Friday 17 February 2017

Scientists: please pay more attention to palaeoart

A few years ago I wrote about how the 21st century is a terrific interval for palaeoart because of the wealth of information, discussion of palaeoart theory and diversity of talent we presently enjoy. Never before has so much data on this topic been available to practitioners of the trade, to scientists and historians, or to curious members of the public. The increase in palaeoart talent, a movement of artists exploring the outer regions of palaeoart science and artistry, and hints of wider interest in art of fossil animals are all traceable to these recent developments.

Any palaeoartist you speak to will tell you that there are honed practises and processes that should be applied when executing a palaeoart study. Several decades worth of influential artwork and writing by the likes of Knight, Hallett, Paul, Antón, Conway et al. and others send a unified message: reconstructions of fossil animals should be produced through close study of animal anatomy, both fossil and modern. They demonstrate that palaeoartworks must pass some basic, almost objective tests to be considered scientifically credible, successful examples of the medium.
  • The proportions of the subject should be accurate to those indicated by its fossil remains
  • The skeleton of the subject should fit within the restored soft-tissue volume
  • Depicted poses should conform to predictions of joint motion
  • The soft-tissue volume should - at minimum - be informed by predictions of muscle bulk derived from fossil remains as well as relevant data from modern, related species
  • Soft-tissues restoration should be informed where possible by fossil data, or else via robust predictive techniques, such as phylogenetic bracketing
These points are not controversial. I'm sure any number of modern palaeoartists would list these as their baseline, entry level requirements for bona fide, scientifically-credible palaeoartwork. Further considerations of posing, composition and behaviour become more subjective and debatable - ideas of what we think 'looks right' or is 'most likely' in these areas are influenced by personal preferences, artistic styles and intent of the artwork. Successful palaeoartists can have contrasting ideas about these latter points, and that's fine: so long as our work remains within the realms of scientific plausibility, we are free to experiment and develop unique styles. But at the core of any palaeoartwork must be a reconstruction of a species that conforms to those fundamental aspects listed above.

Although it feels like we live in an enlightened age for palaeoart, some artworks associated with the very people who should be sticklers for scientific precision and reconstruction plausibility fall well short of the most elementary aspects of fossil animal reconstruction. These are not reconstructions for TV shows or films, where creative forces override scientific input. They are not illustrations for books where an overworked generalist illustrator is given a few hours to render an animal they'd not heard of until that morning. These are artworks produced for papers and press releases where scientists - researching palaeontologists with direct access to fossil material and technical literature - have every opportunity to guide and shape the artistic process. And these works are sometimes so scientifically awful that they're almost insulting to those of us who strive to produce credible prehistoric imagery, being critically flawed at the most basic level.

The level of failure in some artworks (not linked to here out of politeness) is sufficient to question whether those involved knew anything about reconstructing anatomy or if they really cared about the artwork at all. And yes, I think we do have to look at the scientists and researchers as being ultimately responsible here. These are artworks produced directly under their control to be associated with their work, and without pressure from publishers or media producers to be fantastic, weird or sensational. Scientists made the decision to produce these palaeoartworks, chose the artist to execute it, chose the level of input to have during its production, and signed off the final product.

In these artworks, even errors which scientists can objectively veto (such as proportions and bone articulations, elements of form such as skull or tooth shape, well-studied soft-tissue anatomies like feathers arrangements on dinosaur wings) are ignored. The result is that these pieces perpetuate errors that were realised as problematic years ago: under-muscled, 'shrink-wrapped' animals; 'bunny handed' theropods; feathered maniraptorans with three free fingers extending from their wings; ichthyosaurs with visible giant eyes; pterosaurs with enormous torsos and so on. The worst offenders show no grasp of basic aspects of animal anatomy, fossil or modern, with outlandish ideas of muscle distribution or proportions which are falsified by the most cursory glance at reference material. It is no exaggeration to say that some recent scientist-led palaeoartworks would not look out of place if produced in the 1830s. 

And we - educators, scientists, palaeoartists - should feel ticked off about this. Scientist-led palaeoart should be the best there is: carefully-executed, evidence-led syntheses of research conclusions in compelling artworks. It should convey to people how the subject appeared and behaved based on both new, cutting-edge research and the best of the science which preceded it. There is no reason not to take the same attitude to our palaeoart that we do to the rest of our studies. It is frowned upon to take half-measure approaches to descriptions, statistical analyses or cladistic methodology, so why is palaeoart exempt? Making crass, basic errors in animal reconstruction is no different to executing a flawed study or analysis. Both ignore data, advice and theory documented in palaeontological literature, and both show little regard for the techniques developed by pioneers of the process. Moreover, when they make it to publication, both imply that half-baked approaches are worthy of equal consideration to more carefully executed examples. Scientists will know the feeling of frustration when work directly relevant to a paper is not cited: lousy palaeoart is guilty of ignoring the theory and development of an entire field

A baffling aspect to this problem is that scientists routinely seek expertise lacking on research teams. Need a fossil prepared but lack the skulls? Seek assistance from a preparator. Need to crunch some stats but not sure how? Contact a colleague with statistical expertise. What we don't do with our science is assume our intuition and instincts about a topic are enough to guide us alone: we defer to those with the training, specific knowledge and experience to do the jobs we can't. And we would never employ an equally inexperienced individual and guide them through a process we lack all experience of ourselves.

But this is exactly what we do with palaeoartistry. Executing a palaeoart study requires a grasp of anatomy (and not just bones!), an ability to reconstruct/interpret fossil remains, a healthy grasp of living animal form, and an ability to translate all this into an artistic creation. These are not skills that everyone has, or that even all palaeontologists have. There are numerous specialisms in palaeontology, and not all of them are associated with the expertise ideal for consulting on palaeoartworks. Fossil bones do not exude a radiation which means those who work with them automatically know everything about palaeoart methods and theory. And yet it seems some scientists think it does, resulting in ill-founded advice for naive artists and approval of poor, flawed work. I am not the first person to raise this point (it has been mentioned in palaeoart literature since the 1980s), but it seems to fall on deaf ears.

Some readers may be wondering if this matters - so what if we have the odd wobbly looking reconstruction every now and then? Consider these points. Firstly, if scientists are so relaxed about palaeoart that they have no regard for even getting fundamental aspects correct, then what, really, is the point of the art in the first place? What can art of that quality really add to our field? It can't be held up as an accurate representation of the animal itself, and knowledgeable educators will avoid it or abandon it the moment a superior alternative becomes available (which, given the popularity of palaeoart online, is normally a couple of days after a new discovery. Indeed, awful PR palaeoart normally spurs more alternative versions, and with faster turnaround times). Badly produced palaeoart is basically destined to be ignored by those in the know, reflects poorly on those involved in its production, and ends up being an embarrassing aspect of the publication.

Secondly, scientist-led palaeoart is often the basis for derivative artwork, whether it's good or bad. Whereas people might expect prehistoric animals seen in film and TV to be embellished and enhanced, scientist-endorsed artwork carries the weight of expert approval. For non-specialist illustrators, they're an obvious source of information and errors are carried over into next generation work. Scientists need to realise that the half-lives of palaeoart are often much longer than any press articles or even scientific papers: they have long-lasting impacts on public perception and even inform scientific hypotheses. Darren Naish recently wrote more about these issues at length here.

Thirdly, there are scores of competent palaeoartists awaiting opportunities to work with scientists, and their prior knowledge of reconstruction processes and anatomy would fill knowledge gaps in some teams. Not only do these individuals have the skills needed to understand a fossil specimen and technical paper, and are thus able to produce credible artwork without constant academic input, but their experience means they can converse with scientists at (or close to) a technical level. This allows for detailed conversations about the specifics of the reconstruction and development of new ideas and insights into the life appearance of the subject organism. Experienced palaeoartists are more than just people who make pretty pictures: they're peers and colleagues of scientists, and able to augment research when given the opportunity.

Lastly, it is widely known that the palaeoart industry has a problem maintaining employment for even its most talented individuals, and in this context hiring non-specialists, especially if the research team is not palaeoart savvy, is ludicrous - why not hire the right people for the job? There are many early-career palaeoartists available if tight budgets are a concern, as well as numerous veterans who can offer highly polished art and rapid turnaround times if time is tight. Finding these people is as easy as opening modern palaeoart books, asking colleagues for recommendations or even a Google search. The wealth of easily-accessed palaeoart talent makes it inexcusable not to bring specialist artists on board for palaeoart projects.

And 'inexcusable' sums up my feeling on this topic pretty well. The fact that many scientist-led artworks are really amazing shows that high quality palaeoart of this nature is achievable if scientists care enough about its production. But the availability of palaeoart-relevant information, the growing body of literature on palaeoart theory, the willingness and accessibility of talented artists, and the demands of modern scientific standards make academically-driven, scientifically-rotten palaeoart inexcusable in the modern day. I'm not arguing that scientist-led palaeoart has to be perfect. I'm not arguing that scientist-led palaeoart has to conform to specific conventions of style, or to constrained ideas of life appearance. But I am arguing that scientist-led palaeoart should look like someone gave a damn about the final product.


  1. Have you had any experiences with public stations? I'd extend this argument / plea to that realm as well... and be equally optimistic (read: sarcastic) about it being heard. Scientific editors I've dealt with have often lacked 1) dramatic abilities 2) scientific knowledge 3) understanding (or even tolerance) of cg workflows.
    It would put a certain bounce in my walk to see their televised world declining into insignificance if it weren't for the fact that the new social-medial replacement just proved its incompetence in a sequence of geopolitical disasters.
    I DO have some really nice whiskey though.

    1. ...and the palaeontologists adoration of alcohol is explained!

  2. With respect to your first few bullet points: I put forth (in a book review for Natural History some years ago) Holtz's First Law of Paleoart: if the skeleton doesn't fit inside your reconstruction, your reconstruction is wrong.

    1. Hmmm... maybe there's actually a need for a 'how to peer review palaeoart in papers' guide. I imagine a lot of scientists wouldn't touch art commentary with a barge pole, but aspects like fitting the skeleton inside the reconstruction are obvious things to look for, and anyone can do it. I might take that idea up!

    2. David Marjanović19 February 2017 at 03:33

      Specifically, it's the reconstruction that's wrong, not the skeleton, even if it seems ludicrous; and it's wrong, not an alternative fact.

  3. Hi Mark,

    First, I love your artwork -- thank you for filling the world with your beautiful and accurate reconstructions.

    In my experience, the problem comes from two areas: funding and not being an artist. Regarding funding, when one applies for a grant or gets funding, there often is not money available for paleoart (although there should be given it's importance). So even if I really wanted to pay you for your work, I often simply can't. And I certainly don't want to approach an artist about work if I can't compensate them for their time and expertise. Having a brother who is a commercial artist, and having done all the artwork for my book, I understand and appreciate the time and dedication it takes to do things right. Perhaps this is where, as paleontologists, we need to make our voices heard about STEAM (Science Technology Engineering ART and Mathematics). I wonder if it were possible to start a campaign to voice this concern? Given the current politics in the US, I doubt incorporating monies for scientific art into NSF grants is going to be any sort of priority. However, I am in complete agreement with you that paleoart forms an integral part of what we do and how we communicate that with the public.

    The second thing is that I think some scientists don't want to challenge an artist because either someone is doing the art for free or because they feel they are such poor artists that they have no leg to stand on. Paleoart is one of those things that requires open channels of communication and the ability of the artist to hear the scientist and work towards accuracy, even if that somehow destroys a particular aesthetic, etc.

    Anyway, thanks for this blog post and I'd be interested to hear what you and other paleoartists think can practically be done to help this situation.


    1. Hi Matt,

      Thanks for this comment. I took your last point to heart and penned a follow up piece to this article, highlighting how even artistically-challenged scientists can bring their expertise to the table and guide palaeoart production.

  4. Probably entirely coincidental and not at all an inspiration for this piece is the rather sad and depressed, fish munching pregnant Dinocephalosaurus depicted in an otherwise rather important paper: Oh the melancholic look on it's face "someone made me do this, guess I'll just go eat some fish heads". Come on now that animal should be beaming with a proud, pregnant glow on it's face not looking like a rejected fish monger.

  5. I have to say, your comparison of artists to preparators and statistics-savvy colleagues is not entirely accurate - preparators, though fewer and fewer, have always been employed by host institutions have they not? And statistics-savvy colleagues get a publication out of it - or at least some favour in return - neither get one-off commissions...though I wish I got paid a commission for every statistical enquiry I've dealt with in the past...
    Having said that, I think the main issue is funding. Unless there is some specific reason to, I don't think it is very common to budget for artwork in research grant applications. So I don't know how researchers are supposed to take responsibility when there is simply no money.

    1. David Marjanović19 February 2017 at 03:41

      Preparators, though nowhere near enough, are indeed employed by host institutions. You need a cast done, you give your specimen to a preparator, a few days later you get the cast. You need artwork done, you have a problem.

  6. some thoughts...

  7. I think it works this way; if you don't care about it as art, you are not going to care about it as scientific representation.

    Who are the new wave of great Paleoartists since the late 80s - early 90s? I don't think our current value system allows us to hold anyone in esteem at the expense of the next person/method. There's no creative incentive and it becomes contagious.

    Paul W.

  8. David Marjanović19 February 2017 at 03:34

    Yes!!! Thank you!!!

    It is frowned upon to take half-measure approaches to descriptions, statistical analyses or cladistic methodology

    Less than you think.

    1. I remember my statistics teacher awhile ago saying that a lot of people , incl. Scientists, had no idea how to do statistics and would try to publish work with trivial errors like reporting dozens of P values and declaring that a value of 0.04 is clearly significant.

    2. A lot of statistics reported in the literature are either inappropriate, done badly, or simply not well thought out.

  9. I think maybe peer review should start considering any included paleoart along with the actual science. For example:
    "Smith et. al. (2017) misidentified the zygoapostrophe on the fourth quadrilateral pseudoungulate, and also included a truly god-awful life restoration of the type specimen."

    1. Maybe that would be a professionally acceptable way to "shame" scientists into doing better next time. Especially since you were too polite throw any offenders to the wolves of the internet, Mr. Witton.