History becomes possible only when the Word turns into words. Only verbatim traditions enable the historian to reconstruct the past. Only where words that were lost can be found again does the historiographer replace the storyteller. The historians home is on the island of writing. He furnishes its inhabitants with subject matter about the past. The past that can be seized is related to writing. Beyond the island's shores, memories do not become words. Where no words are left behind, the historian finds no foundations for his reconstructions. In the absence of words, artifacts are silent. We have often felt frustrated, but we accept that prehistory cannot be read. No bridge can be constructed to span this chasm.
HISTORY REMAINS a strict discipline only when it stops short, in its description, of the nonverbal past. The critical historian, reading Herodotus or Homer, observes and admires the very creation of Greek words, for the word is a creature of the alphabet and has not always existed. If the historian tries to describe wordless societies, he soon becomes a natural historian, an anthropologist like Aristotle, whose anthroplogein can only be translated as "idle talk"or"tattle tales."Herodotus knew how far the writ of the historiographer ran. A thousand years after the death of Polycrates, he wrote that the tyrant of Samos "was the first to set out to control the sea, apart from Minos of Knossos and possibly others who may have done so as well. Certainly Polycrates was the first of those whom we call the human race."Herodotus did not deny the existence of Minos, but for him Minos was not a human being in the literal sense. He let the architect of the labyrinth live on as the father-in-law of the Minotaur. He believed in gods and myths, but he excluded them from the domain of events that could be described historically. His ability cheerfully to place historical truth alongside the qualitatively different truth of myth stemmed from his having set limits on historiography. He did not see it as his job to decipher a core of describable truths in myth, to explain the sacrifice of Athenian boys to Minos as a tribute to please some lecherous Oriental potentate, as later Greek and Roman historians did. Like Plato, he retained the ability to see the myths as stories that spoke to the illiterate, to children, poets, and old women.
Prior to history, Plato says, there is a narrative that unfolds, not in accordance with the rules of art and knowledge, but out of divine enthusiasm and deep emotion. Corresponding to this prior time is a different truth—namely, myth. In this truly oral culture, before phonetic writing, there can be no words and therefore no text, no original, to which tradition can refer, no subject matter that can be passed on. A new rendering is never just a new version, but always a new song. Thinking itself takes wing; inseparable from speech, it is never there but always gone, like a bird in flight. The storyteller spins his threads, on and on, never repeating himself word for word. No variants can ever be established. This is often overlooked by those who engage in the "reading"of the prehistorical mind, whether their reading is literary, structuralist, or psychoanalytic. They turn Minos into a person, the Minotaur into a dream, and the Labyrinth into a symbol.
Memories of this prehistory become a historical source, a verbatim tradition, only through historiography. Only the historian, writing it down, freezes the source material for his descendants, as Flavius Josephus stresses in his Jewish War: "My task is to write down what I have been told, not to believe everything; and what I am saying here applies to my entire work."Only the original text gives simultaneous rise to source and history.
Every original text is the record of something heard. Some scribe of genius listened to Homer and the result was the one Iliad. Bernardo de Sahagun, the sixteenth-century Franciscan missionary in Mexico, and a pupil of Erasmus, took down hundreds of Aztec songs. He tried to apply the rules of textual criticism to several songs on the same theme all attributed to Prince Netzalhuacoyotl, but failed to reconstruct an original. In their deceptive similarity, each song, when written down, was not a variant but an original. Anthropologists become hunters chasing unwritten materials; tape recorders in hand, they descend on blacks, women, peasants—anyone on whose lips they sense prehistory. Folklorists sieve sagas and legends for fragments of oral phraseology. It is the task of the historian to develop the tools for recognizing which of these records are original sources, that is to say, texts that are not based on other texts, but represent the first fixing of speech. For those records are the flotsam from the oral realm that have washed up on the historiographer's shore, dicta for the first time broken down into words, sung rhythms strung in verses.
Writing is not the only technique we know of for making the flow of speech coagulate and for carrying clots of language along intact for tens or even hundreds of years. When melody, meter, and rhythm combine with a proverb, the result is often an indestructible nugget of language. The drummers of the Lokele who live in the jungle of Zaire, not far from the former Congo River, still know the sayings that fit their tom-tom rhythms. In fact they need the sayings in order to drum the rhythms. But no one now remembers what they mean—or whether they ever "said"anything.
In certain rituals practiced in the Isthmus of Panama, sequences of sounds are used, in which rhythm, melody, and articulation form a three-dimensional counterpoint. The counterpoint effectively prevents any change from creeping in, the chants acting as mummified dicta from a forgotten, prehistoric age. Legal maxims, oaths, spells, benedictions and curses, elements of genealogy, the stock epithets attaching to the name of a god, a hero, or a place, are all very often secured against corruption in this way. The utterance can also be tied to a thing. The tally stick that the Maori orator holds in front of him and to which he hitches his solemn oration, the quipu, or knotted bundle of threads that enables the Incan runner to reel off his news like a rosary, the sequence of pictures drawn on a wall, can support the unchanged repetition of sounds that might make no sense to the speaker. The caste organization of preliterate India can be understood as the social organization of a mnemonic device that enables the Brahmins to preserve the Vedas unchanged. Gestures that coalesce with the liturgical murmurs in a sacrificial ceremony fix language to body movements. Through all these techniques, nuggets of frozen speech can be carried along in an oral culture.
But it would be a grave mistake to view the alphabet primarily as an immense improvement over these mnemonic devices. Only the alphabet has the power to create"language"and"words,"for the word does not emerge until it is written down. Neither the songs of the poets, nor the invocations of the priests, nor the dictates of rulers from prehistoric times are sequences of words. Their immense yet evanescent power eludes description, and those who uttered them were unable, for all their oral skill, to see their own speech as a string on which words are the beads. Prehistory knows nothing of these mono-or polysyllabic atoms of language whose semantic fields we plot with our dictionaries. What prehistory perceives as units can have only audible contours. The sequences of sounds between pauses that characterize speech are not words but syllables, phrases, strophes. It is to these measures of speech alone that the original word or Logos relates. This meaning has become secondary today, although we invoke it when we "give"or"go back on"our word, or when we"have a word"with someone. For us the"real"meaning of word is grammatical building block, before and after which our pen breaks contact with the paper. Plato's slayers, barbarians, and children still live in a prelogical, that is, a"word-less"society.
The historian misreads prehistory when he assumes that"language"can be spoken in that word-less world. In the oral beyond, there is no"content"distinct from the winged word that always rushes by before it has been fully grasped, no"subject matter"that can be conceived of, entrusted to teachers, and acquired by pupils (hence no"Education*,""learning,"and "school"). For it is the record in phonetic writing that first carries what is heard across a chasm separating two heterogenous eras of speech. The alphabetic scribe carries what is spoken from the ever-passing moment and sets down what he has heard in the permanent space of language. Only with this act can knowledge, separate from speech, be born.
As literates, we think of speech as the use of language, and we think of this language as outliving speech, as leaving traces—if not on paper, then in our selves. Before the concept of recording sounds through the alphabet had come into being, speech could not be imagined to leave such a trail. Without a listener (who might be an angel or God), speech could not be perceived as anything but madness, because speech courts attention. And before this sound-recording through the alphabet, a listener could not be perceived as a recorder. The nod indicated that the other person had understood, not that he had recorded the message, accepted the information.
How different speech is from language is made clear by the fact that language is always neuter, while speech is always gendered. With every utterance, the speaker refers back to himself and his gender. It is always the total quality of speech that refers the listener to the speaker's gender, not the grammatical gender of the pronoun"I." (Nowhere, with the possible exception of the oasis of Hadramut, does the personal pronoun have grammatical gender.) In a culture, what sounds feminine and what sounds masculine is determined by convention, and not by the biological nature of the vocal cords. The way men and women speak contrasts in many ways: linguists, anthropologists, and sociologists recognize about two dozen criteria describing these contrasts. In no two places is their configuration the same. The gender contrast in speech is just as fundamental as the contrast in phonemes, but it has barely been remarked. At the very best, recently, linguists have examined the discrimination against women in the so-called "use"of language, which is genderless.
This gender contrast in speech is lost when it congeals as language on the page. It does not survive the jump from pure time of speech into the permanent, spatial dimension of script. To return to Herodotus: The historian's task starts"with those whom we call the human race"that script has brought into being; with men and women when they begin to speak the same language. (We have been tempted to speculate that the story of the Tower of Babel tells of this event.)
If alphabetic writing can be spoken of as bringing the human race into existence, it is only because this kind of writing is unique, as a study of the history and phenomenology of phonetic writing will reveal. Pure, mature phonetic writing, which was discovered only once, albeit in stages, is an oddity among writing systems in the same way that the loudspeaker is an oddity among trumpets. The alphabet records only sounds, and it is only through sounds that it provides meaning. The alphabet does exactly the opposite of what most hieroglyphics and ideograms and, most importantly, what Semitic letters were created to do.
In writing systems using hieroglyphics and ideograms, the reader is expected to speak; the ideogram itself is silent. The statement"1 × 1"says"once one,"or "one times one,"or even "multiplication table."But it can equally be read "jedan put jedan."In all these scripts the reader must find the spoken expression from recollecting what has been said before: Mayan hieroglyphics, for example, provide the clues so that the reader may speak aloud from memory. Through landmarks that are more than just pictograms, they help him find his way orally along an often-traveled path. Ideograms, too, originally point toward utterance. They presuppose that the reader is familiar with the content of ideas whose individual elements are strung in a row before him to be named. Reading thus means retelling the familiar content depicted in accordance with more or less precise rules. Even when—as in the third millennium B.C.—the individual Egyptian hieroglyph or Mesopotamian ideograph become logograms, so that from that point on they had to be named with one and only one word, the word presents itself to the reader without any indication about its sound; the ending and inflection that make it audible must be supplied by the reader.
The early part of the second millennium B.C. saw a series of faltering attempts here and there in the Middle East to bind speech more closely to writing. Convention came to dictate that a particular pictogram or ideogram, which had become a logogram, could be used as a syllable sign. The reader put aside any recognizable meaning of the word and read it into the text for its sound only. As a syllable sign it came to be placed beside the thing sign, making it easier to decipher. Reading became somewhat like solving a rebus. Nowhere, however, did a true syllabary evolve out of this custom—the Indian syllabic alphabet is of considerably more recent origin than the Greek. It is an admirable system of phonetic notation that grew out of the Greek invention.
Quite suddenly, around 1400 B.C., an entirely new kind of script made its appearance on the border between the Egyptian hieroglyphic tradition and the cuneiform of Mesopotamia. This North Semitic alphabet was the first to have signs for sounds only, and only one sign for each group of sounds. Some archaeologists have speculated on a single inventor for this alphabet, so completely does it accomplish both requirements for script from the first moment of its appearance: the universe of heard sounds—an almost infinite variety of sounds in every language, with men and women, children and dotards, singers and ragmen all sounding different—is reduced to a limited number, each of which is then labeled.
However, this Byblos alphabet whose letters stand only for sounds does not have any letters for vowels. The freely voiced qualities of breathing are not indicated, only the consonants, the harsh or soft obstacles the breath encounters. Its script does not yet transform the page into a mirror of speech, but is rather a burial ground for the skeleton of language. Being a purely phonetic notation, it differs radically from all previous scripts, but it can still be read only by someone trained for a special kind of analysis. Only a person who has developed the ability to recognize within the uninterrupted string of consonants groups of two to five elements that act as "roots"can breathe those roots into life. The roots grow into words only when the reader makes them resound according to the semantic function they ought to play in the environment in which they stand.
In a prophetic vision, Ezekiel describes the process: "The hand of the Lord carried me out... in the midst of a valley which was full of bones that, lo, were very dry... and I prophesied as I was commanded, and the bones came together: bone to bone... but there was as yet no breath in them... and the Lord said, 'Breathe upon the slain [literally: Give thy soul, nefesh, to them] that they may live'... and as I did, they stood upon their feet" (Ezek. 37:1-10). It is astounding with what audacity a clutch of pastoral tribes in Canaan claimed the invention as their own. As Exodus relates, Israel overcame "Egypt"intellectually and emotionally with the invention of phonetic writing. The mummies in their tombs are supplanted by roots. No longer is it only priests who can promise the continuation of life after death by deciphering the hieroglyphs. The invention of the Semitic script makes possible a new relationship to the life and death of Osiris.
From now on the written character rescues a sequence of sounds from ephemerality; and living speech is dismembered by the scribe, who as he listens to dictation ponders the speech, examines it for its inaudible roots, determines the (usually) three consonants that compose it, and engraves these into a clay tablet. The letters he has buried tell what roots have been read into the recorded utterance, and these letters can be resurrected at any time alphabetically by the reader.
Greek merchants acquired the string of Semitic consonants from Syrian traders on the coast of Asia Minor. They left the sequence of letters undisturbed, with their shapes recognizable and their names unchanged, but they perverted the use of these letters. While for the Semite beth had a semantic association, because for him it means"house, "for the Greek it is merely the name of a letter that stands for a sound. Four of the Semitic letters were not needed by the Greeks: To the Greek ear they stood for barbaric noises. The Greeks of the eighth century used them to indicate vowels. The consonants are now placed between vocals, the entire word now lies on the page. No more does the reader have to recognize naked bones that must be properly assembled by the eye and then fleshed out only by breathing life into them. The page has become a record of sounds.
Phonetic script could now do the opposite of what the string of consonants had so far done. While the consonants had been used to record units of meaning that the scribe had picked from the flow of speech, the Greeks froze the flow of speech itself onto the page. The scroll had been sounded thus far through an act of interpretation of the letters; alphabetic recording that fixed sound on the page brings an utterance from the past into the present, to which the reader can listen, interpreting what he hears. The Jew searches with his eyes for inaudible roots in order to flesh them out with his breath. The Greek picks the sound from the page and searches for the invisible ideas in the sounds the letters command him to make.
The transformations brought about by Greek literacy are well symbolized by the appearance of Sybil, who replaces her older sister, the Pythia, as the model of the prophetess. Her story is told by Heraclitus, a Pythagorean who, through Cratylus, could claim Plato as a pupil. He was the first to distinguish the consonants (which he divided into the unvoiced aphthonga and the sonant aphona) from the vowels. Plutarch has conserved the passage from Heraclitus in which the Sybil makes her first appearance. In the image of the alphabet, she wrests utterance from its temporal context and turns prophecy into a literary genre:"Sybil, in her mania, makes the oracle of the god ring out a whole millennium, joyless, odorless, and unadorned.... "She spells out the future. For the Sybil first writes her oracle on leaves, then later on tablets. She brings stone slabs to King Tarquinas, who reigned over the Campagne, south of Rome—over Etruscan towns through which the Romans got their alphabet. No one need strain anymore to hear the ominous murmurings of the Delphic Pythia. The menacing future can now be read.