We use cookies to provide and improve our services. They allow us to remember some of our preferences and improve the overall site performance. View our Privacy policy.
About Ilana Bean
Ilana Bean, an accomplished Irish essayist, has been selected as a City of Literature Virtual Writer-in-Residence at Melbourne Connect. This esteemed program aims to foster connections between writers of various disciplines including poets to playwrights, and Melbourne’s diverse community.
Focused on a long-form nonfiction project exploring how medical illustration shapes cultural understandings of the body, Ilana’s residency involves engaging with researchers from across Melbourne Connect. From her insightful exploration, Ilana has written a compelling essay that delves into the intersection of art and science.
For further information on Ilana’s work visit ilanabean.com.

Like many people, Bernhard Siegfried Albinus longed for the perfect body. But he wasn’t concerned with his own appearance: instead, Albinus—an 18th century Dutch anatomist—wanted to work alongside his artist Jan Wandelaar to document and analyse the ideal human specimen.

Albinus believed that all people were flawed variants of the homo perfectus. He didn’t like that medical illustrations, like those done by Vesalius and his artists, often depicted a specific person. By “collecting data from one body after another, and making a composite according to the rule” Albinus believed the “actual truth will be displayed.”

Drawing the perfect human didn’t just mean surveying, measuring, and averaging out the limbs of different cadavers. Albinus was also looking for beauty. He searched for human specimens that met his expectations of ideal proportions, and, as one writer put it: “showed all signs of strength and agility, one that was elegant but at the same time not too delicate, that showed neither juvenile nor feminine roundness and slenderness nor uncouth roughness and clumsiness; in short, one whose parts were all beautiful and pleasing to the eye.” The skeleton he thought came closest to perfection was 167 centimetres—interestingly, the same height as Albinus himself.

Still, even the best specimens he found fell a little short. He eventually chose a model skeleton, but still used parts from other bodies that he thought were a little better, compositing and stitching them together into the perfect form. He thought the way that anatomists could get closest to describing a universal truth was by depicting a body that never existed.

Today, medical illustrators still prioritise ideal, universal figures over specific ones. This makes sense for a lot of reasons. Including an outie belly button, for example, could be distracting, or even make viewers think that navel protrusion is a symptom of whatever condition they’re seeing. The images are supposed to represent a whole population, and teach people how to approach all sorts of bodies. Not to mention, it might be unsettling to open a surgery manual and see somebody you’ve met looking back at you. At the same time, this practice of standardisation calls for consideration. Medical illustration has historically lacked diversity, and this has real life consequences—like higher rates of undiagnosed skin conditions in patients with darker skin. As medical illustrators work to balance diversity and universality, it’s important to ask: Who gets to decide what the “generic” or “ideal” body looks like? How does this affect people whose bodies aren’t classified as “typical”?


Medical illustrators show people parts of the body that are usually hidden—and this responsibility requires a lot of trust. Because the average person can’t simply observe something like a developing human embryo themselves, they have to be able to believe that a medical illustrator is representing the world honestly and accurately. But what happens when artists—perhaps intentionally—distort medical images to make a point?

This may have happened in the case of Ernst Haeckel. Born in 1834, Haeckel was a scientist, artist, and early supporter of evolution. Like Darwin, he believed that humans were descended from ancestors of different species, but unlike Darwin, he didn’t credit this to natural selection. Instead, he thought that evolution has a built in direction—that animals were always advancing to a higher form.

The chart shows Haeckel’s view of phylogeny—a term that refers to a group’s evolutionary history. Haeckel places humans on top of a ladder, which ranks animals in a hierarchy. On the right is a more contemporary version of the chart, which depicts evolution as less like a ladder, and more like a tree. It tracks how species are related, but doesn’t imply that there’s a destination. In this version, animals aren’t evolving into something better, rather than something better suited for their particular environment.

Haeckel believed that during embryonic and fetal development, more advanced creatures like humans went through stages where they looked like their primitive ancestors. First, a human embryo might go through a fishlike phase where they would have gill-like structure. Later, they would turn into a pig-like fetus. Haeckel created a series of images comparing the development of embryos of different species to make this argument. By exaggerating some features and leaving out others (like not showing limb buds in some mammals), he distorted the drawings to argue his point. When you look at his illustration compared to later photographs, you can see how different they are.

Haeckel’s drawings didn’t just stay in his notebook as a fringe theory. They actually appeared in textbooks—and kicked off the long-running myth that human fetuses have gills. Now, scientists see them as inaccurate—and many argue, fraudulent.

Haeckel’s work left a dark legacy. Many of his ideas— including biological hierarchy—had long lasting impacts that fueled scientific racism. His work was read by leaders in the Nazi movement and used to support their pre-existing biases.

We can think of Haeckel as a case study about the danger of dishonesty in medical illustration. Since Haecekl, molecular research has shown that early embryos from different species do have a lot in common. But it’s not because they’re changing into each other: we all start out with similar bases, and then gain specific features that make us ourselves.