Friday 31 March 2017

Scientist-palaeoartist collaborations – what palaeontologists can, and probably should, critique when reviewing palaeoart

Last month I posted a complaint about poor scientist-led palaeoart - those artworks of extinct animals produced under direct control of scientists to promote research, without any interference from TV companies or book publishers, and yet still end up being objectively flawed at a scientific level. I focused a lot of my criticisms at scientists themselves, as they have final authority over factual aspects of palaeoartworks. That doesn't necessarily clear artists from all blame, but it's naive to think that every artist tackling palaeoart has specific training in palaeontological matters, unrestricted access to technical literature, or the anatomical knowledge required to restore fossil animals without instruction. A scientist's main role in a palaeoart collaboration is bringing rigour and information to the table and, when science-led pieces are objectively poor, we have to wonder what happened to those guiding hands.

Our instinct might be to assume that lack of scientific rigour reflects flippancy towards palaeoart and its impact, and I think that is true - in at least some cases. But in a comment posted after my article, Matt Bonnan proposed that scientifically poor artwork might reflect scientists struggling with their role in the palaeoart process - that is, not really knowing how to instruct an artist, or where the line between scientific and artistic considerations lies. I can believe this is true, too. Lots of people - including many scientists and artists - view science and art as incompatible concepts, and are unsure how to approach projects blending the two. For some folks the idea of contributing to an art project is pretty terrifying, perhaps because they're afraid of being seen as naive, or of their contribution letting a project down.

Whatever the reason, I want to follow my previous criticism with something more constructive: some pointers for how scientists might approach their role in producing palaeoartworks. The take-home message is that scientists concerned about getting paint on their fingers shouldn't worry about the artistic aspects of paleaoartworks. The primary role of a scientist is not to understand colour hues, choice of media, composition and so on, but to comment on objective, factual components of the work. Is the restored species the right size in relation to other species and the environment? Is its head the right shape? Has it been restored with the right soft-tissue anatomy? Answering these and other questions does not require artistic training, or any special training at all for that matter, just application of knowledge that most palaeontologists will already have.

This discussion mainly considers scientists involved in producing a novel palaeoartwork, but will also apply to those reviewing artwork before publication or exhibition. This includes peer review of papers with palaeoartworks, and I encourage editors to make sure these aspects are checked alongside other parts of a paper. After all, if an artwork is being presented as scientifically-credible enough to be included in a peer-reviewed publication, it should be held to the same standards as the rest of the paper. I don't want to beat a dead horse by bringing this up again, but we should recall that palaeoart components can remain in use long after the context of its genesis has been consigned to history, and artworks associated with papers can be especially prone to long-term use. We should take all opportunities to steer depictions of the past in the most credible directions, hopefully influencing subsequent generations of artists in the best way possible.

So, what should scientists look to critique in artworks, and how might they go about assessing credibility? A perquisite of guiding palaeoart processes is having a concept of what the subject species looked like. This might seem like a patronising comment, with experts replying 'of course I know what [subject species] looks like!', but really think about it - do you really know what the proportions of your subject are, and what they look like when reconstructed against one another? As in, to the point where you could render a reasonable stick-figure version of it? Could you describe what it looks like outside of plain lateral view, and are these interpretations based on modern technical data, not treatments of the same subject by previous artists? Palaeoart is highly varied in scientific credibility and even those works produced by masters of palaeoart may have dated or contain errors. Ergo, palaeoart consultants should form their basic concepts of appearance from images of fossils, tables of measurements and other primary resources, not previous artistic interpretations. This need for caution applies to skeletal reconstructions too, as these become dated and require modernisation as much as any other reconstructions of prehistoric life. So before dusting off that copy of Romer's Vertebrate Palaeontology for another round of consultancy work, or digging out a skeletal reconstruction from a century ago, consider how kind the last few decades of research have been to those familiar images. Anyone needing an example of how a well-known, seemingly 'safe' skeletal can become dated should check out Scott Hartman's new Dimetrodon skeletal. The animal we all 'know' as Dimetrodon is really Romer's 1927 skeletal - 90 years on, it's looking pretty different.

Role 1 of a palaeoart consultant: know what the basic anatomy of a subject looks like when pinned together. A great poster child for this requirement are pterosaurs, familar animals but with very unfamiliar proportions. It's continually clear from scientist-led pterosaur palaeoartworks that their proportions remain poorly appreciated, even to specialists, a fact made especially obvious by the continued depiction of large torsos in many species. Spend an afternoon piecing together pterosaur fossils, or even just measurements of their bones, and their tiny bodies - shown here via Quetzalcoatlus sp. - become inescapable.
Once good contemporary and credible references have been amassed to account for the technical side of a project, they can be used as core reference material for all parties on the project, giving a common goal to work towards. A useful shopping list for 'core references' might be a skeletal reconstruction of the subject species (either drawings or a good museum mount; a closely related species might do if the specific skeletal is unavailable, and especially if the subject is poorly known or only subtly different from available reference material), lists of bone measurements and ratios, and literature providing a detailed insight into the anatomy of the subject. Good photos of referred specimens, from as many angles as possible, are always helpful too.

Is palaeoart a reliable source of information about the appearance of a fossil animal? Sometimes yes, but oftentimes, no. Talk slide from my 2014 TetZooCon presentation on the cultural evolution of azhdarchids pterosaurs, showing some of the earlier, zanier attempts to restore these animals. There's so much incredulous anatomy here that artists and scientists should steer well clear of these as reference material and go back to primary sources - fossils, descriptions, measurements - to form the foundation of their artwork. (Psst - TetZooCon is happening again soon, details here)
With these references in hand, regular checks on developing artwork can begin. A rule of thumb in palaeoart is that aspects of an artwork should be justifiable one way or another ("these proportions are because of that, this pose seems OK because of this, this speculation reflects this..."). If they aren't, or the defence for that aspect is suspect, the artwork should be modified until a superior interpretation is presented. We can go a long way to bringing palaeoart credibility up to speed by appraising the following, fact-based elements:
  • Anything to do with basic measurements, including the size of the subject relative to its environment and other species, or the proportions of its body. Obtaining metrics from 2D art can be difficult if a subject is obliquely posed or foreshortened, but their rough proportions can be estimated based on their relationship to other body parts. If in doubt, it’s better to get the artist to check their work than to ignore it. Pay particular attention to the proportions of the head to the rest of the body, the size of the torso, and the ratios of the limb bones, as these are prone to errors.
  • Whether the skeleton of the subject fits within the restored soft-tissue volumes. Especially notice the shape of the head and teeth, the cross-section and length of the torso, and the bulk of the appendages, as these are often problem areas. Also make sure the position of the shoulders is correct – it is often more challenging to reconstruct the pectoral region than the pelvic, so the forelimb attachment region can end up in strange places.
  • Whether the chosen pose breaches predictions of joint articulation. Over-stretched limbs, as well as exaggerated neck and tail poses, are key to look at here.
  • Whether appropriate fossil soft-tissues have been factored into the painting. This includes tissue types (e.g. correct integument) and aspects of tissue bulk. Where tissue types are unknown, check that the predicted substitute is based on sensible use of phylogenetic bracketing and comparative anatomy.
  • Finally, note whether the species depicted in the artwork were actually contemporaneous, and that the restored environments and climates are appropriate. 
If, via aid of flux capacitored DeLorean, I was consulting for my own azhdarchid art from 2008, I could make lots of suggestions for improvement on purely scientific grounds. The comments here - concerning proportions, limb bone orientations, bone shapes and so on - could be made from any scientist familiar with recent work and interpretations of pterosaur anatomy, and do not require any forays into the artistic side of palaeoart.
Note that none of these aspects stray into areas of artistry, except - sometimes - a need to interpret 3D shapes in 2D art. Moreover, virtually all of these elements relate to commonly studied aspects of fossil forms. All we're doing is taking the same bone shapes and proportions that inform taxonomic or systematic studies, or the ratios and metrics that underlie functional analyses, and applying them to a different project. We're using information that most scientists already know or have immediately to hand, just set to a different tune.

Because of this, palaeoart consultancy is not as arduous a task as it first appears, nor a total time sink. I'm not going to pretend that good palaeoart consultancy is a job you can do in seconds but, once you have basic references established, most comments simply pertain to nudging the reconstruction in the right direction. As with many academic projects, advising on palaeoart requires the most time investment up front, and then relatively little after. Needless to say, the more prepared you are at the start, the less time investment is needed down the line.

And these points - basic as they might seem - will see just about any palaeontologist able to guide and shape palaeoart production. It should be stressed how continued checking along these lines can make an amazing difference to a palaeoartwork, and thus its success at capturing a hypothesis and future legacy. Correcting a scientific goof not only makes a picture more credible, but it often marks a division between a picture being artistically lacking and coming together. There's a reason artists of living creatures (including humans) are so obsessed with the anatomy of their subjects, and that's because it's essential to producing good artwork. Palaeoart is no different, so don't be shy: help your artist get the information and understanding they need to make your science look great.

If you enjoyed this post, you might enjoy my exclusive Patreon content

This blog is sponsored via Patreon, the site where you can donate a few dollars to your favourite online content creators to help them make a living. If you like what I do here, please consider becoming a patron for as little as $1 a month. In return, you'll get regular updates on projects I'm working on. Currently, this includes two palaeoart-heavy books, and my Patreon pages are filling with advance previews of their content, discussions of the animals concerned and that sort of thing. Plus, you get free stuff - prints, high quality images for printing, books, competitions - for sponsoring my work. You can access all this for $1 a month.


  1. This comment has been removed by the author.

  2. Nice post Mark. Recognizing errant proportions isn't merely a matter of paying sufficient attention, it's a visualization ability that many people aren't good at. So I'd like to second your call for using numbers - I've seen too many cases of scientific consultants supplying advice that "item X needs to be a bit bigger" based on a mistaken impression. And with that sort of qualitative feedback it's not really actionable for the artist even if the advisor is 100% correct.

    P.S. I need to edit before I hit post.

  3. > dimensions as perceived within perspective. 3D is great here because you can dodge the perspective issue with scale-specific viewpoints, but nowadays mastery of perspective is seldom enough among artists... it's understandable if that's a grey point in scientist-artist feedback

    > visual communication
    It's great hearing you talk on this, because your POV is very 'scientist'. Some more considerations from a 'manager / artist' based on experiences, not just with scientists but all involved...
    - visual communication. You post great examples in the images here, but to drive the point home... communicate visually. An email with descriptions ... nah. Open the image, draw on it. Can't draw? Of course you can. Get a tablet... a bamboo or something that let's you scribble. And then scribble directly on the images.

    > modular communication
    Be aware of what it is that is being discussed at what moment. In the earlier steps of interpreting a reconstruction, use drawings specifically suited to the purpose. Later, when discussing the animal in perspective and dynamically posed, these discussions can be referred to independently of perspective, isolating questions of ie. plausible posing of joints.
    Likewise, be aware of the key message needed for the project. Matt's post about communicating the sheer hugeness of Argentinosaurus is a great example of this.
    All of this is concentrated in an iteration... meaning one feedback loop of "artist does a, scientist gives feedback, resulting in artist doing b > repeat process". Progress is not measured in time or effort spent. It is measured in iterations. If you are not reviewing based on a sequentially prioritized list of issues, you are not capitalizing on your most valuable unit... the efficiency of iterations.

    > trust
    The less each participant understands their partner's POV and intentions of the other side, the more trust is an issue. Is the scientist a communicational recluse distrustful of non-objective illustration? Is the illustration his/her wish or forced upon them by the PR department? Is the illustrator interested in palaeo? Is he/she a concept artist, medical illustrator? Each has a wildly differing package of expectations on workflow and focus.
    Generally speaking, any palaeo-interested artist is diving into the material and giving more bang for buck than could be reasonably expected. Still, opportunities for irritation abound: why can issue 'a' be changed trivially, but you tell me that issue 'b' requires a rework of the whole thing? Issues and misunderstandings that are inherently bound to technical toolsets and further complicated by opportunities... I'm fascinated by the prospect of offering interactive 3D content in the context of scientific publishing, convinced it has the potential to make complicated text accessible to a wider audience. How much of traditional imagery making can be re-used in this process?

    Here's to a bright, scientifically illustrated future and all these crazy good artists and scientists paving the way!

  4. Great post! One thing for the scientist to consider is that an honest constructive critique wont hurt the artist's feelings. If the artist in question takes their work seriously they should expect a good critique. Our job is to create as accurate (and compelling) an image of the creature we are depicting. That is something I believe we all strive for. If I were lucky enough to have the opportunity to work directly with a scientist I would would be happy to have their expertise brought to the table.