Words as Things

IS A WORD A THING? It depends, of course, on what is meant by thing. If sensible properties constitute thingness, then a word is certainly a thing. It exists either as a sound to be heard or a mark to be seen. There is a long tradition, however, of denying words the status of things. In the short essay that follows, I will suggest that this tradition begins when words are required to represent things or matter. If words are to give a clear representation of things (empirical or notional), they must forego their own thingness. At the beginning of the seventeenth century, Bacon draws a strong line between words and things. To emphasize the inferiority of words to things, he compares words to three forms of representation:(1) a flourish on the initial letter of a patent or limned book; the statue Pygmalion fell in love with; a painting like Zeuxis's famous still life of grapes that looked so real a blackbird tried to peck them.(2) This mistaking of the unreal for the real is what Bacon terms "Pygmalion's frenzy," a madness like idolatry that fixates on the image rather than the thing the image represents. In all three instances, the forbidden graven image is imagined as being itself immaterial. It offers up nothing of its own to read, to embrace, to eat. But, of course, all these forms of representation do have substance of their own, though it is not the same as that of the thing they represent. The flourish is made of ink on paper, the statue of stone, the painting of canvas and pigment. If these images were granted materiality, they would themselves become things worthy of the desire (to study, love, eat) that is the due of what they represent. Their pursuit then would be impelled not by a mad "frenzy" but by perfectly reasonable interest. If words are to serve as transparent representations of things, their own thinglike or sensible properties must be overlooked. Or else remade in the image of what they represent. Thus Bacon hinted at an alternative system of notation that would work "without the help or intervention of words."(3) Its characters, he speculated, would resemble the things they represented, either physically as pictographs or conceptually as ideographs. In the second half of the century, Bacon's suggestion materialized in the project sponsored by the Royal Society to devise an artificial language. John Wilkins, for example, in his 800-page An Essay Toward a Real Character, and a Philosophical Language (1668) describes a set of characters intended to represent directly the objects or notions common to all men.(4) Each character stands for a thing or an idea, and when properly distributed and combined they are to correspond with empirical observation or philosophical ordering. In these attempts, characters are designed in the likeness of the things they represent; their own material attributes are forged to match what they stand for. Words, it might be said, have been phased into things. Indeed it is not much of a leap to Swift's satire of the Royal Society's linguistic projects. Gulliver stumbles upon an academy whose members have abolished words altogether and communicate solely by brandishing things. If a long conversation is anticipated, huge bundles are assembled; for shorter ones, a few items under the arm will do.(5) The Royal Society was also concerned with improving the use of ordinary language. In his History of the Royal Society (1667), Thomas Sprat commends the Society's commitment to aligning words with things; he reproduces several of its reports, which are exemplary in their description of "so many things almost in an equal number of words."(6) John Wilkins's first consideration is to assign marks or names to all things and notions in order to attain "a just Enumeration." Ordinary language is riddled with two defects that upset this ideal balance: homonyms and synonyms. Either there are not enough words to match things or else too many. In the case of homonyms, one word is used for many things: bill, for example, is used to mean weapon, bird's beak, and written scroll. …