Last summer I walked on an exposed sea floor a mile out to sea. That morning was the lowest tide of the year at Duxbury Reef, an hour north of San Francisco, and a group of us set up camp the night before in a friend’s backyard in Petaluma so we could be at the beach by dawn. The shallow stones emerged as miniature archipelagos or crooked paths, effervescent tide pools swirling between them. Huddling around tide pools is child’s play — you crouch low, you wobble and lose your balance, you poke around the coral’s dark crevices. Walking on the ocean’s floor is also like walking on another planet: You turn your back on the shore’s promise and hope you don’t accidentally step on anything as you walk this craggy turf crawling with weird life.
I wasn’t alone. Duxbury Reef was the subject of a “bioblitz,” which is, like its name suggests, a bombardment of sorts. This one was cheerily titled “Snapshot Cal Coast” and organized by the California Academy of the Sciences. They gather scientists and “citizen scientists” into a “series of bioblitzes” up and down California’s coastline in an attempt to document every living being in a given area by taking geo-tagged photos using the iNaturalist app. The research objective: track “taxa” especially affected by warming oceans or disease. The organizers even included a “Most Wanted” list: all starfish, seaweeds, pink acorn barnacles, the Dark Unicorn Snail, European Green Crabs, Chinese Mitten Crabs, and sea anemone. During these bioblitzes, according to the bioblitz Wikipedia entry, scientists gather data, and hopefully, get lay people “interested in biodiversity.” All one needs is a smart phone and in the case of tide pools, appropriate footwear, preferably rubber boots.
The sea anemone’s message: “Get the fuck away from me”
I had neither, and wasn’t really interested in blitzing, but I supposed I was interested in biodiversity, although I am not sure what that means, exactly — to be interested in biodiversity. The phrase sounds passive and vacuous, an “interest” someone might write on their dating profile: film, politics, biodiversity. What I was actually interested in was sea anemones. They are tentacular carnivores with trigger sensitive stingers that look deceptively plantlike and cover the ground of intertidal zones. They are very strange and very beautiful.
Out on the tide pools of Duxbury Reef, I met a marine biologist who asked that I let her know if I spotted any translucent sea slugs lined with yellow bumps along their otherwise see-through bodies. A few minutes later she spotted one herself and called me over to see what she meant. The slug was no bigger than my pinky finger. Its body looked like it was made out of jelly. She crouched close and took out her iPhone, protected by a waterproof case. “It’s hard to get a clear picture through water,” she complained, and adjusted several angles before she was satisfied with the shot. She uploaded her picture as an “observation” to iNaturalist and suggested an identification (Ancula gibbosa). If two other professional “naturalists” agreed with her, the observation would be certified “research-grade.” That day, 358 observations where uploaded to the app (223 of them research-grade), with 110 species recorded, including 14 Mossy Chiton, 12 Striped Shore Crab, and 11 Sunburst Anemone.
A year earlier, at Muir Beach, I had poked around some exposed tide pools while tripping, and, enthralled, decided to stare a sea anemone down with my desire for contact, for some sort of felt recognition from this being. After a few seconds, my brain felt very hot, almost as if zapped. I instinctively covered my eyes and became convinced contact had been made. The sea anemone’s message: “Get the fuck away from me.” Dubious as it sounds, the takeaway stuck, even in sober reflection. Sure, poke around, but keep in mind: Not all life wants to be seen, or known.
The app iNaturalist is somewhere between a field notebook and an Instagram profile, “a social network where everyone can record and share their observations of living things.” It is also a crowdsource data set of geotagged images used to teach computers how to see and identify life. It’s one in a market of plant and animal identification apps that range from digitized field guides to, as one writer puts it, “real-life Star Trek tricoders.” They have names like PlantNet, Nature Gate, Like That Garden, iBird, Plantifier, Leafsnap, or FlowerChecker. PlantSnap CEO Eric Ralls, in a video about his app, explains this proliferation: “Thanks to the internet,” he tells us, “we instantly find the answer to any question.” Ask Siri how old Mark Cuban is? 58. What’s the capital of Texas? Austin. “What’s the name of this giant tree I am standing next to?” Siri is stumped. Ralls enters the description of an unknown plant (“large plant with purple flowers”) into Google, to no avail. He tries searching for corresponding local fauna, but can’t verify an answer. He too, is stumped. His frustrated reality bursts biologist Daniel H. Janzen’s space-cowboy-scientist fantasy:
The spaceship lands. He steps out. He points [a gadget the size of a cell phone] around. It says “friendly—unfriendly—edible—poisonous—safe—dangerous—living—inanimate.” On the next sweep it says ‘Quercus oleoides—Homo sapiens—Spondias mombin—Solanum nigrum—Crotalus durissus—Morpho peleides—serpentine.’ This has been in my head since reading science fiction in ninth grade half a century ago. I am sure it was in the heads of Linneaus, Alexander the Great, and Timid the Mastodont Stomper. And it has been on the wish list of every other human confronted with the bewildering blizzard of wild biodiversity at the edge, middle and focus of society.
It’s certainly been in the head of Ralls, who, flummoxed by the large plant with purple flowers, concludes that “there is no app for that.” Now, thanks him and others, there are dozens. But Ralls’s and Janzen’s frustrations belie a certain inconvenience that has bated and baffled scientist and programmers with tricorder dreams: taxonomically identifying a species is by no means a simple process, for a human or for a computer. This task requires zoological or botanical expertise that most lay “nature enthusiasts” lack. For example, if you try to identify the giant tree you are standing next to, you’d have to first ask: What species of trees grow in this region? Does this tree bear cones or needle-like leaves? Does its leaves have a scale-like texture? Or are they flat and smooth? If they are flat and smooth, are they simple, compound, or fan-like shapes? If the leaves are simple, are they opposite each other on a branch, or do they alternate? If they are opposite, are the leaves blade-like? Are they palmately lobed? Do they have sharply toothed edges? How deep are their sinuses? How long are their leaves? After living in Los Angeles, I can tell the jacaranda tree from the olive tree from the lemon tree in my backyard. Still, I’d likely struggle to name any other tree on the block, any stab a dubious guess. Even the professional taxonomists — a dwindling vocation — don’t have it easy. For one, there is no “official” list of all scientifically classified species to refer to, and the existing collections often conflict each other, biased in their selections and riddled with inconsistencies and confusions borne of individual taxonomical whims.
Taxonomically identifying a species is by no means a simple process, for a human or for a computer. We too can see only what we are taught to look for
Automated identification creates further problems. A computer might be able to identify a tree the way it is able to surveil and track faces, fingerprints, or handwriting: It “processes” an image — resizes it, compresses it, enhances it, geo-tags it — and then recognizes existing patterns in order to classify it. The Digital Automated Identification SYstem (or DAISY) has been able to do this with some success on insect’s wings, comparing the morphological features it “sees” in an image of, say, a moth to a database of distinguishing taxonomic markings for moths. But life’s variegated forms don’t neatly align into databases. Individuals of any species differ in their morphology, either from genetic distinction, accident, or age. If difference abounds within a species, similarity abounds among species closely related, tipping the scale towards a degree of nuance DAISY can’t quite push.
Other automated systems, most notably iNaturalist, apply artificial intelligence and deep learning to the problem, teaching machines how to teach themselves to see and identify species. The machine’s intelligence, however, is only as good as the datasets it studies, and there aren’t many large image sets of reliably identified species to teach a computer with. Hence iNaturalist’s crowdsourcing scheme: The average app user does the work of taking and uploading images and professional “naturalists” verify an identification. All those “research grade” observations are then used to create the deep learning training sets for iNaturalist’s algorithm. Still, this approach has no way of teaching a computer to deal with a species it has not been trained to see. This is something humans and computers share. We too can see only what we are taught to look for.
Here’s a characteristic anecdote from an iNaturalist user published in the Atlantic: “That same day, a large black insect landed on a flower in front of me, and I snapped a portrait of it before it flew off. It was a dragonfly, but what kind of dragonfly? Many of our experiences of nature take this form. You see something, but you don’t know what it is. You are surrounded by life, but much of it is anonymous.” Here, the curious urbanite witnesses another life form. They observe it, generally, but they don’t quite know it. Their desire for this knowledge is couched by the anxious realization that, when it comes to other life, they know little. It’s a dragonfly, sure, but what kind of dragonfly? They can see it, but can they recognize it? Can they name it out of its alienating anonymity?
The anecdote is reminiscent of another, by the French botanist and naturalist Michel Adanson from 1736. Without the ordering eye of natural science, Adanson writes, the world is “a confused mingling of beings that seem to have been brought together by chance: here, gold is mixed with another metal, with stone, with earth; there, the violet grows side by side with an oak. Among these plants too, wander the quadruped, the reptile, and the insect; the fishes are confused, one might say, with the aqueous elements in which they swim, and with the plants grow in the depths of the waters… This mixture is indeed so general and so multifarious that it appears to be one of nature’s laws.”
Adanson was a contemporary of Swedish botanist Carl Linnaeus, who had published his Systema Naturae a year prior, in 1735. In it, Linnaeus proposed a classificatory schema to classify all plants, known and unknown to Europeans, according to the characteristics of their reproductive parts. Later modes of classification expanded to include all morphological attributes of an organism, but what has stuck to this day is Linnaeus’ system of taxonomy — class, order, genus, species — and the practice of binomial nomenclature, applied to name all plants and all animals. “His schema,” writes Mary Louise Pratt in Imperial Eyes (2007), “was perceived, even by critics, as making order out of chaos — both the chaos of nature and the chaos of earlier botany.”
Linnaean taxonomy was a lens that, when aimed at Adanson’s “confused mingling,” saw the world in static columns of names in Latin. “Natural history,” writes Michel Foucault in The Order of Things “is nothing more than the nomination of the visible,” yet what it actually nominates is what gets to be visible and what is otherwise ignored. Natural history created “a new field of visibility” that was trained to look at other life and “see as little as possible.” Taste, smell, and touch are dwindled out of observation — they are too mutable for analysis. Sight is given privilege, though even what we see must be cropped into its simplest composition: “lines, surfaces, forms, reliefs.”
The curious urbanite witnesses another life form. They can see it, but can they recognize it? Can they name it out of its alienating anonymity?
What gets left out of the picture is the very condition of relation that sustains life — the violet growing side by side with the oak, the insects and reptiles that wander near its roots, the human that rests on its bark for shade, that leans in to sniff the violet’s scent. “The real, geographic and terrestrial space in which we find ourselves,” Foucault writes, “confronts us with creatures that are interwoven with one another, in an order which, in relation to the great network of taxonomies, is nothing more than chance, disorder, or turbulence.” Taxonomy overlays this turbulence, as Pratt puts it, with a “rationalizing, extractive, dissociative” vision, a “picture of the planet appropriated and redeployed from a unified European perspective.”
No one represented, and spread, this perspective more fully than the “herborizer,” a 17th-century nature enthusiast “armed with nothing more than a collector’s bag, a notebook, and some specimen bottles, desiring nothing more than a few peaceful hours alone with the bugs and flowers.” He was the passive cousin of the conquistador or the diplomat, doing the “innocent and imperial” work of Adam in Paradise, as one writer puts it, “walking around as superintendent, sticking on labels.” (An often-cited quote by Linnaeus: “God created, Linnaeus organized.”) His harmless assertion of taxonomical hegemony over Europe and her colonies actually produced commercially exploitable knowledge for the empire’s gain. He was a researcher, classifying, collecting, qualifying and quantifying imperial loot. “The systematization of nature,” Pratt reminds us, “coincides with the height of the slave trade, the plantation system, colonial genocides in North America and South Africa, slave rebellions in the Andes, the Caribbean, North America and elsewhere.… For what where the slave trade and plantation system if not massive experiments in … the systematization of human life, the standardizing of persons?”
“Pattern discrimination” or “pattern recognition” describes the process by which algorithms learn to look for predetermined forms in order to discriminate or recognize information from a turbulent, confused mingling of data. It is the method that data scientists (or the NSA) use to turn disorderly data into a legible form. “As with the gesture of Althusser’s cop,” Hito Steyerl writes, “‘recognition’ creates subjects and subjection, knowledge, authority and as Rancière adds, neatly stacked categories of people” (like Linnaeus’s neatly stacked categories of animals and plants.) It almost goes without saying that these methods of looking are imbued with racist, sexist, and classist views, that they are “fundamentally political operations.”
So, when writer Ferris Jabr recommends iNaturalist in the New York Times Magazine, arguing that “learning the names of wild things changes the way we look at nature and the way we think about it,” in a way, I agree with him. Most of us, he continues, are “oblivious” to other life “even in our own backyards” because “our perception is circumscribed by our ecological illiteracy.” Rectifying this “illiteracy” is “an exercise in perspective” necessary for conservation efforts hoping to forestall our current crisis of extinction. Even the founders of iNaturalist, in their mission statement, affirm that, aside from their scientific aims, their primary goal is “getting people to feel that the non-human world has personal significance, and is worth protecting.”
I would add that this is precisely the reason why we should ask how learning taxonomical names changes the way we look at nature and the way we think about it. What is this exercise in perspective teaching us to see? Taxonomy, and its gotta-catch-‘em-all call to classify and collect, actually propagates “ecological illiteracy.” It sees the world, and the life that saturates it as a vat of information, significant for its economic or scientific exploitability, to be tracked or hacked. It teaches us that we approach and see other life from a distance, as if distanced, or separated from what we observe. In other words, it teaches us to see other life as proximate to us, rather than knowing ourselves as an extension of it.
In 1735, the same year Linnaeus published Systema Naturae, the French geographer Charles Marie de La Condamine went on a scientific expedition to South American equator to help determine the shape of the Earth. The coincidence of these two events, Pratt argues, created a “new vision” she calls Europe’s “planetary consciousness,” a “basic element constructing modern Eurocentrism.” At the time, La Condamine wrote and worked before the taxonomical imperative of scientific classification took hold. As he traveled the Ecuadorian Amazon, measuring longitudinal degrees, he “notes prophetically” that the Amazon’s “diversity of plants and trees” would give botanists “ample employment for many years.” He then adds a note that Pratt argues “would, in scientific contexts, have become nearly unthinkable” by the end of the century:
I speak here only of the labor it would require, to make an exact description of these plants, and to reduce them into classes, and range each under its proper genus and species. What would it be, if we comprehend herewith, an examination into the virtues ascribed to them by the natives of the country? An examination, which is undoubtedly the most attractive of our attentions, of any branch of this study.
Conservationists working to protect biodiversity have begun, haltingly, to think the unthinkable. The Convention on Biological Diversity, a treaty signed at the United Nations Conference on Environment and Development in 1992, included measures to use and protect “traditional knowledge” as a part of a broader stake in conserving biodiversity. This makes sense in regions like the Amazon, where indigenous cultures continue to defend their sovereignty, and where this defense is often inextricable from the defense of the forest they live in. Still, foreign scientific efforts hoping to study, for example, the rainforest’s biodiversity — including its “cultural” biodiversity, so to speak — tend to replicate the very extractive methods their colonial predecessors practiced. Linda Tuhiwai Smith, a professor of indigenous education, writes about the colonial and imperial legacies of “research”— particularly research of or about indigenous people and indigenous knowledge — in Decolonizing Methodologies. “Imperialism and colonialism,” Smith writes, “are the specific formations through which the West came to ‘see’, to ‘name’ and to ‘know’ indigenous communities.” Like the taxonomical lens that scans animal and plant life through a European perspective of discreteness and utility, the methodology of scientific or anthropological research, even when applied to the benign aim of conservation, often conducts such “research ‘through imperial eyes.’”
“Indigenous knowledge,” write James (Sákéj) Youngblood Henderson and Marie Battiste, “is not a uniform concept across all indigenous peoples; it is a diverse body of knowledge that is spread throughout different peoples in many layers. Those who are possessors of this knowledge often cannot categorize it in Eurocentric thought, partly because the processes of categorizations are not part of Indigenous thought.” Here I raise an apprehensive and hopeful question: Rather than turning indigenous knowledge, and indigenous people, into objects of study, can we become its students?
Perhaps the first step towards such an education, what others like Smith might call a process of decolonization or a process of unlearning, is to take off our lenses and reckon with the humbling, bewildering condition of unknowing, to quite the appetite for legibility of the world that leaves us at a comfortable distance from what we cannot understand. We might want to put our phones away as we stick our face into the strange and the beautiful, recognizing its zap not necessarily as a rebuke but rather as the weight of its own stare, for once, meeting our own.