of Saint Albans (STC 3309, 1496). This work contains some illustrations that might be described as diagrammatic to a degree. In the section on fishing, there are images of different kinds of instruments and fishing lines, and these are equipped with explanatory captions (e.g. “The lyne for a pyke”, H4r). Although these images are in our classification placed under ‘General image’, their practical function calls to mind later, more complex technical diagrams.

Of course, the scarcity of tables and diagrams in the earliest English printed books does not mean that such devices did not exist yet. On the contrary, there are various kinds of tabular and diagrammatic devices already in medieval (and earlier) manuscripts. In Peikola & Varila 2024, recently introduced in this blog, we examine the various ways in which medieval text producers instructed their readers in understanding and using the devices. The lack of tabular and diagrammatic elements in late fifteenth-century English printed books may partly be a result of the economic and technical constraints of early printing. During this period, printers in England were still experimenting with the potential of the new technology while also employing previously conventionalised ways of presenting text on the page. Images (woodblocks) had to be prepared separately from the text (type), and it is possible that the costs of commissioning images were easier to justify for narrative texts containing enticing illustrations rather than for practical books with diagrams. Perhaps illustrations were also slightly easier to reuse in editions of other works than diagrams created for a very specific purpose. As for tables, they could in theory be laid out simply by setting the type into a grid formed by rows, columns, and blank space. Although there are no tables in the incunabula that would adhere to a strict modern definition of ‘table’ (typically a ruled grid with labelled rows and/or columns), there is still evidence of tabular formatting in these early books, perhaps most commonly in different kinds of lists as well as paratextual elements such as tables of contents and indices.

Mari-Liisa Varila

twitter Follow us on Twitter!

Facebooktwitterredditpinterestlinkedinmail