This project investigated the visualization of the hidden zones beneath the Earth’s surface and the interior of the Earth, in science and art. The geological underground as a subterranean space is not physically accessible in the same way as a geographical landscape, which can be observed, surveyed, described, and depicted according to the criteria of modern empirical science. In contrast to scientific illustrations portraying the morphology of the Earth’s surface, the production of images of the underground can scarcely refer to pictorial conventions or artistic traditions of naturalistic representation. Empirical examinations of the subterranean world—in the sense of a direct observation of nature, carried out on the spot and with one’s own eyes—are feasible only to a very limited extent. Our conception of the appearance, configuration, and consistency of the underground is based on an intellectual synopsis of extremely fragmentary insights. Every attempt to visualise the interior zones of the Earth therefore has to resort to as many of the available fragments of information as possible. In any case it will always remain largely dependent on the results of scientific speculation and deductive conclusion.
Susanne B. Keller'S assumption was that pictures played a fundamental role in the production of knowledge of the Earth’s interior and the underground. Yet the translation of the fragmentary knowledge of the subterranean realms into images required the development of specific strategies of visualization. For centuries, new modes of representation were repeatedly put to the test. Some found general acceptance, others were discounted and not used again. Attempts to visualize the underground were always conditioned by the contemporary faculties of perception and conceptions of Nature on the one hand; on the other hand they also depended on the specific technical, scientific, and artistic practices of their time.
This project examined selected examples of graphical illustrations of subterranean structures and spaces in a series of case studies. Depictions of the interior of the Earth in cosmological theories of the seventeenth century as they appear in the works of René Descartes, Athanasius Kircher, or Thomas Burnet were developed in an epistemological context very different from, for instance, that of illustrations in books on mining practices. In the latter, pictures became exceedingly important in ensuring effective communication about subterranean structures and for the location of mineral deposits. From the sixteenth century, when Georgius Agricola published his lavishly illustrated De re metallica (1556), up to the elaborate engravings of “coupes des mines” and “géométrie souterreine,” engraved for the passages on mineralogy in Diderot’s Encyclopédie (1751–80), remarkable changes in modes of representation and perspective are obvious.
Geological sections in late eighteenth-century geological and stratigraphical research were for the most part based upon mining experience. However, they often also carried further connotations. They could relate to a variety of chronological, volcanological, seismological, or archeological debates and questions. By the beginning of the nineteenth century, geological sections had largely established themselves as part of the visual language of the Earth sciences.
A major focus of this project was the examination of the word-image relationship in illustrated scientific treatises concerning the underground. It was considered where and how an image could and did complement or even substitute the text, and how the mutual interactions worked. Susanne B. Keller also took into account whether the respective draughtsman or scientist made it explicit that he theoretically reflected upon his choice of a certain mode of representation. Paying close attention to the pictures themselves, Keller investigated how the relationship of the surface to the underneath finds its visual expression. Questions include: What can be deduced from the surface? What concept of observation does this involve? Is there a conflict between naturalistic representation and the need for abstraction? What is the epistemic value of fragmentary, and thus contingent, visual information for a visualization of the whole?