In my last post I wondered what it means to “discover” something that doesn’t actually exist, or to describe something before we have knowledge of it. How much of the unicorn was discovered, and how much was invented? The unicorn was, I think we can all agree, invented, but the point of this post is to consider further this question of the unicorn’s discovery. And to ask, what role does invention—imagination—play in science?
The unicorn is more than a myth, more than wonder word: it sheds light on scientific understanding, in particular on the way analogy works in scientific understanding. Analogy is sometimes called a “predictive metaphor.” 1 While both analogies and metaphors offer comparisons, the comparisons that analogies make are not purely idiosyncratic (“red is the sound of a trumpet” 2, or “effectrix of divine honey” for bee) but predictive (“hand is to glove as foot is to _______”). For this reason, metaphors do not claim any special relationship between the two terms being compared. Rather, the relationship is in–and reveals–the mind of the metaphorist. Analogies, on the other hand, claim a certain knowledge-value: they help our minds move from known terms to unknown ones.
Another way of saying this is that analogies are, as the Stanford Encyclopedia of Philosophy notes, a species of “heuristic”: they provide “aids to discovery” based on the properties of a known thing, which we can extend to the properties of an unknown thing. Heuristics and analogies are like rough-and-ready sketches that allow our minds to approach the unknown. It would be tempting to say that metaphor is to poetry as analogy is to science, but I think a more accurate analogy for analogy would be, analogy is to science as the hand-drawn map is to territory. In other words, we don’t require a satellite-generated topo-map to navigate territory—often a hastily scrawled crayon drawing on a napkin with a few points of interest will get us close enough to our destination. 3
Pliny the Elder’s 1st-century CE description of the unknown beast first described by Ctesias reveals something of the sketchy, predictive power of analogy. In Pliny’s account, the beast
has the head of a stag, the feet of an elephant, and the tail of a boar, while the rest of its body is like that of the horse; it makes a deep lowing noise, and has a single black horn, which projects from the middle of its forehead, two cubits in length. This animal, it is said, cannot be taken alive. 4
The sources of these analogies to stag, elephant, boar, and horse in Pliny’s account are unclear. We know that Ctesias did not make such references to other creatures in his account, except to compare its astragalos, or knuckle-bones, to those of an ox. However, between Ctesias in the 5th century BCE and Pliny in the first century CE, we find a possible source in Gaius Julius Caesar’s first-century BCE Commentaries on the Gallic War. Caesar apparently did a little natural history when not conquering the Gauls, and provided a description of animal in the Hercynian forests of Germany:
an ox of the shape of a stag, between whose ears a horn rises from the middle of the forehead, higher and straighter than those horns which are known to us. From the top of this, branches, like palms, stretch out a considerable distance. The shape of the female and of the male is the, same; the appearance and the size of the horns is the same. 5
In the above image, taken from the Stag Room of the Museum of Hunting and Nature, André Arbus’s statue of Actaeon, the Greek hunter transformed into a stag by the goddess Artemis for the crime of gazing on her while she bathed, takes the form of Caesar’s Hercynian unicorn. Arbus helps us see what Caesar thought he saw in the forest, and what Caesar thought he saw in turn might explain where Pliny got “stag.”
Walter Hyde, in “The Curious Animals of the Hercynian Forest” (1918), attempted to save the “sober historian” Caesar from trafficking in fables of “one-horned deer” and other chimera, noting that “it is now fairly well agreed among Caesar scholars that the whole account of the forest and its wonders is merely an interpolation into the body of Caesar’s work by some unknown scribe.” 6 So we may justly surmise that Pliny was a bit credulous to include this fabulous animal in his Natural History based on this and, before this, Ctesias’s limited accounts.
But then, how did Pliny arrive at his other comparisons? Is he—are all of these observers, from Ctesias forward—merely trafficking in fables as a result of some centuries-long game of telephone? Despite his own suspicion of Caesar’s account, Hyde may give us some clues. He notes that “It is well known that Roman writers called unfamiliar wild animals of large size boves, ‘oxen.’ Thus the elephant was known as the Lucanian ox…. Seals were also called ‘marine oxen.’” This is important, and reveals the analogical significance of Caesar’s—and Pliny’s— account that Hyde doesn’t quite understand, and which he in fact attempts to erase with the help of “unknown scribes”: it shows how one animal can become a heuristic, a map, to another. The animal itself ceases to be a collection of specific details and becomes a rough sketch for thinking, which is then used to compare to other animals in search of underlying similarities. Caesar was simply using “ox” and “stag” to describe something that was ox-large and stag-like. The only “fabulous” dimension of the animal is its single horn—and this would not have seemed especially fabulous to Romans carrying around with them ideas, like Ctesias’s, of single-horned animals.
Caesar and Pliny are employing some analogical reasoning based on such prior eyewitness accounts, using the parts of known creatures to account for the elements of the description of this unknown creature. They are, in a sense, modeling the unicorn using terms known to them and their audience. The destination might be unclear, but the coordinates they are using to plot the route are well known. They have, in other words, discovered a unicorn through analogy—predicting the existence of an unknown thing based on the existence of known things.
To think of these accounts this way reveals their real knowledge-value as heuristical, not simply empirical. In Gregory Bateson’s terms, they may have discovered, not a thing, but a “pattern which connects.” Pliny, in casting about for coordinates, might have freed the unicorn from its binding specificity and rendered it a tool for inquiry.
Bateson called this kind of thinking “transference,” or the seeking of underlying relationships between things—a kind of pattern-seeking in contrast to simple collecting or defining. Transference involves contextualizing details so as to allow for such pattern-recognition. His example is the elephant’s trunk:
What is the elephant’s trunk?…. As you know, the answer is that the elephant’s trunk is its “nose.”… And I put the word “nose” in quotation marks because the trunk is being defined by an internal process of communication in growth. The trunk is a “nose” by a process of communication: it is the context of the trunk that identifies it as a nose. That which stands between two eyes and a north of a mouth is a “nose,” and that is that. It is the context that fixes the meaning, and it must surely be the receiving context that provides meaning for the genetic instructions. When I call that a “nose” and this a “hand” I am quoting—or misquoting—the developmental instructions in the growing organism, and quoting what the tissues which received the message thought the message intended.” 7
Bateson is exploring the process by which we arrive at knowledge, a process I think is at work in early accounts of the unicorn. At the most basic level, we can call the elephant’s trunk its “nose” because of its context in relation to the rest of its features (above this, between that). The definition of “nose” ought not to be a list of characters: this would produce a venn diagram at whose center would be an absence—the noses of many species share a lot of overlap, but no two noses will ever be the same. Maybe we could say something about the “noseness” that all noses share, but that’s pretty metaphysical. Rather, the meaning of “nose” is its relationship to other features, its context. “Nose” is an analogy that reveals connections, relationships, rather than a set of characters.
(Bateson is also suggesting that this is more than a language matter: that the way we acquire our knowledge of the nose bears a relationship to how the nose acquires its knowledge of itself from its genetic instruction. The nose knows, indeed.)
The unicorn is like the nose—it does not exist except in the form of relationships, homologies or samenesses among things. Caesar and Pliny, with their primitive, ready-to-hand, folk categories (ox, stag, horn, etc.), scanned the world for relationships and patterns that connect. Sure such a map can be wrong, because based on the wrong kinds of coordinates (observable and superficial, versus underlying and fundamental, for instance), or the misidentification of coordinates (the single horn is most likely a double). But it is surely the right kind of wrong—a wrong that reveals the mind that connects, that sees patterns. It might not be a stretch to say that the same pattern recognition that went into identifying the unicorn went into the nineteenth-century English naturalist Richard Owen’s identification of the pattern of all vertebrate life in the spine of a fish. Beyond the pattern, everything else is just fill-in-the-blanks, “know-how,” says Bateson.
But we lose the mind in common between Pliny and Owen when we, like Hyde or even Sir Thomas Browne, apply to them only our ham-fisted divisions between the real and the unreal, between waking and dream-life. When Hyde invents “unknown scribes” to rescue Caesar from his naivety, he may only be revealing his own—his inability to think past details to the patterns that connect. If there is a key difference between Classical and Modern thinking when it comes to science, as Hyde or Browne want to suggest, it is in those early naturalists’ willingness to base their knowledge on something other than empirical observation—namely, on imaginary categories (inventions) to be later filled with things (discovered). Men like Ctesias or Pliny dreamed first, and they dreamt in analogies.
In the case of the unicorn, the dream proved both more powerful and more elusive than the reality. To our better-informed modern eyes, these accounts reveal that the unicorn is likely not a single imaginary beast, but a collection or composite of many actual ones—a bestiary, a compendia, of creatures distilled into a single account.
Subsequent natural historians have spent a lot of time trying without success to parse the unicorn into its various components. Eighteenth-century French naturalist Georges Cuvier tried in his own way to put an end to the unicorn as a heuristic, hitching it to a single animal: the rhinoceros (Latin for rhino (nose) + ceros (horn)). This would certainly explain the single-horned appearance, the deep lowing noise, and the fact that the animal would have been hard to take alive. It would also explain the speed of the animal, given that a rhino gains in speed the farther it run due to its bulk. Likewise the ox. Moreover, rhinos are solid-hoofed, like the elephant, not cloven-hoofed, like many other ungulates.
Other features, though, are not accounted for: the 1 1/2 – 2 cubit-length horn; the animal’s color (white); its relative shape (ass- or possibly stag-like). The stag, the ox, the oryx, and the ass would also seem to be likely contenders, but again, these explain some but not all of the characters of the unicorn. It might point to a gazelle-like creature, like the scimitar oryx (Oryx dammah) of north Africa.
But an oryx, like a stag, has two horns–unless of course you’re looking at one in profile and from a great way off. Further, the oryx is cloven-hoofed, meaning that the “feet like an elephant” (or, solid-hoofed) descriptor would rule out this animal. As Aristotle wrote, “Most of the horned ones are cloven-hoofed, e.g., the ox, the deer, and the goat; we have no solid-hoofed animal with a pair of horns.” Cloven-hoofed creatures can’t grow horns in the middle of their forehead because their skulls are divided down the middle, not solid. Finally, the geography is all wrong: north Africa is a ways off from India, and neither of which are close to the forests of Germany.
This is but a partial list of likely candidates for the unicorn. Chris Lavers, in his The Natural History of Unicorns, puts together a comprehensive history of this composite creature that includes even more animals archived within Ctesias’s and Pliny’s descriptions. These include the Tibetan kiang, a fleet animal that can outrun a greyhound, the Tibetan chiru, a notoriously elusive type of goat with long black horns, and the yak, a fierce and unpredictable type of cow.
The “discovery” of the unicorn, we might say, predicted, or at least anticipated, our knowledge of these other creatures whose existence was, at the time, equally unimaginable. In a sense, the invention—the dream—of the unicorn anticipates the existence of things that would only later emerge as real. We could say the early naturalists helped draw a map that led, maybe not to the destination they sought, but to a lot of other places.
This much is certain: someone somewhere saw something(s) they’d never seen before, which they framed, described, using the fragments of things they had seen. In doing so, they invented a known unknown. Call it a map, or maybe the first field guide into nature. Their analogies produced that venn diagram of overlapping circles, and at the center we expected the ding an sich, the thing itself. But instead, we found a pattern that revealed, and modeled, the existence of other unknowns in other contexts—rhino, oryx, kiang, chiru, yak. But some–men like Browne, Cuvier, or Hyde–mistook the pattern for an absence, a yawning emptiness that they called, somewhat angrily, a fable, a dream. Perhaps they were just really disappointed: they were promised a unicorn. Perhaps they weren’t as taken with the ark of creatures we found instead.
In find it, if not analogous, at least metaphorical of this disappointment that Actaeon, the hunter-turned-stag, is, in a bit of unfortunate turnabout, torn to pieces by the dogs that once were his hunting companions.
The unicorn fared little better. It would eventually be destroyed by the very tools of science that it helped to fashion. More on that later. But unlike poor Actaeon, the unicorn did manage to live on, at least for a while, because a pattern is more powerful than a data point. Its unknown destination gave the idea of the unicorn legs, literally and figuratively, and on these it traveled from North Africa to India to Germany, and from the ancient world to the modern. How long those legs were will be the subject of my next post.
- Gillian Beer, Darwin’s Plots: Evolutionary Narrative in George Eliot and Nineteenth-Century Fiction. Cambridge, 2000. ↩
- I borrow this one from Thomas Nagel’s classic essay on the philosophy of mind, “What Is It Like to Be a Bat?” The Philosophical Review. 83:4 (1974). 435-50. ↩
- And some say this kind of analogizing and model-making form the basis of how we think. See for example Gilles Fauconnier and Mark Turner, The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities (Basic, 2003) and Douglas Hofstadter and Emmanuel Sander, Surfaces and Essences: Analogy as the Fire and Fuel of Thinking (Basic, 2013). ↩
- Pliny, Natural History. Trans. John F. Healy. Penguin, 2004. 281. ↩
- Caesar, Commentaries on the Gallic War. Trans. W. A. McDevitte and W. S. Bohn. New York: Harper, 1869. 27. ↩
- Walter Woodburn Hyde, “The Curious Animals of the Hercynian Forest.” The Classical Journal 13. Jan. 1918. ↩
- Gregory Bateson, Mind and Nature: A Necessary Unity. 1979. Hampton Press, 2002. 14. ↩