This article is so wrong it makes me mad. I don't even care that she got a pulitzer for that stupid Seattle earthquake article, too.

She's written a treatise on myth without recognizing the purpose and characteristics of myth. Bigfoot is a large, gentle non-human who lives where people don't go and has the ability to tear your arms off but doesn't. Ditto the Yeti and everything else within its paradigm. The Loch Ness Monster is something lost and fantastic that exists within the realm of the known. Unicorns are something rare and pure created to sell narwhal horn to European royals (who weren't about to understand "whale" but could wrap their heads around horse/goat). Dragons are lizards so big you're allowed to be scared of them (Chinese dragons don't breathe fire, they bring rain - the Chinese didn't burn witches, either).

These myths have some aspect that sounds plausible (people but bigger, dragon/dinosaur that somehow didn't die, horse with weird horns, large lizard) and something that sounds fantastical. The fantastical-sounding aspect allows storytellers to inject metaphor into daily life with a dose of plausible deniability. Sure, virgins can be used to attract unicorns. This makes virgins more valuable because we all know they're valuable, right? The mysteries of the Scottish highlands are so impenetrable that maybe there's a sea monster in that loch. Once upon a time there were giant lizards that required the most virtuous knights to slay them and protect the countryside - which is a tale usually used to prop up the knighthood. NOBODY is doing a calculus of "but there are no vertebrates with three sets of limbs". And, when dealing with metaphorical implausibility, nobody is dividing things into "possible" and "impossible."

Jaron Lanier talks about VR research in which his team constructed arbitrary creatures with arbitrary limbs and then arbitrarily assigned the motions of subjects to control those arbitrary limbs - in order to curl your left tentacle you must contract your right elbow while simultaneously extending your left pinky, for example. To their surprise, adaptation to a new motion schema takes a while but not forever; our framework for connecting muscle with movement is far more flexible than you would expect from a high order vertebrate whose compliment of extremities doesn't vary over our lifetimes.

Metaphor is similar - in screenwriting, we call it "storyworld rules." So long as I explain "if you die in your dreams you die in real life" satisfactorily to the tale I wanna tell, I can do Nightmare on Elm Street, Dreamscape, Inception, you name it. "Casting spells takes mental energy" equals "trains are harder to levitate than pencils" unless I add "and it's all arbitrary" because we understand that unless I tell you physics works completely differently, you'll go with what you know. That's the whole reason myth works - unless I tell you differently, presume the only things that have changed are the things I've told you up front (or made you discover). So stuff like this:

Angels, for instance, are physiologically unlikely: in addition to being able to fly (fine for birds, unheard of in hominids), they manifest a particularly extreme version of the limb problem, since, per various sources, they have not just two but in some cases hundreds of wings.

How fucking dumb can you be while trying to look smart? Angels include in their myth an omnipotent, omniscient God for whom there are no rules. The whole purpose of angels, from a myth-making standpoint, is you cannot know the unknowable. Angels exist solely to confound your understanding of the physical universe. Their mere inclusion demonstrates the failure of the model.

On the other hand, this theory leads us quickly into ontological problems: are we humans more like mermaids, or more like ghosts?

Eat a dick. Mermaids suggest that we do not understand the physical plain; ghosts suggest we do not understand the spiritual plain. Is a circle more like a cylinder or a sphere? The question illustrates the ignorance of the subject.