Keywords

1 Introduction

Earth’s geological past extends far beyond the human capacity for comprehension. Deep time dilates natural history to the extent that human reasoning seems to lose its power; human’s theories of the earth reach their “darkest point […] we start falling into the abysses of time. There, genius’ insight seems not to be enough and, due to lack of observations, it is powerless to guide us forward” (Comte de Buffon 1778, p. 44). While this vast temporal dimension clearly attracts scientists, it also traps them in a void and threatens the production of clear thought and relevant theories. Visualizations are one of the primary strategies used by earth scientists to improve their understanding and make deep time manageable.

Earth sciences are indeed extremely visual disciplines. It is quite difficult to reconstruct the history of, for instance, paleontology or geology without considering all the visual devices and technologies used to picture and investigate earth’s deep past. Indeed, and quite paradoxically, paleontologists sought to develop a series of visual and technological devices to make visible something inaccessible: earth’s deep time. Quoting Steve Shapin and Simon Schaffer’s seminal book Leviathan and the Air Pomp (1985), science historian Martin Rudwick affirmed that “any scene from deep time embodies a fundamental problem: it must make visible what is really invisible. It must give us the illusion that we are witnesses to a scene that we cannot really see; more precisely, it must make us ‘virtual witnesses’ to a scene that vanished long before there were any human beings to see it” (Rudwick 1985, 1). He used Shapin and Schaffer’s term “virtual witness” to emphasize that the process “makes real” that which is otherwise impossible to observe (ibid., 255).

Following Rudwick’s insights, in this chapter, I examine the power and limits of a series of visual devices used mainly in paleontology and geology to make visible, access, and eventually work with earth’s deep past. Accomplishing a phenomenology of visualizations used in paleontological knowledge production,Footnote 1 I will analyze the functions of these visualizations as essential means for working with the past. By so doing, I will defend the argument that visualizations help validate earth science as a field of scientific knowledge production.

The focus on the epistemic powers and limits of visualizations within earth science practices is relatively recent. Although Rudwick, as a consequence of his work as a paleontologist and morphologist during the 1960s,Footnote 2 called for a solid investigation of the visual language of geology as early as 1978, paleontological visualization began to be used as a strategy starting in the twenty-first century, following the broader material and visual turn in science study. From a variety of perspectives, scholars called attention to the different powers of visual media used to explore the earth’s (and human) past.

Before engaging with the phenomenology of deep-time visualizations, one key epistemic issue of earth science disciplines should be mentioned. Earth sciences are peculiar disciplines since they are located between historical and natural disciplines. For instance, one of the most representative earth sciences, paleontology, finds itself at the intersection of history, biology, and the natural sciences. This unique place within the system of science is due in part to the status of its data. The fossil record is always imperfect and incomplete. Only a tiny number of organisms and events that have shaped earth’s deep history have been preserved as fossilized objects. Hence, before using the record of the past for producing paleontological knowledge, paleontologists should analyze whether record of the past is biased. To do so, the fossil record is treated statistically and taken as a bulk to avoid possible biases and misleading interpretations of data.

Having mentioned the main epistemic peculiarity of paleontology, in the following section, I first focus on how paleontologists visualize and order field data; second, I look at the practices used to illustrate and validate knowledge of extinct animals; third, I explore the role visualization plays in supporting the transition between the collection of data and possible explanation of global phenomena; fourth, I investigate the role of computer; and fifth, I move to the recent collision of paleontology and new technologies to make visible some aspects of the past and generate new research questions. In the conclusion, I will reflect on the intersection between visual cultures and knowledge production.

2 Visualizing Field Data

In this section, I focus on the first set of visual devices used to work with fossilized and incomplete data gathered in the field to generate knowledge about the past. Visual practices were adopted during an emblematic paleontological expedition between 1989 and 1913: the Tendaguru expedition. Over the course of 4 years, scientists from the Berlin Museum für Naturkunde [Berlin Museum of Natural History] excavated more than 225 tons of fossils in present-day Tanzania and transported them to Berlin. Among these were the bones of the Brachiosaurus brancai, which would later become the largest erected dinosaur in the world, and which dominates the Dinosaur Exhibition Hall in Berlin to this day. The challenge faced by paleontologists and locals was both quantitative and quantitative: How could they deal with such a bulk of data, make sense of it, and exhibit the previously classified specimens in Berlin?

First, paleontologists advanced three kinds of considerations. Indeed, philosopher Darek Turner noted that in deciding what to collect and what to leave on the floor, scientists always ask whether, first, the specimen might be relevant to someone’s ongoing research; second, whether the specimen is well preserved, complete, or beautiful; and third, how much effort is required to unearth it and prepare it (Turner 2019). Second, as Tamborini described, pictorial representations produced through drawings and other so-called paper-based techniques were central to making sense of incomplete and imperfect fossils (Tamborini 2020). First, fossils were presented through simple means such as paper and pencil. Second, figures drawn on paper were cut and pasted together to present a potentially unified and coherent image of the fossilized organism. This first set of basic steps helped to tentatively visualize and categorize the excavated finds.Footnote 3

Although illustrations played a central role in visualizing deep-time organisms and guiding field work, they were not sufficient either to display a complex extinct organism in a museum hall or to present it as a coherent deep-time scene (sensu Rudwick) in textbooks or scientific papers. In other words, although field illustrations and sketches of fossils provide the semantic space of deep time, they still lack a syntax that brings the illustrations together. A different kind of knowledge and medium contributes to this task: bureaucratic knowledge and its technologies.

The transfer of knowledge between bureaucracy and natural history occurred continuously throughout the eighteenth and nineteenth centuries. Paleontologists, for example, have used tables and quantitative practices, which have already served an established function in the bureaucracy to reconstruct the history of nature (see next section). Natural history field work is characterized by the same transfer of knowledge. As Tamborini noted regarding the Tendaguru expedition, paleontologists on the field behave like bureaucrats. They had to meticulously classify, tabulate, and list the findings and link them to the particular spots on stratigraphic maps previously sketched. In this practice, the metadata assigned in the tables, lists, and book records determine how to re-read the fossils once they’re transported into the artificial space of a laboratory. In other words, they provide a syntax for reading the data decoupled from the field.

The starting point of this practice is the identification of the earth layers with a natural archive (Sepkoski 2017). In the earth’s natural archive, the fossils are preserved together with other important information about, for instance, the geological formation and possible ecological events which might be indicated in that particular earth and temporal stratum. In the field, the paleontologist can read the temporal and environmental features that make up the differences and similarities between different fossil groups. They take notes on every possible element, drawing on the stratigraphic profile of the landscape and earth layers, that might help them to decode all the information once they are in the lab. As a result, the fossils become documents to be read, written, and carefully kept: “Fossils are not stamps or rarities, but they are, in a sense, documents from long ago periods of our earth” (Fraas 1910, 4).

Practiced since the eighteenth century, this bureaucratic method to visualize and keep record of the past was institutionalized during the nineteenth century and reached its peak during the paleontological expeditions of the mid-nineteenth and early twentieth centuries (Rieppel 2019; Nieuwland 2019). As stated in various paleontological and geological guides and textbooks, “When traveling or collecting in different horizons, one should also become accustomed to conscientiously attach to each find a label noting the locality and geological horizon” (Fraas 1910, 8). Labels are the first step toward establishing the grammar useful to coherently visualize the past.

In addition to the proper labeling of excavated fossils and the appropriate selection and preservation of metadata, lengthier journal entries are needed to explain the necessary brief notes written about the circumstances of the find. In diaries, paleontologists cataloged what was excavated at each site so that the earth’s archive could be fully preserved and transported into a different context, such as museum’s laboratory, with minimal error. In addition, these diaries, which in the emblematic case of the Tendaguru expedition resembled a double-entry bookkeeping system, had to clearly establish “which fragments or parts belong together. The drawing of sketches or photographic records with the numbering of parts should not be omitted in the process” (Stromer 1920, 10).

Paper technologies and knowledge from bureaucratic disciplines, as well as background biological theories, were therefore essential to work, identify, and transport data and supporting information from the field into the lab.

3 Illustrations and Validation of Extinct Animals

Like in all biological and taxonomic disciplines, illustrations play a crucial role in paleontology. After fossils have been excavated, transported, and finally reassembled, they are described and classified in painstaking detail. By accomplishing these tasks, these specimens can be eventually shared with the broader scientific community through specific publications. This practice relies on a mixture of words and images. Fossil illustrations are published in compendia, textbooks, scientific articles, or even sent to colleagues as paper placeholders of the material fossil record. In effect, scientific illustrations represent a piece of the record of the past to an audience that cannot directly perceive it; the original record is mostly incomplete, imperfect, and impossible to transport (see the Rudwick’s account of Cuvier’s paper museum of fossil bones (Rudwick 2000)).

Driven first primarily by paper and pencil and a strong cooperation between artists and scientists, this practice and resulting visualizations have received a new impetus and rhetoric with the advent of technologies such as the electron microscope and 3D scanners. These tools enable paleontologists to perceive the so-called microcosmos, thus enabling the scientists to perceive microstructures otherwise invisible and non-representable.

Advertised to the broader public as the conquering of the invisible world, the integration of different kinds of technologies within paleontological research has had three main consequences. First, it re-established the gap between the scientist and the technician. This gap had narrowed to some degree through the drawing techniques learned by the paleontologists themselves during their course of study, when paleontologists learn the motto “to see is to draw”.Footnote 4 As philosopher Adrian Currie noted, “Sketching geological formations is important for training students to see like geologists: picking out pertinent details, learning how to represent them; making judgements about what is of epistemic import in geological formations” (p. 21). The same can be said for archeology and other earth sciences. This kind of formation represents an exercise in tacit knowledge. Sociologist Caitlin Wylie pointed indeed out that “distinguishing coveted fossilized bone from useless rock requires the tacit knowledge of expert visual judgment and manual skill to reveal the fossil by destroying the rock (i.e., the process of fossil preparation). Because of this, the same starting rock can result in a variety of different specimens” (Wylie 2019, 24). However, because of the difficulty of using highly specialized new technological tools, such as the electron microscope or 3D-scanners, scientists are now not generally able to produce images independently. They therefore collaborate with technicians to produce and visualize data. This has produced new social orders, as Wylie pointly illustrated.Footnote 5

Second, through the visualization of invisible and highly aesthetic images and structures (such as the inner composition of a bone, sand dollars (Fig. 1), etc.), the use of these technologies set the preconditions for a multi- and interdisciplinary approach to the morphological study of organisms. Earth scientists, architects, and engineers started collaborating more than ever to understand the principles responsible for the structures which were made visible. A classic example is the transfer of the morphological features of sand dollars from morphology into architectural manufacturing.

Fig. 1
figure 1

Skeleton of sand dollars used for designing research pavilion. Sand dollars’ morphology paved the way for cooperative enterprises – https://www.trr141.de/index.php/research-areas-2/a07/

Full size image

Third, these highly technological practices have pushed the use of images beyond their purely representational character and toward a presentational use (see sections below). Images have been used not merely to bring the past back to light, but rather to create a different planform upon which to investigate the past.

Besides presenting the past, images of fossilized organisms were used to validate the past (as Rudwick posed), to rhetorically promote paleontology to obtain financial support for expeditions and other activities, and in popular books and museum exhibitions to make visible the quite remote dimension of deep time to a broader audience.

4 From Local Data to Global Phenomena: Lists, Graphs, and Diagrams

The visual culture of earth sciences is marked by a polyvalent use of lists, graphs, and charts. As described in Sect. 1, scientists must perform a quasi-bureaucratic task of compiling fossils in the field to make sense of the bulk of incomplete and imperfect data unearthed. The extensive use of lists is also responsible for the transition from the collection and analysis of local data, for example all sorts of data gathered during an expedition, to the generation and analysis of global phenomena, such as phenomena of mass extinctions in the case of paleobiology and stratigraphy.

German geologist and paleontologist Albert Oppel (1831–1865), the founder of the modern zone concept, used lists as epistemic tools for collecting and comparing stratigraphic data on a global scale.Footnote 6 First, he collected all the species found in one specific earth layer and described the layer as the Early Jurassic (Unterer Lias). Second, to facilitate further analysis and comparison of this set of fossils, Oppel applied the name of the fossil most frequently found in the area to the entire zone. For instance, one zone of the Early Jurassic was named Ammonite planorbis after the massive presence of this fossil in this spatial and temporal dimension.Footnote 7 This operation was conducted through the morphological analyses of the fossils gathered, thus using the visual devices described in Sect. 2. Third, Oppel focused on the description of the geographical dispersion of these faunas by listing the places in Germany, France, and England where these fossils have been found. Comparing and bringing together different kinds of lists, Oppel identified the possible patterns of geographical dispersion of the species in Europe.

In the visual culture of earth sciences, lists also serve a function. They can be used to prompt the narration of deep time phenomena. Indeed, they can be seen as raw materials for what has been called data crunching. German paleontologist Heinrich Georg Bronn (1800–1862) came up with a complex system of lists to catalog the fossil record. He named it Index palaeontologicus (D. Sepkoski and Tamborini 2018; Tamborini 2015; Sepkoski 2012b). Successively, he took the previously listed fossils and tabulated them according to their number and place of discovery. Furthermore, he applied a quantitative approach to data proper to bureaucratic sciences to natural history. He used a series of pseudo-quantitative practices called statics to generate numerical patterns concerning the deep past. He successfully transferred another visual culture into the study of earth deep past: he took up the practices used in human history to visualize human past and implemented them into paleontology to narrate the natural history of data (Figs. 2 and 3).

Fig. 2
figure 2

Bronn’s list used to catalog the record of the past. See, Heinrich Georg Bronn, Untersuchungen Über Die Entwickelungs-Gesetze der organischen Welt Während der Bildungs-Zeit unserer Erd-Oberfläche (Stuttgart: Schweizerbart’sche Ver- lagsbuchhandlung, 1858)

Full size image
Fig. 3
figure 3

Bronn’s graph used to narrate earth’s deep history. See, Heinrich Georg Bronn, Untersuchungen über sie Entwickelungs-Gesetze der organischen Welt während der Bildungs-Zeit unserer Erd-Oberfläche (Stuttgart: Schweizerbart’sche Ver- lagsbuchhandlung, 1858).

Full size image

Bronn’s practice and set of visualizations became quite prolific in twentieth and twenty-first-centuries paleontology. For instance, J. John “Jack” Sepkoski Jr’s A Compendium of Fossil Marine Families (1982) lists approximately 3500 marine family fossils according to their first and last occurrence. The times of the first and last occurrence generates “81 stratigraphic intervals ranging from 1.6 to about 20 m.y. long” (Sepkoski Jr. 1994, 133). This Compendium was a model of data. It was namely a tool provisioning information for complex operations on its entries. It was very similar to Bronn’s Index palaeontologicus and other nineteenth-century enterprises. In fact, Sepkoski used his Compendium to study “mass extinction, evolutionary rates, and patterns of diversification” (ibid., 140) in the same was that Bronn used his Index to study the laws of diversification of the organisms in the past.

5 Computing the Past

Visualizations generated by data simulations provide another means of dealing with deep time. One classical example is the MBL Model, which originated at a meeting at the Marine Biological Laboratory (MBL) in Woods Hole, Massachusetts. The resulting computer program simulates the possible fate of a lineage. It can either die out or evolve, with or without speciation (Sepkoski 2016). As historian David Sepkoski commented, this model “was an application of a randomization process known as a Monte Carlo simulation. The computer was used to randomly draw numbers to determine outcomes with prespecified possibilities, much as a dealer might randomly draw cards from a deck and arrange the outcomes into hands. At the end of the run, the program output the results graphically in the form of a branching phylogenetic tree” (Sepkoski 2012a, 2228). Computer-based simulations present possible and highly contingent historical scenarios thus offering working hypotheses regarding evolutionary processes. They offer possibilities for intervention in a deeper temporal dimension that would otherwise have remained inaccessible.

A similar approach can be found in the study of the morphology of extinct organisms. American paleontologist David Raup pioneered this field of investigation. In the early 1960s, following D’Arcy Thompson’s analogical analyses of form development, Raup identified four parameters responsible for the shells of coiled gastropods and used a “digital computer with automatic plotting equipment […] to make graphical reconstructions of a shell from any given values of the four parameters” (Raup 1962, 150). He then presented a hypothetical snail form simulated according to four previously identified parameters. In this case, form resulted from growing processes. In addition, he conceived of a virtual space (the so-called morphospace) in which all theoretically possible shapes could be simulated starting from the parameters responsible for their formation. The particularity of a morphospace lies in the way the forms were visualized. As Raup put it, these “can be combined to define a ‘four dimensional’ space which contains most of the theoretically possible shell forms. When the geometries of naturally occurring species are plotted in this space, it becomes evident that it is not evenly filled. Evolution has favored some regions while leaving others essentially empty. In the empty regions, we are presumably dealing with forms which are geometrically possible but biologically impossible or functionally inefficient” (Raup, 1965) (Fig. 4).

Fig. 4
figure 4

Raup’s morphospace of possible shell forms. See, Raup, David M. “Geometric Analysis of Shell Coiling: General Problems.” Journal of Paleontology 40 (1967): 1178–90.

Full size image

In 1969, Raup co-authored another seminal paper in which he combined his computational ability to simulate theoretically possible shape processes, i.e., Raup’s use of morphospace, with the morphological viewpoint adopted by German paleontologist Adolf Seilacher. Their idea was that “meander patterns produced by ancient sediment feeders can be simulated by a digital computer with x-y plotter output” (Raup and Seilacher 1969, 994). Assuming a hypothetical animal, the program could sense its immediate surroundings and convert the resulting information into behavioral instructions. However, the interesting part of their computer-based design was that “changing behavioral controls from one run to the next causes more pronounced differences in the model, which are comparable to the differences actually observed between species and genera” (ibid.). Raup and Seilacher concluded that “many more species models could be simulated with little or no program processing. This would lead to a better understanding of the change in parameters required to transform one variant species into another and would also have some effect on the classification of these trace fossils” (ibid.). Thus, shape transformation resulted from changing parameters in the morphogenetic program. This, in turn, could be studied through computer simulation.Footnote 8

As for the use of images described in Sect. 2, in this case, visualizations of the past do not merely reproduce something that happened in the past nor make visible different aspects of a possible explanandum. Instead, they present new kind of phenomena, thus extending paleontologists’ access to deep time. Lorraine Daston and Peter Galison defined this activity as follows: presentation is “no longer necessarily focused on copying what already exists – and instead becomes part of a coming-into-existence” (Daston and Galison 2007, 383). By computing the past, paleontologists were able to create new phenomena, in effect “making real a scene or a process” (sensu Rudwick) not otherwise visible.

6 Virtual Reality, CT Scans, and Robots

In this section, I will take a step forward and explore how technology and different use of visualizations are essential not only for creating a possible access to deep time (as in the cases analyzed above), but also for generating new research questions. Following the architect Kostas Terzidis, I will examine the transition from the computerization of paleontology to computer-based paleontology. According to Terzidis, the computerization of a discipline is characterized by the insertion, manipulation, or storage on a computer system processes already conceived in terms of the project (Terzidis 2004, 2006). Computerized research allows scientists to ask new questions and possibilities that would not have been possible with classical technologies. To examine this transition, I will now turn to the characteristics of recent virtual paleontology.

As explained at the beginning of this chapter, fossils are imperfect and incomplete. They are also extremely fragile and difficult to separate from the rock in which they were fossilized. Some more or less effective techniques have been developed to free fossils from their rocks. The classic approach is to use tools to physically remove the fossil. Alternatively, chemical preparations can be used to try to remove the fossil from the rock, or the fossil can be chemically dissolved to release a cast that can in turn be filled. In all these cases, however, there is a risk of fracture or deformation of the fossil and, above all, the internal morphology of the inaccessible remains. The insertion of recent technology into the field, called virtual paleontology, provides new solutions to these problems. Today, fragile fossil material is displayed on the computer screen after being scanned. These new methods make it possible to experience the microcosm of the fossil and view it from all directions. In addition, they make it possible to perform a range of activities and investigations that would not otherwise be possible in a more traditional paleontological practice. For example, the scientist can non-invasively zoom in and out and rotate objects directly on their computer screen. The three-dimensional virtual visualizations of the material constitute the basis for descriptive publications and for further experimental and quantitative approaches to the evolutionary dynamics of organisms in deep time. In addition, the data generated by 3D scans and visualizations can be easily manipulated, reproduced, and shared with other scientists. Although there are still many problems associated with sharing fossils between institutions,Footnote 9 Cunningham and colleagues enthusiastically explain that “digital datasets have been touted as a panacea for the problems of limited access to fossil specimens. In principle, they can be shared online to make them available to the entire community” (Cunningham et al. 2014, 353). Digitization has the potential to bring the “openness enjoyed by other biological sciences” to paleontology” (ibid.).

Through digitization, fossil structures can also be made visible that could not otherwise be accessed without damaging the original. This technology-driven virtual approach “is not just a solution for problematic material, but a powerful new set of techniques for re-viewing fossils preserved in three dimensions” (Garwood et al. 2010, 96–97). 3D modeling of fossils, therefore, allows for technical access and intervention on the workings of individual structures of fossilized creatures. As Cunningham and colleagues assert, “Computer-assisted visualization and analysis of fossils has revolutionized the study of extinct organisms” (Cunningham et al. 2014, 347).

Furthermore, augmented reality is also being used to visualize the fossil record on the computer screen and merge it with other specimens. In this case, “a smartphone, a tablet, or any device with a camera and internet connection would suffice to grant access to the actual 3D morphology of the specimens and their tiniest features” (Bartolini Lucenti et al. 2020, 11; Bartolini-Lucenti et al. 2021) (Fig. 5).

Fig. 5
figure 5

QR code and augmented reality marker used by the scientists to reproduce the comparison of teeth morphologies of C. etruscus from Upper Valdarno (blue), Canis from Dmanisi (red), and extant C. lupus (grayish). See, Bartolini Lucenti et al. 2020, 10

Full size image

To analyze how virtual paleontology is actually changing the way the geological and evolutionary past can be both accessed and studied, I will focus on two case studies. The first took place at the Naturkunde Museum Berlin. In recent years, the skeleton of Branchiosaurus Brancaii has been studied using virtual paleontology techniques. In a study published in 2020, paleontologists focused in particular on the morphology and musculature of the tail. The classical method of inferring the muscle-skeletal structure of an extinct organism is based on a comparison of this structure with that of evolutionarily related animals. In the case of dinosaurs, for example, paleontologists have used the morphology of birds and crocodiles as a possible comparison. The paleontologists of the Museum für Naturkunde in Berlin accepted this classical method of comparison and implemented it in the context of virtual paleontology. They digitally reconstructed the tail of this sauropod using “photogrammetric 3D digitization and 3D modeling tools in combination with information provided by dissections of extant crocodilians (Alligator mississippiensis) […] and an Extant Phylogenetic Bracket approach […] comparing the anatomy of the caudal vertebrae and muscles of Giraffatitan with that of extant crocodilians” (Díez Díaz et al. 2020, 2). Following these analyses, the paleontologists were able to determine that Branchiosaurus Brancaii had an extremely powerful tail that played a functional role in the body: It helped “in its stabilization, and propulsion, but also as counterweight for the presacral part of the body” (ibid., 14). In addition, they determined that the total weight of the tail was about 250 kg. These results are extremely significant because the musculature of an extinct organism is never fossilized; since the muscles are made of soft material, they are not preservable.

By using 3D photography and various modeling techniques, paleontologists are able to gain access to something that is not a given. They used these results as possible platforms to experiment and work with deep time. Then, by examining the fossil virtually, new research questions were asked about the posture, strength, and biomechanical properties of the real fossil. In other words, reality (the material specimen) and virtuality (what can be visualized through different technologies) merged and supported each other. The resulting “methodology allows a better-constrained reconstruction of muscle volumes and masses in extinct taxa, and thus force and weight distributions throughout the tail, than non-volumetric approaches” (ibid., 1).

Another emblematic case of the dissolution of boundaries between biology and technology is the use of virtual methods to overcome the lack of fossilized information. As explained above, fossils are inherently imperfect and incomplete. In addition, the process of fossilization can cause severe deformation. Using virtual paleontology techniques, paleontologists examined the severely deformed fossil of Equus stenonis, one of the most widely distributed horse fossils of the European Pleistocene. Fossil horses are often deformed in the skull due to their long and thin morphology and the presence of multiple pneumatic sinuses that occur inside the skull. Paleontologists introduced “a new virtual reconstruction protocol, termed Target Deformation, which takes advantage of recent progress in digital restoration of fossil specimens […] digital alignment of disarticulated portions […] and 3D thin plate spline transformation (tps3d) […] to provide the virtual reconstruction of badly deformed, partially incomplete cranial material by using target fossil remains of the same species” (Cirilli et al. 2020, 2). Target deformation aims to eliminate asymmetric changes due to taphonomic processes by applying a set of corresponding bilateral reference points.

Like in the case of the Brachiosaurus brancai muscle reconstruction, paleontologists in this study used technology in addition to fossil remains of other organisms to merge the virtual with the real. In this case, two fragmentary but well-preserved skulls of Equus stenonis from Olivola, Italy, and Dmanisi, Georgia, were used. Paleontologists showed that the 3D virtual reconstruction protocol used was “capable of virtually restoring a severely crushed specimen by using partially complete skull specimens, such as the fragmentary E. stenonis skulls from Olivola and Dmanisi” (ibid., p. 9).

One last element should be briefly mentioned here. Presently, the visual culture of earth science is merging with an engineering approach to morphology. As a result, paleontologists are teaming up with engineers to design robots that resemble extinct organisms. This robotic fabrication of extinct forms allows scientists to indirectly experiment and manipulate the deep past. The rationale of this practice is to create robots able to generate further biological questions that wouldn’t have been possible otherwise. In this process, the robotic and the digital merge in the materialization of the morphology of extinct organisms. This integration has been recently labeled the “twenty-first-century material turn in the study of form” (Tamborini 2021, 2022b).

7 Conclusion: Visual Cultures and the Dynamics of Knowledge Production

Paleontology, geology, and other earth sciences address “configurations that [can] not be adequately conveyed by words or mathematical symbols alone” (Rudwick 1967, 151). Within this arena, different kinds of visualizations support earth scientists’ reasoning, in effect enabling scientists to work with deep time. In this process, science, theory, and technology are fused together to produce knowledge. With the imperfect and incomplete record of the past as a point of departure, scientists have developed a set of practices to both represent and present deep time. Technology has allowed for researchers to overcome the imperfection and incompleteness of the record of the past. As a result, a working version of the past is produced which allows scientists to think with pictures.

At its core, the visual culture of paleontology is based on circulation of skills, practices, and technologies between different disciplines.Footnote 10 For example, several circulations of practices took place between bureaucratic sciences and paleontology. The same can be said for the exchange of knowledge between paleontology and human history as well as for paleontology and engineering sciences. This has two main implications. First, paleontology, and broadly earth sciences, can be seen as methodological “omnivorous disciplines” (A. Currie 2015). They take up and implement every possible set of practices, visualizations, and tools to make real and visualize something otherwise non-perceptible. Second, as Rudwick (1967), Hentschel (2014), Sepkoski and Tamborini (2018) have aptly indicated, the transition of knowledge is made possible by scholars who were between different disciplines and therefore could pre-adapt and translate a particular visual culture, say the bureaucratic one, into another discipline such as paleontology or geology. This transition of media and visual practices characterized not only the development of earth sciences, but it also was a common issue in other disciplines.

Second, as my phenomenology of visualizations has indicated, every medium and visual tool has its own specific epistemic and ontological power. The images of extinct organisms used for taxonomic scopes have a quite different appeal, function, and epistemic power from the form simulated and visualized through a morphospace. The former represents the past, the latter presents a possible platform for creating the conditions for the manifestation of deep-time phenomena. The same can be said for the different modalities of working with data variously visualized and diversely visible. As seen, there is a wide range of epistemic possibilities when working with data previously listed, tabulated, or put into a graphical representation: they convey their representational meaning in quite different and sometimes complementary ways.Footnote 11

However, although every visualization conveys its meaning differently, diverse visual devices should be put into relation to efficiently work with the past. One peculiarity of the earth sciences is that both the fossil record and its representation can be considered to be both data for further works and the sort of phenomena the scientists seek to explain (the so-called explanandum). In paleontology, data, such as fossilized bones, a displayed dinosaur, or the diversity curve generated through a database (such as the phanerozoic diversity curves generated by Heinrich Bronn or similar computer-based enterprises produced in the contemporaneity), are data used for understanding broader biological phenomena (such as the broader patterns and mechanisms of adaptability or biodiversity). At the same time, this data is what paleontologists seek to explain – they try to figure out why an organism had this particular form or why the data show a logarithmic instead of an exponential curve. To handle this dual function and the related epistemic and practical issues, scientists adopt a continuous flow of representations in which the visualized object becomes the starting point for generating new phenomena and visualizations, which in turn become the objects to be explained.

Third, knowledge production and visualizations are tied with the use of different technologies – ranging from pen and paper, through computers, to robots and virtual and augmented reality. Therefore, to understand how knowledge is produced and visualized in earth sciences, the productive intersection between technology, background theory, and images should be taken into account. Earth sciences’ visual culture cannot be fully understood if studied in isolation. Rather, image production should always be put into the broader technological and theoretical framework adopted by the specific earth science disciplines.

Last, by working on the different modalities of visualizing the past, paleontologists and earth scientists often produce images which have a strong aesthetic appeal. The structures and beauty of fossilized structures have been used as guiding model to design buildings or as the inspiration for paintings. Geological and stratigraphic maps have inspired literary and poetic works. What is worth emphasizing from this is twofold. First, the aesthetics of the past have paved the way for many interdisciplinary and joint scientific projects. These were set up to understand how technical form could be best represented and built emulating the form-function-beauty complex of deep time organisms. In turn, paleontologists took up from these joint projects a nuanced engineering perspective toward organism construction. This has enabled paleontologists to develop new practices and vocabulary to visualize the past. The role of aesthetics as a binding element between different practitioners and scientists from diverse disciplines is a further element of the visual culture in earth science.

The aesthetics of the past fosters also its possible epistemic investigation. The aforementioned Tendaguru expedition provides one emblematic example for the intersection between aesthetics and knowledge production. Once the giant bones were transported to Berlin Museum of Natural History, they were vertically displayed in a museum main exhibition room, evoking all the majesty and beauty of the columns of an ancient temple. The director of the Museum used this to start a successful crowdfunding campaign to support the expedition and broadly paleontological work. Hence, as Turner rightly noted, “Collection decisions are heavily (though not exclusively) guided by aesthetic norms and goals. Some of the reasons not to collect a Hadrosaur fossil, […], are distinctively aesthetic” (Turner 2019, 13). This point implies what Turner called a rejection of the bias toward the epistemic.Footnote 12 He did so “by showing that paleontological research is a form of aesthetic engagement with fossils and with landscapes” (ibid., 1). Particularly, he argued that “aesthetic engagement and historical investigation are mutually facilitating” (ibid., 29).

Hence, and to conclude, the aesthetics of deep time should be also considered as relevant components when working toward an understanding of the role and function of different media and technologies in producing and sustaining deep-time knowledge production. As a result, earth sciences’ visual culture intersects with the history of aesthetics, design, technology, and biological sciences. To fully comprehend this visual culture, a transdisciplinary effort is required.