David Weinberger
Photo: AP/Steven Senne Editor's note: Following David Weinberger's introduction is an excerpt from his recently published book Everything Is Miscellaneous.
In the age of social tagging and folksonomies, where we all get to classify the things of the world into categories we make up on the spot, Linnaeus the Taxonomist seems quaint. But we shouldn't feel too smug. We are struggling against the same limitations as he did … and this time we don't have an excuse.
Before Linnaeus, it was assumed in the West that God has perfectly ordered the universe. Each thing has its place, we assumed, clustered with other things like it, but also each essentially different from the other things in the cluster. The clusters are themselves clustered, creating a Tree of Everything, each branching determined by a perfectly unambiguous definition.
But, although God knew the definitions, it was sometimes hard for humans to tell what He had in mind. What were the relevant principles of likeness and difference? Does it matter more where they live, how they look, or how they behave?
Linnaeus gave up. Instead of trying to read God's mind, he proposed a classification scheme that would at least let the world's scientists agree they were talking about the same thing, even if they didn't know how Jesus would classify a platypus.
Linnaeus was confident there was a real order, and hoped that he got much of that order right, but like a folksonomist, he wasn't willing to wait for the revelation.
Unlike folksonomists, however, Linnaeus was constrained by his medium -- paper -- to come up with a single order. Despite the impression sometimes left, folksonomies are not bottom-up taxonomies that replace top-down taxonomies. Thinking about folksonomies that way misses their most important virtue: Folksonomies contain the information required to let us have multiple simultaneous categorizations. The majority of taggers may tag "las vegas" as "vacation," but those who think of it as "sodom" can find their way through the data as well.
That's the big change the rise of the miscellaneous brings. We're adding massive amounts of metadata -- tags, links, playlists, even taxonomies -- to all of the resources available on the web without prior planning and coordination, making a huge mess. But, that mess actually enhances the available ways we can find and make sense of what's available to us. All that unplanned metadata lets us pull pieces together, and then it helps us contextualize and understand those pieces.
Until we started digitizing everything, we organized either the physical things themselves (what Everything Is Miscellaneous refers to as the first order of order) or we physically separated the information about the things and organized that (the second order): Think of books and card catalogs, or merchandise on racks and a catalog of products. With the third order, for the first time we can organize information, ideas and knowledge free of the limitations of the physical. And that enables us to get past the notion that there must a single right order, whether it's Aristotle's, God's, or Linnaeus' best guess.
__Laundry and Linnaeus
__Aristotle lays out a task for all those who want to know their universe: Go forth and lump and split.
"Lump" and "split" are not Aristotle’s words, but, surprisingly, they are technical terms among professional indexers. Seth Maislin, a member of the board of directors of the American Society of Indexers and a consultant on indexing to the likes of the United Nations and Microsoft, explains: "A lumper takes things that seem disparate and combines them because they have something similar. A splitter tends to take two things that are lumped together and separate them into smaller categories." Indexers tend to be one or the other, their technique driven by their personality.
Every day we face the same choices as professional indexers. Some of us store all our bed linens in one pile in the closet, while others of us separate them by bedroom, color, weight, and season ... and then arrange each little pile so the least-worn sheets are on top. When asked if we know the way to San Jose, our directions lump together a long stretch of road rather than counting the precise number of lights -- "just keep going for a while" -- but split out the strip mall on the right because we think it will be a useful landmark. And much of our conversation is about the right lumping and splitting. Your friend didn’t like the movie last night because she thought it was supposed to be a comedy. No, you say, it wasn't supposed to be a laugh-out-loud comedy. It was more of a chortle-inwardly comedy ... more like Amélie than Animal House. We are constantly negotiating life's lumps and splits, from trying to decide which kid gets to ride in the front to arguing over health-care reform.
The remarkable fact is that we have built systems for understanding the universe using the same technique we use for putting away our laundry: Split the lump of cleaned clothes by family member, split each family member’s lumps by body part, then perhaps split by work or play, by season, or by color. If you're not sure whether the ski socks go with the normal socks, the winter wear, or the sports wear, you're still going to have to pick one because they have to go somewhere, and they can go only one place. That's just how atoms work. At the end of the process, you will have created a conceptual tree of clothes, with each family member as a main branch, the body parts as subbranches, and the ski socks hanging from whichever branch you finally chose. That's how we get through laundry day. Yet Linnaeus, the father of our modern way of organizing nature -- the person who split the universe into the animal, vegetable, and mineral lumps that start almost every round of Twenty Questions -- used those very same laundry-sorting principles to describe the structure of the entire natural world, from rocks to primates.
Born in 1707 to a pastor and his wife in the small town of Stenbrohult, Sweden, the young Carolus Linnaeus was fascinated with botany. He became a physician, but he spent most of his life devising and applying a system for classifying natural objects. His way forward was cleared by a heroic act of lumping accomplished by the French naturalist Joseph Pitton de Tournefort thirty-five years earlier. Tournefort introduced the notion of the genus (plural: genera), clustering the six thousand known plant species into just six hundred groups. That pared down the number enough so that Linnaeus could classify plants by looking at the shape, number, relative size, and arrangement of their stamens and pistils, the plants' reproductive organs. Linnaeus's system had enough variations -- 5,776 by his calculations -- to let it account for all the plant genera. Because the system utilized easily observable characteristics, with just a little training anyone could examine a plant’s parts and know where it fit. And because the parts of a plant come in identifiable and countable units, there were no messy borderline cases.
Linnaeus's system worked. But, even though he was the son of a clergyman, and even though he knew Genesis by heart, Linnaeus did not believe that the system of classification he had published revealed God's order. Although at times he hinted that it did -- nature, after all, was assumed to be a book written by God -- Linnaeus seemed comfortable with the idea that he had spent his life devising an order that was useful if not true. This was a century before Darwin showed how animals could be grouped not by mere similarity but by causation: Humans and chimps go on the same branch because the chain of causality leads back to a common ancestor.
It's hard to imagine today what it meant to categorize living creatures without evolutionary theory. With evolution, the tree of species is a family tree. But if creatures didn’t evolve one from another, then putting them on the same branch signified only that they resembled one another. Is a bat a type of bird, because it flies, or is it a type of mammal, because it has fur? Linnaeus didn't think he could resolve such questions finally because he could not read God's mind. But, like Melvil Dewey, Linnaeus believed in the efficiency of rationality. Instead of advocating spelling simplification, Linnaeus came up with a highly efficient and orderly way of naming and organizing species, so scientists could agree on what species they were talking about, a condition for scientific progress.
Linnaeus also promoted the "binomial" system of naming, replacing names such as Grossularia, multiplici acino: seu non spinosa hortensis rubra, seu Ribes officinarium (the European red currant) with two-word phrases such as Ribes rubrum. The first word of a binomial is the genus and the second indicates something specific to that species. So humans are Homo sapiens ("men who are wise") and the painted sage plant is Salvia viridis (“a healing plant that is green”). Aristotle himself laid the groundwork for binomials in the Metaphysics by pointing out that if a classification in the tree is right, anything more than a genus-species name is redundant: There’s no need to call Socrates a "human biped animal," because all bipeds are animals. (Carolus Linnaeus’s own name -- his father changed the family name from Ingemarsson to the Latin Linnaeus in honor of a three-trunked linden tree in their yard -- is itself sort of a binomial, albeit in species-genus order.)
But a system so important in the history of natural science can't be explained purely through whims of personality. Linnaeus had a political aim: He -- like Dewey -- wanted to democratize knowledge. Linnaeus was a doctor and a teacher committed to spreading knowledge far and wide. Twice a week in the summers of the 1740s, he would lead up to three hundred people, including women, on twelve-hour natural history walks. Linnaeus's method of classifying plants was easy to teach and didn’t require special equipment. To determine a plant’s class, you first check the stamens and pistils. If they’re in the same flower, it's "monoclinous," which translates to a marriage in which the husband and wife share a bed. If there is one stamen and one pistil, it's a monander -- one husband in the marriage. If there are two stamens, it's a diander -- two husbands in one marriage, and so on. Even Linnaeus, the pastor’s son, knew: Sex sells. In fact, one German botanist declared the system’s reliance on sexuality immoral.
You can see one more inspiration for his system by visiting the headquarters of the Linnean Society, in London. The society's entrance is tucked away in a courtyard shared with the far larger Royal Academy of Arts. A set of statues outside the Royal Academy provides a small tree of intellectual heroes: Cuvier and Leibniz to Linnaeus's right, and Newton, Bentham, Milton, and Harvey above him. Inside, the headquarters are very British nineteenth century, done in mustard and parchment, wood and brass. The Swedish patriot’s lifework is here because Linnaeus's widow sold it to a rich young British scientist eager to make his mark; she needed the money for their daughter's dowry.
On the first floor of the building's library, heavy cloth covers a couple of glass-topped tables that exhibit some of Linnaeus’s original specimens, the husks of the beings Linnaeus held in his hand when he said, "I name thee ... thus!" They serve as the reference points for disputes about whether a particular binomial refers to this or that creature. The official brochure notes that the collection includes 14,000 plants, 158 fish, 564 shells, and 3,198 insects. The collection room itself is below ground, protected by a six-inch-thick metal door and designed to survive a nuclear bomb. "The whole of the taxonomic world depends on the legal concept of the type," Gina Douglas, the society's librarian and archivist, explains. It makes sense to bury first- and second-order organizations such as this one and the Bettmann Archive. Specimens made of atoms are fragile and need protection.
Inside the vault, the room feels oddly homey for a bomb shelter. It's only about fifteen feet square, but the wood and brass of the specimen drawers and bookshelves that line its walls lend the cramped space the air of a reading room at a gentleman’s club. Douglas spreads out some framed specimen pages, each with one plant specimen, gray as dust, attached. "Notice the K on that one," she says, pointing to a small letter at the bottom of the page. "That tells us who collected it. It's rare for a page to have that information." Too bad, because some of the specimens in the cases upstairs had been misidentified. The note card for Solanum quercifolium explains that Linnaeus grew the plant from seeds he thought were from Peru but were actually from somewhere near Mongolia. If he'd had the name of the collector -- important metadata -- he might have avoided that mistake.
Douglas opens a first edition of Systema Naturae, in which Linnaeus dared to classify all of nature in just eleven pages. Of course, Linnaeus had to make the book the size of a small coffee table to fit all of nature into it. Douglas gingerly turns the pages -- like turning down the sheets on a bed -- to reveal the three double-page spreads at the tree's root: animals, vegetables, minerals. Linnaeus has caged the animal kingdom in six major boxes, four for vertebrates (mammals, birds, reptiles, and fish) and two for invertebrates (insects and a category called "worms" that included everything else -- a squirming mass of the miscellaneous). In the upper-left box for animals with backbones is a box for mammals, topped by primates, into which Linnaeus – radically -- put monkeys next to humans, referring to orangutans as Homo sylvestris, "feral man of the woods."
Although the boxes are nested, Linnaeus maintained his version of the Great Chain of Being, ranking each of the species within them. The method is like having separate classrooms for the advanced, average, and slow students and arranging the chairs in each according to the individual student’s grade point average -- or like Melvil Dewey clustering books by topic and subtopic, but still assigning each a decimal number so they can be laid out in neat rows on shelves. The principle of Linnaeus's ordering of the species is harder to figure than a grade point average, however. The medieval version of the Great Chain sorted creatures by how much "spirit" they had versus how much matter, putting angels above humans, humans above oysters, and oysters above rocks, which seems intuitively right. Linnaeus used a more worldly criterion – complexity -- to come up with an ordering that gets the angels-humans-oysters-rocks ordering right. But complexity is itself a complex notion. Are we sure that rats are more complex than peacocks and that caterpillars are more complex than willows?
A hidden hand guided Linnaeus, determining the general shape of his scheme. Douglas withdraws a thin pile of paper cards as soft as handkerchiefs from one of the drawers. On each, Linnaeus has recorded in his fine hand the name of one species. If you have one species per card, you do what we’ve seen Mendeleev did with the elements: You play solitaire. You lump and split the cards, putting them near other cards like them. As you do so, you are drawing yet another map of knowledge. Linnaeus named the largest units in his classification kingdoms not because animals, vegetables, and minerals lord it over the creatures within their borders but because kingdoms are the most inclusive territories on political maps; Linnaeus explicitly likened the five levels of his classification system to "Kingdom, province, territory, parish, village" and hoped to see the three kingdoms "depicted in maps or paintings, printed under the title Geographica Naturae." That's why Systema Naturae is oversized: a map makes the most sense when you can see it all at once.
Linnaeus's system not accidentally shares properties with the paper that expresses it: bounded, unchanging, the same for all readers, two-dimensional, and thus only with difficulty able to represent exceptions and complex overlaps, making all visible in a glance, with no dark corners. Linnaeus's organization took the shape it did in part because he constructed it out of paper. Indeed, he classified plants by their stamens and pistils instead of by the more obvious sign, their flowers, in part because he wanted to be able to publish illustrations in black and white. That way the cost of the books would be low enough for the multitudes. Linnaeus wasn't putting away clothing, but because he used paper -- atoms -- to think through the order of the natural world, the organization he came up with repeated the general shape of an orderly household on laundry day.
Copyright © 2007 by David Weinberger. All rights reserved.
Listen to David Weinberger interview the people at the eye of information chaos here.
The Legacy of Linnaeus 
What's in a Name? The Future of Life 
Order Is in the Eye of the Tagger 
Organizing the World in the Age of DNA 
Gallery: Growing the Trees of Life Folksonomies Tap People Power
