Some years ago, I acquired a seven-volume set of the 1907 Grand Larousse French dictionary. The set is pretty tattered by now, but it's still glorious, with its dark red covers embossed with trees and gold letters, its art nouveau frontispieces and letter pages, and above all its intricate engravings and maps and its resplendent color plates -- plates of animals, birds and insects; plates of costumes and furniture; plates of eggs, locomotives, and fencing positions; and lots and lots of plates of military uniforms.
That's the same dictionary that Jean-Paul Sartre recounts reading as a child in his grandfather's study in Alsace, in his autobiographical novel Les Mots:
The Grand Larousse was everything to me; I would take down a volume at random, behind the desk, on the next-to-last shelf. A-bello, belloc-Ch, or Ci-D. . . (these associations of syllables had become proper names that denoted the sectors of universal knowledge: there was the Ci-D region, the Pr-Z region, with their flora and fauna, their cities, their great men and their battles). ... Men and beasts were there in person -- the engravings were their bodies, the text was their souls, their unique essences.1
A child is less likely to have that experience of the dictionary nowadays. Whatever their virtues, most of our dictionaries aren't books to take us lands away. They may include a handful of color plates, but they're basically textual affairs -- the illustrations mostly serve mostly to break the unrelieved monotony of the columns of type. Those dictionaries do fine on souls, but there's not a lot to satisfy our hunger for bodies.
Still, even those dictionaries have their visual charms, and it was nice to see a piece in the Sunday New York Times Magazine the other day on Jeffrey Middleton, who's the illusrator of the new eleventh edition of Merriam-Webster's Collegiate. Middleton is responsible for the elegant little pen-and-ink drawings inserted every couple of pages or so alongside the entries, in a style that hasn't much changed in the past 75 years.
Dictionary traditionalists argue that those drawings do a better job of rendering the idea of a word than elaborate plates or photographs. As the American lexicographer Sidney Landau once put it, ""Photographs are necessarily of unidealized individual things, whether zebras, geese, or medieval churches [whereas] drawings may represent a composite distillation."2
That says as much about the popular conception of photography as it does about dictionaries. It's a fair bet that Pierre Larousse would have used photographs in his grand dictionary if he'd had the technology to print them properly. In that age, people didn't have any problem thinking of photographs as the representations of abstract ideas or imaginary settings. Victorian photographers like Oscar Rejlander and Henry Peach Robinson produced staged¬allegories and dramatic scenes with titles like "Youth and Age" or"The Two Ways of Life." Julia Cameron did photographic illustrations for Tennyson's "Idylls of the King," and Henry James allowed the use of photographs to illustrate a 1909 edition of The Golden Bowl.
And Darwin's cousin Francis Galton made composite photographs aimed at isolating the physiognomic traits of various types and classes -- criminals, Jews (see photo), and the members of the Academy of Sciences.3
Abstraction and displacement were the accepted goals of photography -- in the words of the critic Charles Caffin, a collaborator of Alfred Stieglitz: "the artist [photographer] must make some abstract quality the prime feature of his picture."
It wasn't until the the early 20th century that people began to think of photography as a pure record of the concrete facts before the lens. For Paul Strand, Andre Kertesz, and Henri Cartier-Bresson, there could be no photographs but of things. As Strand once said, "[T]he camera machine cannot evade the objects that are in front of it."
Within a few decades, the notion of photographing abstract ideas brought to mind the kitschy propaganda photos that the Nazis and Soviets made, with titles like "Sunday Volunteers" or "The Next Generation." And modernists might allow you to make a film of a novel, but you couldn't use stills from it as illustrations when the book was reissued.
So it's understandable that a lot of modern lexicographers should believe that a photograph can only represent a particular badger, not the idea of badgerhood. And even today no American dictionary uses photographs to illustrate the meanings of words apart from the American Heritage. (I should say that I'm associated with that dictionary, though they keep me pretty far from the art department end of things.)
Illustrations for rampant, skunk, and other words from Merriam-Webster's Eleventh Collegiate
Illustrations for espadrille, brocade, and brioche from the
American Heritage Fourth Edition
Yet contemporary photographers put those austere modernist scruples behind them some time ago. Fabulists like Joel Peter Witkin, surrealists like Sandy Skoglund, narrative photographers like Tina Barney or Sam Taylor-Wood -- all of them take the object in front of the lens as standing in for something other than itself. I like to imagine some avant-garde Grand Larousse of the future that's illustrated with modern photographs, like Cindy Sherman's depictions of herself in the guise of a housewife or a Raphael, or the photographs that Laurie Simmons calls "Food," "Clothing," and "Shelter," where figurines of leggy women wear hot dogs, gloves, and houses on their upper bodies like the dancing cigarette packs in the Old Gold commercials of the fifties. That may take a while, though.4
But the problem with the way most modern dictionaries are illustrated isn't what they imply about photographs, but what they imply about words. They leave you with the impression that words are colorless abstractions at an eternal remove from the concrete realm of the senses. You think of the angels in Wim Wender's "Wings of Desire," who float in black and white above the streets of Berlin and never make contact with sensory experience. Yet when you hear words like bagel or bullfrog, what comes to mind isn't a sketchy silhouette. Meanings are part of the world, too, and they have color and texture just like everything else. You can even take pictures of them.5
Notes
1. Mais le Grand Larousse me tenait lieu de tout: j'en prenais un tome au hasard, derriˆ¬re le bureau, sur l'avant-dernier rayon, A-Bello, Belloc-Ch ou Ci-D, Mele-Po ou Pr-Z (ces associations de syllabes ˆ¬taient devenues des noms propres qui dˆ¬signaient les secteurs du savoir universel...) ; je les dˆ¬posais pˆ¬niblement sur le sous-main de mon grand-pˆ¬re, j'y dˆ¬nichais les vrais oiseaux, j'y faisais la chasse aux vrais papillons posˆ¬s sur de vraies fleurs. Hommes et bˆ�tes ˆ¬taient lˆÝ, en personne : les gravures, c'ˆ¬taient leurs corps, le texte, c'ˆ¬tait leur ˆ¬me, leur essence singuliˆ¬re ; hors les murs, on rencontrait de vagues ˆ¬bauches qui s'approchaient plus ou moins des archˆ¬types sans atteindre leur perfection : au Jardin d'Acclimatation, les singes ˆ¬taient moins singes, au Jardin du Luxembourg, les hommes ˆ¬taient moins hommes. J.-P. Sartre, Les Mots, Gallimard, 1964, p. 16. return
2. Dictionaries: The Art and Craft of Lexicography, Cambridge University Press, 1984. return
3. Galton himself regarded the exercise as a partial failure, since his composites of criminal faces didn't convey the malevolence of any their individual constituents. As Sandra Phillips observes, "Since [Galton's composites] are abstractions, they now appear uncannily beautiful." return
4 The notion of an "avant-garde dictionary" strikes most people as a contradiction in terms -- if you think of dictionaries as the custodians of tradition, you're likely to expect them to look a little musty, too. Even the Grand Larousse was well behind the fin-de-siecle curve. At the same time the engravers in the Larousse offices in the Boulevard de Montparnasse were copying the works of Corot and Courbet for the dictionary's illustrations, Picasso and Brancusi were working in studios just a few meters away. But dictionaries do catch up, not just with the language, but with the changing look of the world. return
5. At the risk of sullying this point with technical arguments, let me add a note for semanticists only. In fact, there is nothing metaphorical or figurative about saying that we can photograph the meanings of common nouns. Two points are relevant here. First, the particular subjects of photographs can be taken as standing in for kinds or other objects by a process analogous to what Quine calls deferred ostension (see my article "Indexicality and Deixis" for more on this, in addition to Nelson Goodman's discussion of "sampling" in Languages of Art). For example, we can point at a particular bird and say "That is almost extinct," where we intend to refer not to the bird itself, but to the kind it exemplifies. Analogously, a photograph of a particular bird simply IS a photograph of its type, if visual culture will only permit it to be.
It might be objected that the types for which photographs stand in aren't the "meanings" of the expressions, this on the assumption that lexical meanings are invariably Fregean senses. But in fact the majority of noun definitions are not of the type <e, t> -- that is, predicates -- but rather of the type e -- or more specifically, of kind-denoting expressions. For example, Merriam-Webster defines sherry as "A Spanish fortified wine with a distinctly nutty flavor." That doesn't entail that sherry applies to whatever fortified wine is Spanish and has a distinctive nutty flavor, nor¬ does it entail that there could not be a variety of sherry that didn't have a distinctively nutty flavor; rather the NP denotes a particular kind of wine that typically or characteristically has those properties, without implying that every nutty-tasting Spanish fortified wine must count as sherry. Note that the indefinite makes this clear, since wine is treated as a count noun only when it denotes a kind; if the definition read simply "Spanish fortified wine with a distinctly nutty flavor," we'd take it as expressing a predicate; i.e., as being of type <e, t>. (The point is also easy to see when we look at the way quantifiers are used in noun definitions; when the American Heritage defines skunk as "any of several small, mostly carnivorous New World mammals of the genus Mephitis and related genera...," we assume that quantification is over kinds, not that there are several particular skunks in the world.) return
