Epistemological troubles at the museum

Text Michael Støen

There is a hidden tension that surfaces when modern science is communicated and visualized. Michael Støen discusses how this tension reminds us of the instability of the current epistemological paradigm.
Museum of Natural Historiy, New York, // Foto: Aditya Vyas/Unsplash
Over the course of two and a half millennia material reality seems to have gained epistemological ground over immaterial reality. Plato believed that ideas were more real than objects and subsequently did not consider the physical world to be a valid source of knowledge – a notion which, to a degree, persisted through the Middle Ages in the form of Christian divinity. The epistemological grip of Catholicism would, however, eventually loosen throughout the Renaissance and the Reformation and gradually give way to the ideas of 17th century thinkers like Descartes and Bacon.
Descartes described nature as a mechanical realm of blind causality yet he still held a special place for the immaterial aspects of consciousness. His dualism, however, impacted the philosophical trajectory for centuries to come, which eventually arrived at a firm and hard empiricism – the beginning of modern science. In broad terms the hard sciences of today still hold the same mechanical view of nature that Descartes espoused, the main difference being that they have abandoned his special place for consciousness which is instead simply seen as a causal consequence of the exceedingly complex machinery of material organisms. The mind has become part of the machine, and knowledge has ultimately become a product of the physical. On the other hand, the mind has become a part of the machine which represents the personal agenda of some individual, and so, in that sense, is seen as a less reliable source of information than the rest of the material world.
The late modern period has, of course, offered a fair share of challenges to this materialistic view. Heidegger, to pick a highly influential example, with his focus on being (Dasein), was not ready to grant physicality any special epistemological status in comparison to consciousness. In fact he was not even willing to accept that a categorical separation between object and subject was a meaningful distinction. Yet for many a modern person and certainly for many a hard scientist materialism is still the predominant perspective.
It would, however, be naive to assume that this struggle has found its final historical state, despite the fact that physicality currently has the upper hand. Practices of scientific visualization, for example, still often offer interesting manifestations of these epistemological tensions, as they necessarily deal with both the materiality of objects and (to varying degrees) imaginative representations of these objects which require the immateriality of human subjectivity to be produced. What, then, can current historical studies on scientific objects and their visual representations tell us about this struggle?

Dinosaurs on display
In “Bringing Dinosaurs Back to Life” (2012) Lukas Rieppel explores the tensions between the “conflicting goals” of American natural history museums, specifically in regard to the displays of dinosaur fossils in the early 20th century. [1] In doing so he identifies an interesting problem that the museums faced: how does one display an incomplete physical scientific finding in a way that is both engaging yet does not appear speculative?
In identifying this issue Rieppel applies the useful dichotomy, originally coined by Charles Sanders Pierce, of “indexical” vs “iconic” representations. [2] Rieppel defines indexical representations as those which are, in some way, a direct result of the object they represent. Iconic representations, on the other hand, are indirect recreations of the object they represent. A painting of a dinosaur, for example, is iconic, whereas an exact replica of a dinosaur fossil, say, made by creating a mold and filling it with liquid material which solidifies, is indexical. Rieppel uses a colloquial example to clarify: “a snow angel indexically represents the motions of a child at play, but it is an iconic representation of biblical creatures.” [3] An important point to this distinction is that indexicality is defined as a product of material causation whereas iconicity is defined as a product of human consciousness. This means that a display can be both indexical and iconic to varying degrees. A replica of a fossil which has been mounted in a life-like position to suggest the movement of the dinosaur, for example, is indexical in its basic form but iconic in that the particularities of its life-like positioning are the product of human imagination, which often includes a fair amount of speculation regarding the animals motivations, behaviours and physical dynamics in a living state.

A very particular example, in Rieppel’s work, of such a tension between indexical and iconic aspects which museum curators struggle with has to do with the fact that dinosaur fossils are often incomplete and therefore require manufactured pieces, and/or bones from other fossils, in order to be mounted into complete displays. [4] This means a certain amount of human imagination is infused into a display and its indexicality diminishes while some iconicity is introduced. Rieppel describes how attempts were made to navigate this situation, for example by marking the manufactured pieces in a manner which would allow experts to distinguish them while minimizing the risk that casual visitors would notice the difference—an approach which was criticized by scientists as deceptive. [5] Such tensions may seem relatively minor from an outside perspective but the picture Rieppel ends up painting, in fact, is one of absolute fear on the behalf of curators that their displays should in any way be identified as products of human ideas rather than pure presentations of physical findings:

Although curators at the American Museum recognized the power of icons to induce vivid imaginative experiences, they were loath to display products of human consciousness. Curators worried that icons were vulnerable to distortion by the subjective and perhaps even erroneous beliefs of whoever had fashioned them, thereby calling the museum's authority to speak for prehistoric nature into question. [6]

Historically, as historians of science Lorraine Daston and Peter Galison have made clear, this tendency to fear human intervention in scientific visualization is a relatively new development. [7] Previously, human intervention was seen as a positive necessity. For example the intuition and knowledge that a veteran naturalist had developed over time and which was infused into their “iconic” natural illustrations was something to be trusted as fact. An important aspect to this point, which connects this observation to the broader historical context, is that these illustrations of naturalists were not meant to represent the particularities of some singular physical specimen. The imperfections of whichever particular specimen was serving as a model for the illustration would be worked around by the naturalist, using their mental image which had been acquired through looking at many such specimens. As Rieppel points out, in reference to Daston & Galison, the purpose was to illustrate the “essence” of the species rather than the particularities of the specific individual. [8] Essence is, of course, something immaterial.

Works of George Dyonysius Ehret.

This approach is perhaps most firmly epitomized in Carl von Linné’s botanical work Hortus Cliffortianus and its luscious illustrations by the German botanical artist George Dyonysius Ehret. [9] Hortus Cliffortianus was first published in 1737, and the “truth-to-nature” period, to which Daston and Galison refer, stretched from this period into the early nineteenth century. It took place, then, during a time in which the broader cultural shift mentioned in the opening of this article—the shift away from epistemic value being put into immateriality rather than materiality—was already well into it’s arch. Yet despite the truth-to-nature scientists' dedication to empiricism and trained observation, it seems that there still existed, at that point, a lingering Platonic notion that the objective truth of a thing lies in its “idea” or “form” rather than its physicality. Furthermore, it was the knowledge and expertise of the naturalists which enabled them to illustrate that truth, whereas by the mid 19th century the truth was instead to be located in the particularities of the specific object itself, and the contribution of human consciousness (even that of experts) was something to doubt and consider biased or speculative rather than a mark of authority.

Daston and Galison describe how, in the broader scientific context, this new development, in combination with technical advances, led to a preference for indexical visual representations of objects that were, using various technologies, imprinted by causality rather than created by human imagination—something which they refer to as “mechanical objectivity”. [10] Again, Rieppel picks up on this notion in his analysis of museum displays:

Among other things, this manifested itself in a widespread enthusiasm for technologies of the index like photographs, X-rays, and the camera obscura. [...] The widespread appeal of “mechanical objectivity” during the late nineteenth century can thus be understood as an attempt to “extirpate human intervention between object and representation.” [11]

Yet despite this enthusiasm for indexicality the actions of curators seem to suggest that the communication of facts is not really possible by such means alone. Although museums attempted to counterbalance their iconicity by “highlighting the indexicality of certain materials (such as fossils) over others (such as plaster),” they never came anywhere near abandoning iconic representations. [12] Rieppel even goes as far as to describe museum displays as “mixed-media installations”.

Making models of archaic humans

While Rieppel’s observations are historical, and while the enthusiasm for mechanical objectivity is described by Daston and Galison as having now given way to a more relaxed epistemic period which they call “trained judgement”, the tensions Rieppel discovered do not seem to be a mere phenomenon of the past but can still be seen in the mixed-media installations of today’s museums. Take for example the sculpted, life-sized reconstructions of archaic and early modern humans that are currently on display at various museums around the world; the most famous of these are created by the twin hominid paleo-artists Alfons and Adrie Kennis, who The Guardian describes as “among Europe’s most sought-after – and controversial”. [13] The Kennis brothers’ reconstructions are materialized through an epistemically perplexing mix of indexicality and iconicity, a process which begins with a focus on the indexical: they scan original fossils into digital models which they then print with 3D printing technologies to create life-size duplicates. As they are often forced to combine scans of several incomplete fossils in order to print a complete skeletal reconstruction, however, even this highly indexical process involves a similar influx of iconicity as that of the dinosaur displays described by Rieppel which required combining different fossils as well as adding sculpted plaster bits to create the illusion of a whole skeleton. Nevertheless these skeletal reconstructions are largely created through indexial means.

The Kennis brothers. // Foto: Simon Claessen/Creative Commons

From this point on, however, the process is entirely based on iconicity, although there are different levels of subjectivity infused at various stages. As The Guardian points out, the brothers’ “career has coincided with huge leaps in DNA testing” enabling them to “accurately recreate genetic traits such as hair type and eye colour” and skin color. [14] The visualizations of these particularly visual facts however remain iconic as the phenotypic expressions of the DNA are long gone and cannot be photographed or mechanically copied in any way. Instead the brothers use their subjectivity and creativity to visualize the scientific knowledge they have acquired through various modeling materials. There is then yet another layer of iconicity applied, which is further removed from the hard facts while still somewhat informed by their anthropological expertise:

By studying people from more isolated or primitive societies, the brothers believe they can see through a window into the past. […] For this reason, Kennis models often exhibit unique hairstyles and tribal tattoos, and strike poses that appear startlingly modern, provoking strong reactions. [15]

Adding to this the brothers give their models extremely individual and human expressions in yet another layer of iconicity which is entirely untethered to the hard facts and taken rather from their own experiences and sensibilities. Still, this could be argued to be the “truest” aspect of their models. It is what makes seeing them feel like actually looking at prehistoric individuals full of emotions and thoughts, as opposed to mere scientific specimens. Interestingly, while the truth-to-nature naturalists applied subjectivity to get at the essence of a specimen rather than the particulars of an individual, the subjectivity here is applied for the exact opposite reason; the brothers don’t “simply depict a generalised early man, but a specific man or woman, an effect that allows onlookers to glimpse human prehistory with immediacy, even familiarity.” [16]

In some ways these contemporary museum objects can be seen as expressions of the fact that the obsessions with indexicality which were evident in the period of mechanical objectivity have subsided, as Daston and Galison describe, yet when looking at the ways in which museums discuss and present such objects publicly it is clear that aspects of the tensions described by Rieppel persist. The presentation of the 2018 Kennis model of “Cheddar Man”—the oldest near-complete skeleton of a human ever found in Britain—created by the Kennis brothers for the London Natural History Museum, is a good example of this; while the museum acknowledges that the model is “part art and part science” they are keen to emphasize the indexical aspects of the object, pointing out how the “artists took measurements of the skeleton, scanned the skull and 3D printed a base for their model” yet neglecting to describe in detail how the subjective aspects were incorporated. [17] Media coverage, in 2018, of the newly unveiled sculpture also showed a similar tension of wishing to minimize the subjective creativity infused in the model and highlight instead the indexical aspects. [18]

The downplaying of subjectivity in the communication of such models, in order to maintain scientific authority while still affecting visitors effectively, exemplifies the tense relationship that natural museums still have between the need to communicate knowledge through iconicity and artistry yet still project objectivity through indexicality. Clearly, much knowledge of what archaic and early modern humans, like Cheddar Man, might have looked like only exists in the minds of experts and therefore such highly iconic models do not merely communicate subjective opinion but have real factual value that cannot be located in the physical aspects of the scientific objects they represent. On the other hand, the amount of subjective imagination involved in the creative modelling process makes it hard to avoid second guessing much of the factual value such objects embody.

Neurons by Santiago Ramon y Cajal (1852-1932).

Drawing neurons

Perhaps a more constructive exploration of this dichotomy can be found in the article “Drawing into Abstraction”, by Sarah de Rijcke (2008), in which she discusses the techniques developed by the Nobel prize-winning neuroscientist Santiago Ramon y Cajal (1852-1932) for visualizing neurons. As an expert in both microscopy and illustration Cajal utilized the indexical knowledge that mechanical techniques could offer but also overcame their limitations through iconicity. Because neurons extend in three dimensions the very short focal depth of the microscope could never fully see the phenomenon in one setting. [19] Rather a scientist would assemble an accurate idea of the phenomenon through constantly shifting the microscopes focal points. Apart from such limitations it also took an extremely trained eye to even know how to look at what was being presented through the scope at all. Cajal’s process would result in a compiled mental image of the neuronal structures, rather than one recorded through direct indexical means, which he would represent in his carefully constructed drawings. This process also enabled him to leave out any confusing or unimportant visual phenomena which would have been present in a photo.

De Rijcke reveals how Cajal’s expertise, in combination with his drawing skills, resulted in very flat and clear illustrations of neurons which, through the expert’s mind, had been filtered into an abstraction that could communicate more knowledge than a photograph of a specific microscopic setting might have. [20]

In the drawings, Cajal brought out the fundamental aspects, and played down superfluous detail. These decisions were based on what he saw as the complete 'mental image' of the nervous tissue at hand. He quite literally reduced highly complex, dynamic viewing experiences into flattened, static, comprehensive images of neurons. Cajal's drawings were material articulations of an expert selection process. [21]

Cajal’s utilization of iconicity, however, was not meant to achieve idealistic imagery but quite the opposite. Despite the fact that he felt photography was an inferior representational technology in comparison to illustrations he was a strong proponent of the idea of objectivity and believed his process contributed to the minimization of subjective interference which he accused some of his colleagues of succumbing to. Cajal, in fact, is used by Daston and Galison as an example of a particularly complex type of proponent of mechanical objectivity, as he promoted strongly its epistemic values of subjective restraint and the rejection of notions of essence and the ideal, yet found in his case that purley indexical technologies of visualization were unsatisfactory in reproducing the mental image of an expert who practiced those values.

Cognitive dissonance among curators

Which brings us to our question of what current historical studies on scientific objects and their visual representations can tell us about the historical epistemic struggle between physicality and immateriality. Rieppel’s description of the fearful state of museums underscores the premise of that question: that the modern mind would rather put its epistemological faith in physical objects than the immateriality of consciousness. The difficulties faced by museums in doing so, on the other hand, reveal an ongoing philosophical tension. It is almost as if curators suffer from a kind of cognitive dissonance. On the one hand they are committed to an epistemology of physicality (their appeal to indexicality and mechanical objectivity), on the other hand they see more value in communicating knowledge through iconicity than simply presenting scientific objects in their naked physical form. Perhaps this is because they intuit something close to what De Rijcke illuminates in her article—that there is knowledge in an expert’s “mental image” which is not available to the rest of us when we look at what they are looking at. In which case, when it comes to displaying an object, there might be epistemic value in trying to represent what it is that the expert “sees” rather than what is physically there. A paleontologist looking at an incomplete set of bones lying on a table might, because of their knowledge, mentaly see something closer to what the average person sees when looking at a mounted museum display, or reconstructed model, with all its imaginative iconicity.

Human subjectivity has its downsides when it comes to the production and communication of knowledge. It can certainly be limited and biased. The value of minimizing such downsides is something which the development of the scientific method has championed. When put into a historical context, however, the tendency to attribute factual knowledge to mechanical, indexical, and purely causal modes of representation is a notion which reveals itself to be less stable than one might think. Studies on the visual cultures of science can contribute to a more complex understanding of our modern use of the term “objectivity”, and in the broader sense, to a more complex understanding of the history and development of science, knowledge, and epistemology at large. Despite a decisive historical movement from truth being located in immaterial realities toward truth being located in the physicalities of materialism, the epistemological values of subjectivity and iconicity persist.


  1. Lukas Rieppel, "Bringing Dinosaurs Back to Life: Exhibiting Prehistory at the American Museum of Natural History", Isis 103.3 (2012): 460.

  2. Rieppel (2012): 462.

  3. Rieppel (2012): 462-463

  4. Rieppel (2012): 470.

  5. Rieppel (2012): 477.

  6. Rieppel (2012): 463.

  7. Lorraine Daston and Peter Galison, Objectivity, (New York: Zone, 2007): 55-105.

  8. Rieppel (2012): 464.

  9. Daston and Galison (2007): 55.

  10. Daston and Galison (2007): 115-183.

  11. Rieppel (2012): 463.

  12. Rieppel (2012): 465.

  13. Kit Buchan, "Meet the ancestors… the two brothers creating lifelike figures of early man", The Guardian 2018-05-05. Accessed 2020-05-16. https://www.bbc.com/news/science-environment-42939192

  14. Buchan, The Guardian 2018-05-05.

  15. Buchan, The Guardian 2018-05-05.

  16. Buchan, The Guardian 2018-05-05.

  17. Kerry Lotzof, “Cheddar Man: Mesolithic Britain's blue-eyed boy”, NHM (website, n.d.). Accessed 2021-05-10. https://www.nhm.ac.uk/discover/cheddar-man-mesolithic-britain-blue-eyed-boy.html

  18. See, for example, Paul Rincon, "Cheddar Man: DNA shows early Briton had dark skin", BBC 2018–02–23. Accessed 2020-04-13. https://www.bbc.com/news/science-environment-42939192

  19. Detta är det första föremålet i din lista

  20. Detta är det andra föremålet i din lista

  21. Detta är det tredje föremålet i din lista


Buchan, Kit. "Meet the ancestors… the two brothers creating lifelike figures of early man", The Guardian 2018-05-05. Accessed 2020-05-16. https://www.bbc.com/news/science-environment-42939192

Daston, Lorraine, and Peter Galison. Objectivity. New York: Zone, 2007.

Lotzof, Kerry. “Cheddar Man: Mesolithic Britain's blue-eyed boy”, NHM (website, n.d.). Accessed 2021-05-10. https://www.nhm.ac.uk/discover/cheddar-man-mesolithic-britain-blue-eyed-boy.html

Rieppel, Lukas. "Bringing Dinosaurs Back to Life: Exhibiting Prehistory at the American Museum of Natural History". Isis 103.3 (2012): 460–90.

Rincon, Paul. "Cheddar Man: DNA shows early Briton had dark skin", BBC 2018–02–23. Accessed 2020-04-13. https://www.bbc.com/news/science-environment-42939192