After three months of runtime and using over 8,000 compute cores, a set of cosmological hydrodynamical simulations called Illustris was complete in 2013. Spanning over 13.8 billion years of cosmological evolution, this set of simulations represents a region of the universe in a cube (fig. 18.1). In the largest simulation, the side length of the simulation cube is 106.5 megaparsecs (a unit used to measure astronomical distances), which corresponds to 350 million light years.1 This scale means that the simulation can represent the universe as a whole. We focus on visualisations based on Illustris, tying the question of mimesis to recent visual representations of the cosmos.
This visualisation represents the large scale structure of the universe, known as the âcosmic webâ. The density of the so-called âdark matterâ is shown here as the pink filaments, where the brightest areas represent the most dense concentrations of dark matter in Illustris. Original caption: âExterior view of the dark matter density distribution in the full Illustris-1 box at redshift zeroâ (Illustris Collaboration, 2015b)
Credit: Illustris Collaboration, courtesy of Mark V ogelsberger
Illustris is the product of an international collaboration based primarily at Massachusetts Institute of Technology, Harvard-Smithsonian Center for Astrophysics, and Heidelberg Institute for Theoretical Studies. According to the Standard Model of Cosmology (also known as the LCDM model), normal (or baryonic) matter â that of which the Earth, stars and galaxies are made â is believed to account for only 5% of the cosmos, whilst the invisible dark matter and dark energy are thought to take up 26% and 69%, respectively.2 Illustris marks a breakthrough in its realistic reproduction of a range of phenomena at different scales. Supermassive black holes, galaxy formation, and the large-scale structure of the universe come together in the simulation representing both visible and invisible matter as it develops over time. The output from simulations such as Illustris is often referred to as âsynthetic dataâ, and in the case of Illustris takes up 200 terabytes of storage.3 To perform research based on Illustris, an astrophysicist would typically download only part of the data output. This would then be stored as a matrix describing the characteristics and coordinates of particles in the three-dimensional space of the simulation cube (as an example, see fig. 18.2). The particles of the virtual universe of Illustris represent phenomena in the universe such as stars, dark matter, or black holes, and these particles have different properties depending on what they represent. The largest simulation, Illustris-1, contains 6,028,568,000 particles representing dark matter alone. This can be overwhelming, even to the trained eye of an astrophysicist working with computational theoretical physics.
Example of synthetic data output from Illustris
Photo: Martin Sparre
Several theorists have described big data as âmessyâ.4 As the historian Orit Halpern (2015, 5) writes, âdata is not always beautiful. It must be crafted and mined to make it valuable and beautifulâ, recalling the Greek verb kosmeÅ, as described by the classics scholar Gregory Vlastos (2005, 3):
In English cosmos is a linguistic orphan, a noun without a parent verb. Not so in Greek which has the active, transitive verb, kosmeÅ: to set in order, to marshal, to arrange [â¦] In the Greek the affinity with the primary sense is perspicuous since what kosmos denotes is a crafted, composed, beauty-enhancing order.
Vlastosâ explanation of kosmos could certainly be relevant in relation to the visual dimension of Illustris, where the overwhelming matrix of numerical values is transformed into colourful visual representations showcasing the virtual universe. Halpern (2015, 21) reminds us that âIn the present, visualization is often understood not only as a process but also an object, a subject and a discipline, a vocation, a market, and an epistemologyâ. Despite the attention given to cosmological simulations in journals, magazines, preprint articles, the news, and social media, many of them have yet to be explored in existing scholarship within the humanities. Taking this as our cue, the present chapter will read visualisations based on the Illustris simulation against the theme of mimesis, as it appears in Platoâs corpus, with an emphasis on the Timaeus. Mimesis here is seen on a cosmological scale, which is helpful for tackling the representational questions that are being rearticulated with a project such as Illustris. This chapter will examine visualisations of both normal and dark matter in Illustris, in order to show how the notion of mimesis can be used as a vehicle for unpacking these visualisations. We will employ the ambivalence of mimesis in Platoâs corpus, from Republic X to the Timaeus, to qualify and inform the discussion of the visualisation of cosmological simulations. As the art historian Ernst Gombrich (1960, 83) writes in Art and illusion (1959), âThere are few more influential discussions on the philosophy of representation than the momentous passage in the Republic where Plato introduces the comparison between a painting and a mirror image. It has haunted the philosophy of art ever sinceâ. In the Timaeus, however, the demiurge, a divine craftsman thought to be the personification of reason, transforms chaos into cosmos by giving shape to the visible world, based on the harmonic proportionality of the eternal model.5 Mimesis is often translated to âimitationâ in concise definitions.6 However in Platoâs work, mimesis plays a range of roles. As the classics scholar Stephen Halliwell (2002, 70â71) writes in his attempt to âdiagnose [â¦] Platoâs prolonged and profoundly ambivalent relationship with mimesisâ, there are two main ways in which mimesis is used in Platoâs corpus.
The first, a kind of ânegative theologyâ, which leads sometimes in the direction of mysticism, is that reality cannot adequately be spoken of, described, or modeled, only experienced in some pure, unmediated manner (by logos, nous, dianoia, or whatever). The second is that all human thought is an attempt to speak about, describe, or model reality â to produce âimagesâ (whether visual, mental, or verbal) of the real. On the first of these views, mimesis, of whatever sort, is a lost cause, doomed to failure, at best a faint shadow of the truth. On the second, mimesis â representation â is all that we have, or all that we are capable of. In some of Platoâs later writing this second perspective is expanded by a sense that the world itself is a mimetic creation, wrought by a divine artist who, at one point in the Timaeus (55c6), is expressly visualized as a painter.
In the Timaeus, the demiurgeâs creation can be seen as mimesis on a cosmological scale. Through reason, the visible universe, âthat which becomesâ, resembles the eternal order âthat which always isâ.7 The cosmos, as it appears in the Timaeus, is composed by geometrical shapes fitting perfectly together, reflecting the eternal, immutable model. This brings us back to Vlastos (2005, 3), who traces kosmos back to âthe active, transitive verb, kosmeÅ: to set in order, to marshal, to arrangeâ. In the present chapter, the team behind the Illustris project is viewed as a modern demiurge: in the creation of the simulation, massive amounts of theory and data from observations are used, and âset in orderâ through the code, giving each particle its place in the virtual universe. The visualisations based on the simulation furthermore transform the âmessyâ big data output to a harmonious re-creation of the cosmos.
As the philosopher of science Laura Perini (2010, 148â151) writes, it is important to distinguish between scientific models and visual representations in science (see also Anna Maerkerâs contribution in this volume). Visualisations from Illustris are visual representations, based on a simulation which is produced from AREPO (the simulation code used to construct Illustris), using both theoretical physics and astrophysics as input. The simulation is based on the Standard Model of Cosmology, but should not be conflated with the model. While mimesis has previously been applied to studies of representations in science, for instance in the case of scientific models, this chapter focuses on the visual aspect of the Illustris Project.8 Although research on images in science has developed in recent decades, astrophysics is a discipline which has so far received little attention.9 With this chapter we contribute an account of visualisations of both dark and normal matter. Key to this last category of visualisations is the distinction between real and âmockâ observations.
1 Mimesis and âMock Observationsâ
A researcher wishing to create a visualisation based on the data output from Illustris will typically start out by searching for the relevant synthetic data from Illustris, such as dark matter in a certain section of the simulation cube during a particular point in the time evolution of the simulation. The researcher then writes a code in Python, controlling the angle from which the pixels representing the particles are seen, as well as the colours used to represent certain phenomena.10 Once the visualisation has been produced, they can zoom in and out of the image, to get an overview of particular pixels in the visualisation representing the data output. âMock observationsâ are visualisations based on the synthetic data output from a simulation, but produced with the purpose of resembling the real universe as it appears through observations. By creating images from Illustris in a fashion similar to the construction of Hubble Space Telescope (HST) observations, astrophysicists are able to compare the virtual universe of Illustris to observations.11 On the website of the Illustris Collaboration (2015c), one of the mock observations is presented as a recreation of âone of the most iconic images in astronomy, the Hubble Space Telescope âUltra Deep Fieldâ (the image below is split in half, to the left and right â one half is real, and one is simulated, can you tell which?)â (fig. 18.3). In order to find the separation between the two images, one has to look closely â the left-hand side of the image is the âreal observationâ from the HST, while the mock observation from Illustris can be found on the right-hand side. It thus becomes apparent how the Illustris visualisations are carefully crafted to resemble images over whose representational power there is already a consensus.
Original caption used to describe the visualisation on the website of the Illustris Collaboration (https://www.illustris-project.org/media/, see Illustris Collaboration, 2015b): âHubble eXtreme Deep Field observations (2.8 arcmin on a side) in B, Z, H bands convolved with Gaussian point-spread functions of sigma = 0.04, 0.08, and 0.16 arcsec, respectively. Divided down the middle: real observation (left side) and mock observation from Illustris (right side)â
Image credit, left: NASA, ESA, G. Illingworth, D. Magee, and P. Oesch ( University of California, Santa Cruz), R. Bouwens (Leiden University), the HUDF09 Team. Public domain. Image credit, right: the Ill ustris Collaboration, courtesy of Mark Vogelsberger
A major result from Illustris was its realistic reproduction of different types of galaxies. Therefore, an important part of research done on it entails comparing âsynthetic imagesâ of galaxies in Illustris with images of galaxies as they appear from observations.12 Many astrophysicists performing research based on Illustris choose to investigate a single galaxy. When the simulation reaches the present time, however, it shows 41,416 galaxies.13 To help astrophysicists navigate in this massive dataset, a catalogue of galaxies, called the âIllustris Galaxy Observatoryâ, is available from the Illustris website (Illustris Collaboration 2015a). By means of a search tool, researchers can adjust several parameters, such as âBlack Hole Mass Limitsâ or âGas Mass Limitsâ, enabling them to find the type of galaxy they are interested in amid the âmessinessâ of big data (the sample of mock observations in fig. 18.4, for instance, shows only disk galaxies).
Original caption used to describe the visualisation on the website of the Illustris Collaboration (https://www.illustris-project.org/media/, see Illustris Collaboration, 2015b): âSample of 42 blue, disk galaxies, showing the stellar light distribution (SDSS g,r,i band composites)â
Image credit: Illustris Collaboration, courtesy of Mark Vogelsberger
In the Timaeus, we also find a frequent alternation in scales between macrocosm and microcosm, aptly described by philosopher Thomas Kjeller Johansen (2004, 6), whose work we will discuss several times throughout this chapter, because of his engagement with the intersection between mimesis, ekphrasis, and natural philosophy in Platoâs natural philosophy:
Whilst devoid of neither argument nor conceptual analysis, the work equally persuades by painting a picture in words of our world as predominantly good and beautiful [â¦] As a picture, the work draws us in by its detail and completeness, âfrom the creation of the cosmos down to the nature of manâ (27a6). The Timaeus-Critias can in part, then, be viewed as a philosophical ekphrasis, or depiction in words, of the whole cosmos.
One could add that it is not only âby painting a picture in wordsâ, but also by painting one in numbers, that Plato expresses the beauty and harmony of the cosmos. In the cosmogony of the Timaeus, the creator of the cosmos is described as moulding existence by forming a harmony out of parts. In Tim. (35c2â36b5), Timaeus combines words and numbers in his description of the role of proportions in the creation of the universe.
These connections produced intervals of 3/2, 4/3, and 9/8 within the previous intervals. He then proceeded to fill all the 4/3 intervals with the 9/8 interval, leaving a small portion over every time. The terms of this interval of the portion left over made a numerical ratio of 256/243.14
Throughout the Timaeus, the reader travels between different scales and follows the ways in which each component fits into the whole. The geometrical aspect of the image of the cosmos âpaintedâ by Plato in the Timaeus is what the âfull-blooded mathematical [Platonist]â Johannes Kepler (1571â1630) would build upon in his modified version of the Copernican system in Mysterium Cosmographicum (1596) (fig. 18.5).15 The Platonic Solids, described in Timaeus as the construction elements of the universe, are five geometrical figures, where the tetrahedron represents fire, the octahedron air, the cube earth, the icosahedron water, and the dodecahedron is the shape of the universe itself. When nested inside each other, the distances between these convex regular polyhedra resemble the distances between the orbits of the planets in the solar system (Gaukroger 2006, 176â178).
Johannes Keplerâs illustration of the orbits of the planets in the Solar System, in Mysterium Cosmographicum (Tübingen 1596)
Public Domain Mark
In Platoâs vertical world of thought, mathematics ranks very high (Shapiro 2005, 3). It is also central in relation to astronomy. In âone of the most disputed passages of Greek literature, Plato in the Republicâ tells us that we should âstudy astronomy by means of problems, as we do geometry, and leave the things in the sky aloneâ.16 In one school of thought, the passage is regarded as favouring a âpurely speculative study of bodies in motion having no relation to the celestial bodies that we seeâ, while others hold that âwhat Plato meant was that astronomers must get to know the real motions of the heavenly bodies as opposed to their apparent motions as seen by us on earthâ (Bulmer-Thomas, 1984, 107). How can this vertical line of thinking in relation to the world of appearances, contrasted with the eternal forms modelling the construction of the universe, inform our present understanding of Illustris as a simulation meant to build a bridge between theory and observation? The purpose of mock observations is to appear as close as possible to real observations. Recalling mimesis as it appears in Republic X, mock observations might be seen as imitations of observations of visible phenomena. In line with this interpretation, one could speak of mock observations as an inversion of the vertical thinking of Plato, using mathematics and reason in order to reach downwards, to replicate the visible phenomena best.17 If we look at the ontology of the virtual universe in relation to the visualisations, however, the simulation is bound together by encoded mathematics and data: a harmonic unification of numbers, resembling the âmimetic model of the cosmosâ found in the Timaeus.18 In the Timaeus, if we follow Johansen, we see a painting of the universe in words, attracting the reader with the harmonic interplay between the whole and its parts. The part of this painting describing the Platonic Solids is not a representation in the sense of a mirror image of the visible world â rather, it is the cosmos seen from a perspective above sensory perception â a universe of theory, painted with words and numbers. The data visualisations based on Illustris, too, show us a universe recreated through the use of theory. Mortals, Timaeus tells us, are not able to completely grasp the otherworldly order. All we can hope to do is to provide a likely myth or story (eikÅs muthos) or a likely account (eikÅs logos).19 The account of the universe could be seen as yet another level of mimesis present in the Timaeus. Following Johansen, in Critias, Critias describes his own account, as well as the explanation given by Timaeus, as:
âimitationsâ (mimÄsis, apeikasia 107b5), perhaps echoing Timaeusâ wish that his account be received as a mere eikÅs logos or eikÅs muthos of an eikÅn of an intelligible paradigm (Tim. 29d). Both logoi, then, are presented to us as imitations of a sort.20
The reader, Johansen continues (2004, 31), is encouraged to take into consideration the famed passage in the Republic describing mimesis (595aâ608b), in relation to the status of the account in the Timaeus-Critias, through the reference to painting found in Critias:
It is inevitable, I suppose, that everything we have all said is a kind of representation and attempted likeness. Let us consider the graphic art of the painter that has as its object the bodies of both gods and men and the relative ease and difficulty involved in the painterâs convincing his viewers that he has adequately represented the objects of his art.21
However, whereas in Republic X the artistâs painting is negatively viewed, on account of the superficial reproduction of the appearance of the phenomena, akin to that of a mirrorâs reflection, the account of Timaeus is another story altogether. Timaeus is described as being very knowledgeable in astronomy, and the explanation of the cosmos is based on his expertise. On this point, Johansen (2004, 35) argues, mimesis as it appears in the Timaeus-Critias differs from mimesis in the notorious passage in Republic X. Furthermore, Timaeus acknowledges that he cannot give certain answers to the origin and hidden nature, of the cosmos â by emphasising the status of his account as âlikelyâ, he avoids the deceptive character of mimesis, unlike the seducing mirror-like painting described in Republic X. Rather than seeing mock observations as imitations of mirror images, recalling Republic X, one could also argue that Illustris works within the Platonic world of thought: Illustris visualisations are produced by specialists in the field of astrophysics, making the simulation one of the most âlikely accountsâ of the cosmos seen today. The likelihood of the account is emphasised through the publication of images such as the Hubble eXtreme Deep Field mock observations from Illustris.
2 Touching Dark Matter
The Illustris Project has gained international attention in major news media, with eye-catching headlines such as: âUniverse evolution recreated in labâ (7 May 2014, BBC news), âUniverse recreated in massive computer simulationâ (7 May 2014, The guardian), âStalking the shadow universeâ (16 July 2014, The New York times), and âHow our universe grew upâ (8 May 2014, CNN).22 Much like large-scale projects such as the HST, communicating the success of Illustris would have been important to its creators as they prepared for new simulations using AREPO. Results from two projects building upon Illustris have already been released: a series of simulations in a project called IllustrisTNG, which improved upon the original version, and the Auriga Project, which gave a detailed view of the Milky Way.23 Several studies of recent astronomical imaging have found that publicity and funding are significant factors in pushing researchers to produce observations intended for communication to the general public.24 False-colour observations, sometimes referred to as âpretty picturesâ, have been a topic of contention amongst the art historians who have studied them. While James Elkins introduces them as âhopped-up versions of legitimate photographs, with the colours intensified or falsifiedâ, Elizabeth A. Kessler argues that false-colour images from the HST are aesthetically valuable, as they are processed in such a way as to evoke the feeling of the sublime.25 Our aim here is not to contribute to this highly interesting discussion of the aesthetic value of false-colour observations. Rather, we want to shift the focus slightly by noting, firstly, that we see a similar development within computational theoretical physics, where it has become common practice to produce colourful, dynamic, and sometimes interactive visualisations from cosmological simulations. Secondly, we note that although funding and publicity likely played a role in pushing researchers to produce the compelling âsynthetic imagesâ, Illustris is one example of how the demarcation between âpretty picturesâ and images used in communication to specialised audiences within astronomy or astrophysics is not clear-cut. Although the visual representations most commonly used in peer-reviewed literature on Illustris are plots (understood here as graphical visual representations of data), synthetic images are also used in articles within journals such as Nature and Monthly notices of the Royal Astronomical Society. Moreover, the ways in which the visualisations are used are similar across communication to intended audiences of varying degrees of specialisation. Synthetic images work to persuade the viewer of the accuracy of the simulation as aids for gaining an overview of Illustris, or for navigating within it. The comparison between the HST observation and the synthetic image in fig. 18.3, for example, appeared in a press release as well as in Nature, in both cases functioning as a visual argument for the quality of the simulation.26 Another example is navigation within the virtual universe. We saw how Illustris Galaxy Observatory uses synthetic images to help researchers find galaxies relevant to their research. Similarly, both in videos used in communications with the general public, and in oral presentations to specialists, information about scale and time provides a way for viewers to orient themselves both spatially and temporally within Illustris. Studies of diagrams in theoretical physics have shown the important role diagrammatic âpaper toolsâ have played to help students and researchers approach challenging topics.27 Although Illustris visualisations do appear in print form â in 2018, for instance, one of them came to decorate an official stamp in Germany (fig. 18.6) â they are more commonly encountered on screens of sizes ranging from smartphones to planetariums. Where Penrose diagrams helped âphysicists struggling through the notorious conceptual and mathematical subtleties of GRâ, virtual tools in the form of synthetic images are used to help viewers grasp a virtual universe on the scale of megaparsecs (Wright 2013, 134).
German stamp featuring a visualisation based on the Illustris simulation
Courtesy of Andrea Voss-Acker, Wuppertal (design) and © the Illustris Collaboration (image), courtesy of Peter Saueressig
Quite unlike synthetic images, the visual style of the plots used in peer- reviewed literature on Illustris are perhaps best characterised by what art historian Kemp has called non-style â the âdraining of obvious ornamentation, stylishness, and pictorial seductionsâ.28 In various ways, synthetic images are instead crafted to look naturalistic, depending on whether they represent dark or normal matter. In the case of mock observations of galaxies, we saw how their close resemblance to observation was emphasised. Visualisations of invisible phenomena instead appear naturalistic because they are shown in perspective, whereby some strands within the cosmic web appear closer than others. Visual representations using a naturalistic style do not necessarily portray objects as they are observed, or even phenomena which exist.29 With visualisations of dark matter, the viewer sees dark matter illuminated, as though crawling up from Platoâs cave. Indeed, the Latin illustris can be translated to âbrightâ, âbrilliantâ, âshiningâ or âpervaded with lightâ.30 Although dark matter cannot be observed, in synthetic images the elusive matter is not only luminous but, as we shall see, becomes yet more tangible and material through rotation and close-up shots in videos.
Some videos from Illustris showcase the simulation in its full size, where the viewer sees the simulation cube from an outside viewpoint; while in others the viewer appears to be placed inside the simulation. In âA virtual universeâ, a video appearing in a Nature news feature on Illustris, moving visualisations from both within and outside the simulation are used.31 In the following, we study this feature as an example of how the visualisations are contextualised and used to communicate Illustris to non-specialists, as well as the ways in which they allow for audiences to interact with Illustris.
âA virtual universeâ begins by zooming in on an animation of a galaxy, whereupon we see an explosion, while in voiceover, we are told about the Big Bang. This is followed by an explanation of Illustris: âTo test our theories scientists have built a computer model of the universe. A simulation so complex that calculating it on a single desktop would take 2,000 yearsâ. We now see the simulation as a whole in a cube, spinning before us, before shifting to a different video from Illustris. Here, we start out by viewing dark matter, zooming in (see Figure 18.7). The field shifts, showing stellar light from a galaxy instead. The video then zooms out, while the scale in parsec is visible on the left side. âThe modelâ, the voiceover narrates, âdoesnât just tackle the universeâs huge range of scales. It also richly describes the forces at work. Much better than previous attemptsâ. The next video shows gas temperature in a cube on the left side, and dark matter on the right. Another cut is made to an Illustris visualisation where we see the cosmic web spinning before us, in what could be described as a close-up, portraying the dark matter as if the viewer could stretch out their hand and touch the large-scale structure of the universe. The field shifts from dark matter to normal matter, where the distribution of colour (from blue through green to red, and finally white) represents temperatures from low to high (fig. 18.8).32
Screenshot from the original video on the Illustris Projectâs website [Illustris Collaboration, 2015b]. Original caption: âContinuous zoom-in from the scale of the entire simulation volume (100 Mpc) to the scale of an individual spiral galaxy (10 kpc), highlighting the diversity of structure across spatial scale, the large dynamic range of the simulation (106 per dimension), and the relationship between dark matter, gas, and starsâ [Illustris Collaboration, 2015b]
Credit: Illustris Collaboration, courtesy of Mark Vogelsberger
Screenshot from the original video on the Illustris Projectâs website. Original caption: âTime evolution of a 10Mpc (comoving) region within Illustris from the start of the simulation to z = 0. The movie transitions between the dark matter density field, gas temperature (blue: cold, green: warm: white: hot), and gas metallicityâ [Illustris Collaboration, 2015b]
Credit: Illustris Collaboration, courtesy of Mark Vogelsberger
Supermassive black holes dominate the largest galaxies, and cause gas to be blown out âin these violent, white burstsâ. Again we see a shift, as the video now shows the chemical composition of the region within the simulation, and soon we see dark matter again. At this point, we are close to the universe in its present state. â14 billion years of the universeâs evolution has pulled a tightly woven dark matter web into a looser network of giant galaxy clustersâ, the voiceover tells us, while the visualisation of dark matter gradually shifts to show galaxy clusters.33 Again, a video which zooms in and out of a galaxy is shown, followed by an animation of an explosion. Finally, we see an animation depicting galaxies, moving close to an individual galaxy, before the animation shows a space telescope in the foreground, and the Earth in the background.
In these videos of the virtual universe, we see current theories and data âin actionâ, with explosions, zoom effects, and rotation in 360 degrees. Commenting on the mapping of dark matter, David Turnbull has argued that dark matter and energy represents a âwhole new ontology and epistemologyâ, with the new understanding of mass and existence bringing with them new forms of cosmological research. Turnbull has described this as a âhyperbolicâ line of argumentation used in order to make what is invisible visible. We suggest a more subtle view of the ânew ontologyâ introduced with dark matter visualisations, one which considers their contextualisation.34 Although dark matter can be â[brought â¦] into existenceâ (Turnbull 2017, 208) through maps or visualisations from simulations, these often appear in the context of verbal descriptions stressing the lack of certainty around what, as physicist Max Tegmark (2014, 70) puts it, âis really little more than a name for our ignoranceâ. In the case of Illustris, dark matter is illuminated and made to appear yet more material using a range of visual techniques, but is at the same time framed as artificial and virtual. This framing of the visualisations as a way of representing the ontology of a virtual universe is characteristic of communications about Illustris targeting the general public. This was likely encouraged by members of the Illustris Collaboration, as the press releases and interviews take the same approach. Interestingly, we see the same in the case of visualisations of normal matter in peer-reviewed papers, where they are referred to as mock observations, constructed from synthetic data, or alternatively as âmock UDFâ (i.e. âUltra Deep Fieldâ), âmock imagesâ, âmock data productsâ, and âsynthetic imagesâ.35 While the accuracy of the simulation is emphasised, so is its artificiality, which, when combined, serves to showcase the abilities of the creators of the virtual universe. Although science has travelled far from the view of the universe seen in the Timaeus, Illustris seems to be driven by a desire resembling that expressed in the Timaeus: to reveal the laws governing the behaviour of the universe, in order to recreate this harmony.
3 âMove over, Matrixâ
At the thought of divine power, the philosopher, politician, and writer Edmund Burke (1729â1797) tells us, âinvested upon every side with omnipresence, we shrink into the minuteness of our own nature, and are, in a manner, annihilated before himâ (Burke 2015, 56). In the Illustris visualisations, the cosmos is visualised on a scale which surpasses that of the Earth, since planets are too small to appear in the simulation. While we cannot speak of anything resembling a geocentric worldview in the visualisations, we could describe the view of the cosmos, given to us through Illustris, as anthropocentric, since creators of Illustris and beholders of its visualisations can gain a sense of control of the virtual universe. Kemp (2006, 35â36) makes a similar observation, although in connection with a much earlier illustration of the universe â Johannes Keplerâs reinterpretation of the Platonic Solids in the Timaeus, in his illustration of the Solar System (fig. 18.5):
Kepler has depicted his scheme for the construction of his own cosmological model [â¦] as if he were acting as a microcosmic emulator of God [â¦] The model, whether Godâs or made by human agency, is by implication something which can potentially be viewed and envisaged perspectivally from any point within or outside the system.
While this Godâs-eye view of the cosmos offers a myriad possibilities in relation to perspective, it is placed on a plinth. From this foundation we can clearly determine the angle from which we view the model. On its plinth, the âgreat folding plateâ is exhibited as a model, resembling, indeed, an armillary sphere, as if placed in the study of a scholar. The model, then, seems to illustrate the possibility of a human omnipresent gaze, enabled by natural philosophy. The same could be said about the Illustris visualisations, which are exhibited as man-made constructions, as images based on âsynthetic dataâ. It is in taking up a Godâs-eye view that Earthâs place in the cosmos shrinks into the minuteness. Through the Illustris visualisations, the beholder is shown a rotating three-dimensional view based on the Standard Model of Cosmology, at multiple scales. The virtual universe is shaped by the modern demiurge of the Illustris Collaboration. The viewer takes on the same position as that of the reader of the Timaeus, gaining access to a âpaintedâ picture of the cosmos, attracting us with the harmonic interplay between whole and parts in the account of the formation of the cosmos. If we regard the Illustris simulation as an anthropocentric view of the cosmos, it is so in the sense that the virtual universe is constructed, and under the control, of the team behind Illustris. Yet as the output of AREPO is not predictable, Illustris escapes from that control. The sense of control a viewer can experience comes from the virtual medium in which they encounter the visualisations: in the videos of the development of the cosmos from shortly after the Big Bang to the present, the viewer can fast-forward, stop, or rewind the visualisation of the cosmological evolution. On the back of discussions following Thomas Nagelâs concept of the âview from nowhereâ, the historian of science Charlotte Bigg has analysed planetariums in the early twentieth century.36 Bigg shows how viewpoints were embodied in pedagogy through certain placements of bodies of students and other spectators visiting, or preparing to visit, planetariums. Here we see a context in which disembodiment is an important part of the point of view, but in the context of a visualisation where dark matter is given body.
Unlike the imagined viewpoint of mock observations (a telescope), in dark matter visualisations, the point of view is not framed as a telescope, since dark matter is not visible through observation. HST observations, some of which the mock observations from Illustris seek to emulate, have given us magnificent views of the cosmos â from false-colour images showing âa universe filled with glowing gases in vivid colours, galaxies swirled together in bands of light and dark, and innumerable starsâ to the famed Hubble Deep Field images.37 âAs a mechanical eyeâ, Kemp (2006, 242) writes, âthe Hubble telescope stands in a long succession of human endeavours to create the ultimate form of sightâ. To Kessler (2012, 19), the HST âstretches humanityâs vision beyond what Galileo ever imaginedâ from its orbit above the Earthâs atmosphere. Elkins (2008, 101) focuses on the intense pursuit to expand the limits of what can be observed. With the Hubble Deep Field North, the âmost wonderful visual act was the attempt to see beyond the faintest galaxies on the plate â to see something in the black regions between the faint bright spots, at the very end of the visible universeâ. Kemp and Kessler characterise the HST in terms of a technological extension of human vision, in part due to its physical extension into outer space, and Elkins in terms of an expansion of the boundaries of the visible universe, through the approach and treatment of observations. While the HST observes in all directions, it is still bound in its orbit around the Earth. With the Illustris visualisations, the viewerâs position is disembodied: any point in the simulation can be taken as a centre, and become a point of view. This sense of control over a virtual replication of our universe brings to mind works of science fiction such as Lana and Lilly Wachowskiâs movie The Matrix (1999).
Indeed, there is a direct connection between The Matrix and the communication of Illustris. On 7 May 2014, a press release introducing Illustris was published on the website of the Harvard-Smithsonian Center for Astrophysics. The Matrix was here used to introduce the reader to the universe of Illustris, appearing in the first sentence: âMove over, Matrix â astronomers have done you one better. They have created a realistic universe using a computer simulation called âIllustrisââ (Aguilar & Pulliam 2014). In The Matrix, the majority of the human population lives in the Matrix, unaware that their bodies are in fact placed in cells, in order to exploit body heat and electrical impulses as sources for energy. The film has been tied to several philosophical discussions and traditions, including Platoâs allegory of the cave. The Matrix is a simulated reality, created by machines that dominate the territory of the Earth. Human rebels fight against agents, computer programmes disguised as humans within the Matrix. Towards the ending of the film, the protagonist, Neo, is shot and killed by Agent Smith, but rises again. When agents shoot at Neo, he stops the bullets. Neo now sees the code of the Matrix and is able to bend the laws of the system through his insight. The viewer, here, sees the universe through Neoâs point of view: the hallway and agents, written in code. There is a similar allure to the visualisations from Illustris: they showcase the current conception of the universe, as it is reproduced from a code written by astrophysicists. âThe matrixâ, as the philosopher and lawyer Paul W. Kahn writes (2013, 122), âis a perfect system of representation, on the one hand, and a completely illusory world, on the other [â¦] There is a logic â the code â that guarantees coherenceâ. The reference to The Matrix in the Illustris press release works to emphasise the closeness of the Illustris simulation compared to observations. Yet Illustris is not naturalised in the way the Matrix is. In the press release from the Harvard-Smithsonian Center for Astrophysics, Shy Genel, a member of the Illustris Collaboration, is quoted saying: âIllustris is like a time machine. We can go forward and backward in time. We can pause the simulation and zoom into a single galaxy or galaxy cluster to see whatâs really going onâ.38 The possibility of travelling through time and space in the virtual universe of Illustris recalls the ending of The Matrix where âwe see [Neo] flying, free of gravity, above other humans, as he dissolves the Matrix and offers them releaseâ (Freeland 2002, 213). Similarly, no source of gravity limits the beholder to a particular point of reference within Illustris. This brings us back to Keplerâs illustration. While being able to see both individual galaxies and the cosmic web in 360 degrees, viewers of an Illustris visualisation are unable to position themselves within the virtual universe. There is no plinth indicating what viewpoint the viewer takes up within the simulation cube. It is not only the Earth which can be said to âshrink into the minutenessâ in the Illustris simulation â by achieving the omnipresent Godâs-eye view, the viewer, too, disappears into the space of Illustris.
4 Conclusion
In this chapter, we have examined âmock observationsâ and visualisations of dark matter from the Illustris simulation in terms of mimesis on a cosmological scale. The Illustris Collaboration can be seen as a modern demiurge, constructing an orderly cosmos out of the chaos of big data, connoting not only the Timaeus, but the root of the word kosmos, found in the Greek verb kosmeÅ. Through data visualisation, the team behind Illustris opens up the black box of big data, enabling the viewer of the Illustris visualisations to access this Godâs-eye view of the virtual universe. Videos from Illustris permit the viewer to see the simulation in rotation, in multiple fields and scales, and to perform a virtual travel through 13.8 billion years of cosmological evolution. The view of the cosmos seen in the Illustris visualisations seems to have left behind anything resembling a geocentric image of the world â planets which could have created associations to the Earth are too small to appear in the simulation, and unlike the HST, which orbits the Earth, in Illustris any point in the simulation cube can become a point of view.
Whereas the HST is described as a technological extension of the human gaze, or a âmechanical eyeâ, the viewpoint in the Illustris visualisations is a disembodied one. While the Illustris visualisations cannot be described as in any sense resembling a geocentric view of the cosmos, they may be understood as anthropocentric, in the sense that the virtual universe is under the control of the Illustris Collaboration, and the viewer of the visualisations can achieve a similar experience as that of the Illustris Collaboration when controlling, for instance, a video showing the development of the universe. Whereas Kepler places his illustration of a cosmological model on a plinth, with the visual dimension of Illustris, the achievement of the omnipresent gaze renders the beholder unable to establish their own viewpoint in relation to the visualisation.
Since the completion of Illustris in 2013, two sets of simulations have been produced using the AREPO code. New cosmological simulations are constantly being developed, growing in both scale and complexity. One could argue that the visual dimension of Illustris is one of the ways in which the public can best view the current state of understanding of the universe within astrophysics. To the layman, the data output of Illustris is entirely inaccessible. Yet the striking visualisations based on the simulation open the black box of these data â through the visual dimension of the project, everyone, including the general public, can view the cosmos, as it is reproduced based on the Standard Model of Cosmology. As noted in the beginning, despite the attention given to cosmological simulations in academic writing as well as public media, many of these simulations have yet to be explored in existing scholarship within the humanities. The release of cosmological simulations gives us a way to study portrayals of our view of the cosmos given by astrophysicists, and, due to the widespread attention they gain, the ways in which a variety of audiences engage with them. As yet more cosmological simulations are released, future studies are needed to analyse their production; the written, verbal, and visual communication of them; their circulation in printed and online media; as well as the responses they solicit, and ultimately how these simulations contribute to how we think about and imagine the universe we inhabit.
Bibliography
Ade, P. A. R., N. Aghanim, M. Arnaud., M. Ashdown, J. Aumont, C. Baccigalupi, A. Banday, R. Barreiro, J. Bartlett, & N. Bartolo, âPlanck 2015 results. XIII. Cosmological parametersâ, Astronomy & astrophysics, (2016), 594, A13.
Aguilar, David A. & Christine Pulliam, âAstronomers create first realistic virtual universeâ, Harvard-Smithsonian Center for Astrophysics (7 May 2014), online: https://www.cfa.harvard.edu/news/astronomers-create-first-realistic-virtual-universe [accessed 30 June 2018].
Aiden, Erez & Jean-Baptiste Michel, Uncharted. Big data as a lens on human culture, New York 2013.
Banks, Catherine M. & John A. Sokolowski, Principles of modeling and simulation: A multidisciplinary approach, Hoboken, N.J. 2009.
Bertone, Gianfranco & Dan Hooper, âHistory of dark matterâ, Reviews of modern physics, 90, No. 4, (2018), 045002.
Bigg, Charlotte, âThe view from here, there and nowhere? Situating the observer in the planetarium and in the solar systemâ, Early popular visual culture, 15, No. 2 (2017), 204â226.
Blackburn, Simon, Oxford dictionary of philosophy, New York 2008.
Borgman, Christine L., Big data, little data, no data: Scholarship in the networked world, Cambridge, Mass. 2015.
Brisson, Luc & F. Walter Meyerstein, Inventing the universe: Platoâs Timaeus, the big bang, and the problem of scientific knowledge, Albany, N.Y. 1995.
Bulmer-Thomas, Ivor, âPlatoâs astronomyâ, The classical quarterly, 34, No. 1 (1984), 107â112.
Bunnin, Nicholas, The Blackwell dictionary of western philosophy, Oxford 2004.
Burke, Edmund, A philosophical enquiry into the sublime and beautiful, New York, 2015 [orig. 1757].
Cambrosio, Alberto, Daniel Jacobi, and Peter Keating, âEhrlichâs âbeautiful picturesâ and the controversial beginnings of immunological imageryâ, Isis, 84, No. 4 (1993), 662â699.
De Swart, Jaco, Gianfranco Bertone, & Jeroen van Dongen, âHow dark matter came to matterâ, Nature astronomy, 1 (2017), 0059.
De Swart, Jaco, âClosing in on the cosmos: Cosmologyâs rebirth and the rise of the dark matter problemâ, 257â284, in: The renaissance of general relativity in context. Einstein studies, vol. 16 (A.S. Blum, R. Lalli, & J. Renn, eds.), Cham 2020.
Durán, Juan Manuel, Computer simulations in science and engineering, Cham 2018.
Elkins, James, The domain of images, Ithaca, N.Y. 1999.
Elkins, James, Six stories from the end of representation. Images in painting, photography, astronomy, microscopy, particle physics, and quantum mechanics, 1980â2000, Stanford, Calif. 2008.
Freeland, Cynthia, âPenetrating Keanu: New holes, but the same old shitâ, in: The Matrix and philosophy: Welcome to the desert of the real (W. Irving, ed.), Peru, Ill. 2002.
Frigg, Roman & Matthew C. Hunter, Beyond mimesis and convention. Representation in art and science, Dordrecht 2010.
Frigg, Roman & Julian Reiss, âThe philosophy of simulation. Hot new issues or same old stew?â, Synthèse, 169, No. 3 (2011), 593â613.
Galison, Peter, Image and logic: A material culture of microphysics, Chicago 1997.
Gaukroger, Stephen, The emergence of a scientific culture: Science and the shaping of modernity, 1210â1685, Oxford 2006.
Ghosh, Pallab, âUniverse evolution recreated in labâ, BBC news, 7 May 2014, online: http://www.bbc.com/news/science-environment-27299017 [accessed 30 June 2018].
Gombrich, Ernst H., Art and illusion: A study in the psychology of pictorial representation, Oxford 1960.
Goodman, Alyssa A., âPrinciples of high-dimensional data visualization in astronomyâ, Astronomische Nachrichten, 333, Nos. 5â6 (2012), 505â514.
Grand, R. J. J., F. A. Gómez, F. Marinacci, R. Pakmor, V. Springel, D. J. R. Campbell, C. S. Frenk, A. Jenkins, & S. D. M. White, âThe Auriga Project. The properties and formation mechanisms of disc galaxies across cosmic timeâ, Monthly notices of the Royal Astronomical Society, 467, No. 1 (2017), 179â207.
Gregory, Andrew, Platoâs philosophy of science, London 2000.
Halliwell, Stephen, The aesthetics of mimesis: Ancient texts and modern problems, Princeton, N.J. 2002.
Halpern, Orit, Beautiful data: A history of vision and reason since 1945, Durham & London 2015.
Hannestad, Steen, âVisualizing the invisible universeâ, 64â86, in: The aesthetics of scientific data representation: More than pretty pictures (L. Philipsen & R. S. Kjærgaard, eds.), New York 2018.
Hentschel, Klaus, Mapping the spectrum: Techniques of visual representation in research and teaching, New York 2002.
Hentschel, Klaus, Visual cultures in science and technology: A comparative history, New York 2014.
Hopwood, Nick, Haeckelâs embryos: Images, evolution, and fraud, Chicago 2015.
Illustris Collaboration, âIllustris galaxy observatoryâ (2015), online: http://www.illustris-project.org/galaxy_obs/ [accessed 30 June 2018].
Illustris Collaboration, âMedia. Images, videos, and other data visualizationsâ (2015), online: http://www.illustris-project.org/media/ [accessed 30 June 2018].
Illustris Collaboration, âPublicâ (2015), online: http://www.illustris-project.org/about/ [accessed 30 June 2018].
Johansen, Thomas Kjeller, Platoâs natural philosophy: A study of the âTimaeus-Critiasâ, Cambridge 2004.
Kahn, Paul W., Finding ourselves at the movies: Philosophy for a new generation, New York 2013.
Kaiser, David, Drawing theories apart: The dispersion of Feynman diagrams in postwar physics, Chicago 2005.
Kemp, Martin, âTaking it on trust. Form and meaning in naturalistic representationâ, Archives of natural history, 17 (1990), 127â1188.
Kemp, Martin, Visualizations: The nature book of art and science, Berkeley & Los Angeles 2000.
Kemp, Martin, Seen/unseen: Art, science, and intuition from Leonardo to the Hubble Telescope, New York 2006.
Kemp, Martin, âStyle and nonâstyle in anatomical illustration. From Renaissance humanism to Henry Grayâ, Journal of anatomy, 216, No. 2 (2010), 192â208.
Kepler, Johannes, Prodromus dissertationum cosmographicarum, continens mysterium cosmographicum, de admirabili proportione orbium coelestium, deque causis coelorum numeri, magnitudinis, motuumque periodicorum genuinis & propriis, demonstratum, per quinque regularia corpora geometrica. Tubingae: excudebat Georgius Gruppenbachius, Tübingen 1596, ETH-Bibliothek Zürich, RAR 1367: 1, online: http://dx.doi.org/10.3931/e-rara-445 [accessed 22 February 2018].
Kessler, Elizabeth A., âPretty sublimeâ, 57â74, in: Beyond the finite: The sublime in art and science (R. Hoffmann and I. Boyd Whyte, eds.), New York 2011.
Kessler, Elizabeth A., Picturing the cosmos: Hubble Space Telescope images and the astronomical sublime, Minneapolis 2012.
Klein, Ursula, Experiments, models, paper tools: Cultures of organic chemistry in the nineteenth century, Stanford, Calif. 2003.
Kusukawa, Sachiko, Picturing the book of nature: Image, text, and argument in sixteenth-century human anatomy and medical botany, Chicago 2012.
Landau, Elizabeth, âHow our universe grew upâ, CNN, 8 May 2014, online: http://edition.cnn.com/2014/05/08/tech/innovation/universe-simulation/index.html [accessed 30 June 2018].
Lynch, Michael & Samuel Y. Edgerton Jr., âAesthetics and digital image processing. Representational craft in contemporary astronomyâ, 184â220, in: Picturing power: Visual depiction and social relations (G. Fyfe, ed.), London 1988.
Lynch, Michael & Samuel Y. Edgerton Jr., âAbstract painting and image processingâ, 103â124, in: The elusive synthesis: Aesthetics and science (A. I. Tauber, ed.), Dordrecht & London, 1997.
Mayer-Schönberger, Viktor & Kenneth Cukier, Big data: A revolution that will transform how we live, work and think, London 2013.
Nagel, Thomas, The view from nowhere, Oxford 1986.
Nelson, D., A. Pillepich, S. Genel, M. Vogelsberger, V. Springel, P. Torrey, V. Rodriguez- Gomez, D. Sijacki, G. F. Snyder, B. Griffen, F. Marinacci, L. Blecha, L. Sales, D. Xu, & L. Hernquist, âThe Illustris simulation: Public data releaseâ, Astronomy and computing, 13 (2015), 12â37.
Nelson, D., A. Pillepich, V. Springel, R. Weinberger, L. Hernquist, R. Pakmor, S. Genel, P. Torrey, M. Vogelsberger, G. Kauffmann, F. Marinacci, & J. Naiman, âFirst results from the IllustrisTNG simulations: the galaxy colour bimodalityâ, Monthly notices of the Royal Astronomical Society, 475, No. 1 (2018), 624â647.
OLD, Oxford Latin dictionary (P. G. W. Glare, ed.), London 1968.
Overbye, Dennis, âStalking the shadow universeâ, The New York times, 16 July 2014, online: https://www.nytimes.com/2014/07/16/science/space/stalking-the-shadow-universe.html[accessed 30 June 2018].
Pauwels, Luc, Visual cultures of science. Rethinking representational practices in knowledge building and science communication, Hanover, N.H. 2006.
Perini, Laura, âScientific representation and the semiotics of picturesâ, 131â154, in: New waves in philosophy of science (P. D. Magnus & J. Busch, eds.), New York 2010.
Pillepich, A., D. Nelson, L. Hernquist, V. Springel, R. Pakmor, P. Torrey, R. Weinberger, S. Genel, J. P. Naiman, F. Marinacci, & M. Vogelsberger, âFirst results from the IllustrisTNG simulations: the stellar mass content of groups and clusters of galaxiesâ, Monthly notices of the Royal Astronomical Society, 475, No. 1 (2018), 648â675.
Plato, Plato. Complete works, Indianapolis 1997.
Plato, Platoâs cosmology. The Timaeus of Plato (F. Macdonald Cornford, ed. & annot.), Indianapolis 1997.
Sample, Ian, âUniverse recreated in massive computer simulationâ, The guardian, 7 May 2014, online: https://www.theguardian.com/science/2014/may/07/universe-recreated-computer-simulation-model-big-bang [accessed 30 June 2018].
Shapin, Steven, The scientific revolution, Chicago 1996.
Shapiro, Stewart, The Oxford handbook of philosophy of mathematics and logic, Oxford 2005.
Snyder, G.F., P. Torrey, J.M. Lotz, S. Genel, C.K. McBride, M. Vogelsberger, A. Pillepich, D. Nelson, L.V. Sales, D. Sijacki, L. Hernquist, & V. Springel, âGalaxy morphology and star formation in the Illustris Simulation at z=0â, Monthly notices of the Royal Astronomical Society 454, No. 2 (2015), 1886â1908.
Springel, Volker, âE pur si muove: Galilean-invariant cosmological hydrodynamical simulations on a moving meshâ, Monthly notices of the Royal Astronomical Society, 401, No. 2 (2010), 791â851.
Springel, V., R. Pakmor, A. Pillepich, R. Weinberger, D. Nelson, L. Hernquist, M. Vogelsberger, S. Genel, P. Torrey, F. Marinacci, & J. Naiman, âFirst results from the IllustrisTNG simulations: matter and galaxy clusteringâ, Monthly notices of the Royal Astronomical Society, 475, No. 1 (2018), 676â698.
Stephenson, Bruce, The music of the heavens: Keplerâs harmonic astronomy, Princeton, N.J. 1994.
Stoddart, Charlotte & Elizabeth Gibney, âA virtual universeâ, in: E. Gibney, âModel universe recreates evolution of the cosmosâ, Nature news, 7 May 2014, online: https://www.nature.com/articles/d41586-019-00415-3 [accessed 30 June 2018].
Tegmark, Max, Our mathematical universe. My quest for the ultimate nature of reality, London 2014.
Torrey, P., G.F. Snyder, M. Vogelsberger, C.C. Hayward, S. Genel, D. Sijacki, V. Springel, L. Hernquist, D. Nelson, M. Kriek, A. Pillepich, L.V. Sales, & C.K. McBride, âSynthetic galaxy images and spectra from the Illustris Simulationâ, Monthly notices of the Royal Astronomical Society, 447, No. 3 (2015), 2753â2771.
Turnbull, David, âMapping dark matter and the Venice Paradoxâ, 203â231, in: Aesthetics of universal knowledge (S. Schaffer, J. Tresch, & P. Gagliardi, eds.), Cham 2017.
Vertesi, Janet, Seeing like a rover: How robots, teams, and images craft knowledge of Mars, Chicago 2015.
Vlastos, Gregory, Platoâs universe, Las Vegas 2005.
Vogelsberger, M., S. Genel, V. Springel, P. Torrey, D. Sijacki, D. Xu, G. Snyder, S. Bird, D. Nelson, & L. Hernquist, âProperties of galaxies reproduced by a hydrodynamic simulationâ, Nature, 509 (2014), 177â182.
Wright, Aaron Sidney, âThe origins of Penrose diagrams in physics, art, and the psychology of perception, 1958â62â, Endeavour (new series), 37, No. 3 (2013), 133â139.
Wright, Aaron Sidney, âThe advantages of bringing infinity to a finite place. Penrose diagrams as objects of intuitionâ, Historical studies in the natural sciences, 44, No. 2 (2014), 99â139.
Endnotes
Hydrodynamics is the study of fluid motion, which in this instance refers to cosmic gas. A simulation is here understood as an âimitation of the operation of a real-world process or system over a period of timeâ, following Banks & Sokolowski 2009, 5. See works by Galison (such as 1997) and sources cited therein for literature on simulations within the history and philosophy of science, Frigg & Reiss 2009 for a critical review of literature in philosophy, and Durán 2018 for a general introduction. While several simulations exist within the Illustris Project, the Illustris simulation is often referred to in the singular, since Illustris-1, the simulation containing the most particles, is usually used for research and to create visualisations. For information on the various simulations, see Nelson et al. 2015.
These are estimates based on the 2015 results from the Planck Satellite (Ade et al. 2016).
See e.g. Vogelsberger et al. 2014 and Springel 2010 on Illustris, and Borgman 2015, 89â90 for an introduction to synthetic data in astrophysics.
See Aiden & Michel 2013, 19; Mayer-Schönberger & Cukier 2013, 32â49.
See Vlastos 2005, 26; Plato & Cornford 1997, 27; Johansen 2004, 95, note 6.
See for instance Blackburn 2008, 235; Bunnin 2004, 434. For discussions of this translation and the etymology of mimesis, see Halliwell 2002, 13â14, 17â22.
Tim. (27d5â6), following Plato 1997. Original emphasis.
For a discussion of the use of the concept of mimesis in relation to different kinds of representations in science, amongst others, see Galison 1997 and Frigg & Hunter 2010, and references cited therein. For an analysis of the Standard Model of Cosmology in light of the account of the cosmos in the Timaeus, see Brisson & Meyerstein 1995.
See Cambrosio et al. 1993, 662 for an introduction to pioneering work within this field. For an overview of research on scientific imaging, see Hentschel 2002, 2014; Pauwels 2006, and further references cited in Hopwood 2015, 309, n. 10. For existing research on contemporary astronomical imaging, see Lynch & Edgerton jr. 1988, 1996; Elkins 1999, 2008; Hannestad 2018; Kessler 2011, 2012; Turnbull 2017; Vertesi 2015, and references cited therein. See Hentschel 2014, 258 for literature on visualisations from the Millennium simulation.
Most visualisations based on Illustris are constructed using Python, a standard programming language in astronomy (Goodman 2012, 7).
See Torrey et al. 2015 for a detailed description of the construction of mock observations.
For an example of this, see Snyder et al. 2015.
Vogelsberger et al. 2014, 177.
Tim. (35c2â36b5), Plato 1997.
Shapin 1996, 59. On Keplerâs so-called polyhedral theory, see Stephenson 1994 and references cited therein.
Bulmer-Thomas 1984, 107; Rep. VII (530b5â6), in Plato 1997.
Regarding astronomy in the Timaeus, Plato allows for an exchange between reason and perception (see, e.g., Tim. 47a1âb2). See Johansen 2004, 160â176 for a detailed analysis of astronomy in the Timaeus, as well as Gregory 2000 and Vlastos 2005.
Halliwell 2002, 321, n. 24.
Tim. (29c8âd3; see also 68câd and 59câd); Tim. (30b, 48d, 53d, 55d, 56a, 57d, 90e).
Johansen 2004, 31. Critias is the unfinished dialogue following the Timaeus, in a projected trilogy where Hermocrates â likely never written â is assumed to be the third dialogue.
Crit. (107b5âc2), translated by Diskin Clay in Plato 1997.
Ghosh 2014; Sample 2014; Overbye 2014; Landau 2014.
For IllustrisTNG, see Pillepich et al. 2018; Springel et al. 2018; Nelson et al. 2018. For the Auriga project, see Grand et al. 2017.
Lynch & Edgerton Jr. 1988; Elkins 1999, 2008; Kessler 2011, 2012; Vertesi 2015.
Elkins 2008, 87; Kessler 2011, 2012.
For the press release, see Aguilar & Pulliam 2014; for the Nature publication, Vogelsberger et al. 2014.
On âpaper toolsâ, see Klein 2003. On paper tools in theoretical physics, see Kaiser 2005; Wright 2013, 2014.
Kemp 2000, 4; id. 2010.
Kemp 1990; Kusukawa 2012.
OLD 1968, 830.
Stoddart & Gibney 2014 links to the video.
Transcriptions from the voiceover in the video âA virtual universeâ; see Stoddart & Gibney 2014.
Stoddart & Gibney 2014.
Turnbull 2017, 207. See also de Swart et al. 2017, de Swart 2020, Bertone & Hooper 2018, and sources cited therein for the history of research on dark matter.
Vogelsberger et al. 2014, 177â178; Torrey et al. 2015, 2753.
Nagel 1986, Bigg 2017: see 204â205 for an introduction to the discussion following Nagelâs work.
Kessler 2012, 57.
Quoted in Aguilar & Pulliam 2014.