Truth, context, perception, and planets

Rather differently from my usual fantasy doodles and occasional opinionated rants, today I've got a bit of an art history essay to share.

I recently watched a very intriguing and thought-provoking lecture by Randall Rosenfeld, archivist for the Royal Astronomical Society of Canada, where he discussed the late 19th-century astronomical illustrations of Étienne Léopold Trouvelot. (You can view a gallery of Trouvelot's lovely renderings here: https://digitalcollections.nypl.org/collections/the-trouvelot-astronomical-drawings-atlas)

At the crux of Rosenfeld's presentation was a question he had no satisfactory answer for: why are some of Trouvelot's illustrations almost photographic in their accuracy (such as the Orion Nebula), while others appear highly foreign and almost caricatures of their subjects' true appearance (such as Jupiter)? Trouvelot was a trained and very skilled artist, using the best telescopes of his time. When his lithographs were published, they were widely praised by professional astronomers who saw nothing wrong with the way he had depicted anything. Other artists before him actually rendered these subjects with more accuracy (you can see a good selection of early Jupiter artwork here), so what was going on?

Rosenfeld ended the lecture with the question remaining open. But as I was absorbing the information he presented, a hypothesis sprung into my mind and I wondered if it might not be along the right track. I emailed Rosenfeld but never heard back from him, so I'm posting the contents of that email here (slightly edited to read less like an email and more like a formal essay), in the hopes that maybe it will help someone along in figuring out the answer to this interesting historical conundrum.

My main observation when I analyze Trouvelot's work is that his renderings of deep-sky objects are very accurate while his solar system objects are much less true-to-life. I believe this disparity has to do with how little was actually known about the physical nature of solar system objects during the time frame in which these illustrations were produced.

By the late 1800s, astronomical spectroscopy was well-established and had been used to examine the chemical makeup of stars, nebulae, and galaxies, and it was known by the 1870s, by virtue of their spectra, that nebulae (both true nebulae and galaxies) were gaseous, dusty objects. On the other hand, next to nothing was understood about the Sun and the planets. Nobody knew what made the Sun shine. Most people were firmly convinced Mars was inhabited by intelligent life and covered in canals. And it wasn't even known if Jupiter and Saturn had solid surfaces, to say nothing of Uranus and Neptune which were simply blue-green dots for all intents and purposes.

We here in the 2020s are extremely spoiled by in-situ space missions and decades of intensive research in planetary geology and solar science; the 1870s did not have such a luxury. Trouvelot's renderings of Jupiter, Saturn, Mars, and sunspots look "wrong" to us now because we actually know what their features are, and what they are supposed to look like at any given scale. And, speaking as an illustrator myself, I can say that it is extremely difficult to accurately render something when you have no idea what it is. 

I suspect that Trouvelot's nebulae fared better because his artist's intuition knew they were clouds of gas and dust, which he likely had much experience observing on Earth and drafting. On the other hand, when he set himself to the task of attempting to record surface features on Mars, he had absolutely no idea what he was looking at. With no context of background knowledge of the geology and atmosphere of Mars, the best he could manage was some abstract shapes, and honestly it was no worse than what other people were doing, making a mess of the Martian globe with networks of canals. 

(I recently came across what I feel is a really compelling hypothesis that Percival Lowell, and probably many other late 19th-century astronomers, when they observed Mars, accidentally turned their telescopes into ophthalmoscopes when they reduced the aperture in an attempt to cut down glare. When they thought they saw canals, they were actually seeing the blood vessels in their own eyes, and Trouvelot may have done the same thing with his Mars observations. Here is the lecture where this is mentioned; the link goes to the timestamp where the speaker begins discussing it. In fact, this may explain that weird prominent dark spot in the middle of Trouvelot's Mars lithograph, not really corresponding to any real feature on Mars--it may have actually been Trouvelot seeing his own optic nerve.) 

As for the Moon, to me Trouvelot's renderings appear "plasticized" because they're lacking the rugged texture of the lunar surface caused by it basically being craters all the way down--but nobody knew that prior to the 1960's. I imagine Trouvelot had to make some judgment calls about the smallest craters he could make out, and with a lack of evidence to the contrary, portrayed the rest of the lunar surface as smooth. To us nowadays that sounds absolutely ludicrous, but right up until the Apollo era, some people argued the Moon was covered in a thick layer of dust. There is so much we simply didn't know, and didn't anticipate, about the other bodies in our solar system, until we actually went there and investigated.

There is the question of why Trouvelot's actual observational sketches are so much more accurate than the published lithographs, but I think this may be because when the lithographs were produced, Trouvelot "cleaned up" the images for public consumption, and inevitably made some errors for reasons discussed above. However, also for reasons discussed above, these errors were not commented on by professionals because nobody knew any better, and what the planets looked like "up close" was anyone's guess. 

I think what Trouvelot did with his rather art-deco Mars and Jupiter was to sharpen the color features on these planets in a well-meaning, but ultimately uninformed attempt to better display their structure, to produce something both aesthetically pleasing for the layperson and instructive to the student. I honestly think his 1882 Jupiter lithograph is a really good rendering of Jupiter by someone who has absolutely no idea what is happening on Jupiter (nobody even knew the Great Red Spot was a storm until the Pioneer missions of the 1970's). 

With Trouvelot's rendering of the Leonids, he was attempting to show not an accurate image of what one might see at any given moment in a meteor shower, but to give the impression of how meteor showers work and what one might expect to see over the course of a night's observing. As these were the days before motion pictures and YouTube, scientific illustrators had to get creative if they wanted to give an informative depiction of a transient phenomenon.

To our modern eyes, which are connected to our brains which are well-stocked with images from Voyager, Cassini, the Hubble Space Telescope, etc., we look at even the smallest, blurriest image of a planet and instantly know that we are "seeing" craters, ice caps, or storms. In reality, what we are seeing are shapes and colors, and our brains are filling in the gap with information that was simply inaccessible to people in the 1800s.

A really good analogue can be found in paleoart. Back when dinosaurs were first being discovered, they were only known from fragmentary evidence. However, that didn't stop people from using the information on hand to reconstruct entire animals, such as the dinosaur sculptures created for the Crystal Palace in London in the 1850s. The sculptures were executed by a very talented wildlife artist, overseen by the professional paleontologist who coined the clade Dinosauria, became explosively popular with the public, and are completely inaccurate in every conceivable way. Nowadays, they're viewed as total jokes, but back in their day, they were regarded as being at the cutting edge of scientific research, because people simply didn't know any better.

As the phrase goes, you don't know what you don't know. I'm sure there's plenty we're getting wrong about science right now that future generations will think is hilarious or weird. I believe that was the case with Trouvelot's renderings; he was doing the best he could with what little information he had, and unfortunately time and scientific progress have simply proved many of his artistic judgments inaccurate.