The Writing Revolution

by Amalia E Gnanadesikan

  Figure 6.2 The Linear B syllabary. At the top are the signs of the basic syllabary. Also listed are optional symbols that seem to have been used occasionally to avoid certain ambiguities in spelling. At the bottom are symbols which occur so rarely that their value has not yet been established.

  Another of Kober’s triplets, , now read ?-no-so, a close match for Knossos. That made the unknown syllable, , ko. Working in this way with known place names and the string of inferences that could be made whenever a value was added to the grid, Ventris was able to identify the spellings of many places in Crete. This gave him the values of many signs, but the language of the inscriptions was still unknown, as place names are often quite independent of regional language. Deciphering only place names such as Connecticut and Quebec, for example, would not tell you if a coded text was in English, French, or even a Native American language.

  But then Ventris turned to the longer members of Kober’s triplets, to the words for “total,” and to those for “girl” and “boy”. The longer forms of the Amnisos triplet were a-mi-ni-si-jo and a-mi-ni-si-ja, which could be construed as “men from Amnisos” (“male Amnisians”) and “women from Amnisos” (“female Amnisians”) in Greek. “Boy” and “girl” were ko-wo and ko-wa, plausible syllabic spellings of what these words were in early Greek, probably korwos and korw. “Total” was to-so in the masculine, and to-sa in the feminine; again, it looked Greek.

  Could Linear B actually be Greek? Was Evans wrong? Why would the Minoans write in Greek? Claiming that Linear B was Greek was a bold step to take, and Ventris did not take it all at once. At first he only claimed to have found some Greek words in the inscriptions.

  Ventris had good reasons to hesitate. For one thing, the Greek that emerged was very poorly spelled. Greek is not a good language for a CV syllabary. Many phonemes were omitted from spellings, and many distinctions between phonemes were not made. For another thing, the Greek – if it was Greek – of the tablets was four hundred to six hundred years older than the earliest known alphabetic Greek. No language stays the same for that amount of time, so the texts were bound to look odd to a student of Classical Greek. Some archaic words were known from Homer, who lived early in the period of renewed Greek literacy and deliberately used archaic language to describe ancient events, but Homer used only words that fit the meter of his poetry. Other archaic words were lost. It is also true that the tablets probably contain some words that are not Greek: Mycenaeans living in Crete might well have incorporated some native Minoan words into their vocabulary.

  One of the first to believe Ventris’s suggestion of a Greek solution was John Chadwick, who assisted in the final stages of the decipherment. Chadwick’s ready acceptance of the fractured syllabic spelling of Greek may have stemmed from his knowledge of Japanese, which made him familiar with the way a language (or writing system) with a simple syllable structure AD apts words that do not fit the allowed patterns (in Japanese, for example, “Christmas” is kurisumasu). Thanks to his academic training, he was also able to guide Ventris in matters of ancient Greek dialectology. Ventris had been concerned because he had found no trace of a definite article in the tablets – no “the,” in other words. Chadwick was able to reassure him on that score: while Classical Greek had a definite article, historical linguists had already worked out that Greek of earlier periods did not.

  Acceptance of the Greek hypothesis was slow in coming, the spelling system of Linear B being one significant reason for it. Trying to spell Greek in an open CV syllabary is very much a business of fitting a square peg into a round hole. Furthermore, not all the CV syllables of Greek are distinguished. The difference between syllables starting with p- and those starting with b- (a sound which is like p, but is pronounced with the vocal cords vibrating) or ph- ([ph], pronounced like p, but aspirated, and only pronounced f in much later Greek) was not indicated, nor was the difference between the related sounds k- and g- and kh-([kh]), or between t- and th-([th]). The difference between long and short vowels was not indicated, nor was the difference between r and l.

  As a result of these mismatches between the syllabary and the spoken language, words in Linear B look strange to scholars of Greek, even accounting for the antiquity of the language. But there is systematicity, even to the mismatch. Syllable-final consonants (l, m, n, r, s) are regularly omitted in spelling. Thus khalkos “bronze” is spelled ka-ko and patr “father” is spelled pa-te. If more than one consonant begins a syllable, a copy of the upcoming vowel is used to turn the single spoken syllable into two written ones: ti-ri-po spelled tripos, a tripod cauldron. (This spelling convention inspired Knorosov’s principle of “synharmony,” with which he began the phonological decipherment of Maya glyphs.) An initial s- is omitted, however, so that pa-ka-na was the spelling of sphagana, “swords.” The system is regular, but it leaves some significant ambiguities, especially to the modern reader, who is not a native speaker of the language and does not have the clear understanding of a tablet’s context that a Mycenaean clerk would have had.

  Startling confirmation of Venris’s decipherment came in 1953 when Carl Blegen, the excavator of Pylos, began studying some tablets he had found the previous year but had not yet published. Using Ventris’s values, one of the tablets matched spelled words and pictograms too well to be coincidence: the word ti-ri-po (tripos) occurred with an ideogram of a tripod cauldron, . Ti-ri-po-de (tripode, “two tripods”) occurred next to the same tripod ideogram and the numeral 2. There followed a list of vessels, described in syllabic Greek as having three, four, or no handles – with perfectly matching accompanying illustrations. Not only were these Greek words, but the use of the correct archaic dual form (in “two tripods”) showed that the words were being used with Greek grammar, not just sprinkled into another language as loanwords. Linear B was Greek.

  The mystery of Linear B was solved, but the solution created a number of further puzzles. What was Mycenaean Greek doing on Crete? Why did the Mycenaeans take up writing only to abandon it? Where in all this were Evans’s culturally superior Minoans? Presumably theirs was the language recorded with Linear A. At some point their island was conquered by the Mycenaeans, who may have considered it expedient to rule the land from which so many of their imports came. Around 1450 BC, the palaces of Crete were destroyed, leaving only Knossos standing. This destruction marks the Mycenaean invasion. The Mycenaeans were accustomed to borrowing from the Minoans culturally, however, and they left little mark of their residence at Knossos for later archaeologists to uncover. They brought with them their language and their weaponry, but apparently very little else. They AD opted many Minoan ideas, including the technology of syllabic writing.

  The mainland Linear B tablets date from somewhat later, around 1200 BC. The new technology may have taken a while to spread back to the mainland. But it is also true that only destruction by fire preserved the tablets, and so we have the last, but not the first, tablets that were written. At the end of the Mycenaean period, the cities of the Mycenaean world were destroyed, preserving their tablets. Knossos fell first, and later Pylos, Mycenae, and the other Greek city-states of the time. Civilization declined, and around 1100 BC Greece entered its Dark Age.

  With the loss of central administration came the loss of writing. The Greeks had used writing for the same purpose as the early Sumerians had: as a means of bureaucratic record keeping. Over time the Mesopotamians went on to discover other uses of writing – as literature, correspondence, and instructional texts – but the Mycenaean Greeks never took this leap. The second stage of the writing revolution – the expansion of writing from its original record-keeping function – seems never to have happened in the Mycenaean world. Literacy was probably restricted to employees of the palace administration, and may quite literally have died with the sack of the cities that preserved the last of the Linear B tablets for rediscovery in the twentieth century. The idea of writing as a medium of culture apparently never occurred to the early Greeks.

  Perhaps the ambiguities i
nvolved in writing Greek in Linear B helped to persuade them that the potential applications of the technology were limited. They did not, after all, have the VC and CVC syllabograms that the Akkadians developed. The texts on the tablets would have been fairly clear in their administrative context (just as proto-cuneiform was), but the chances for misunderstandings would have been much greater in other contexts. Had they really wanted a better writing system, however, they probably could have created one. Already the script contained some signs that were not simple CV syllables, such as pte, and the diphthongs ai and au. These signs were considered optional, and were used sporadically, apparently to reduce ambiguity (see figure 6.2). More systematic use of such signs would have been a first step toward a more flexible script.

  Whatever the reason, the Bronze-Age Greeks, unlike their classical descendants, were not very interested in the written word and left us only the uninspiring texts of the Linear B tablets. It is possible that a certain amount of writing on perishable materials has been lost. References to “this year” and “last year” in the tablets suggest that a clay tablet was normally kept for only a year, and the clay then softened in water and recycled. Whether the year’s records were first collated in some other form – on some more valuable but perishable material such as papyrus – is unknown. It is telling, however, that the range of preserved uses is so small. The older and more scantily preserved Linear A shows up occasionally on stone or metal objects, recording what may have been religious dedications; but not so Linear B. Not only is there no surviving trace of writing used for anything but bureaucratic purposes – no religious, monumental, or private texts – but there is no writing imported from elsewhere. Cuneiform was used all over the Near East at the time, but not a trace of cuneiform – not a letter, not a treaty, not a receipt – has been found in Crete or Greece, despite obvious trade contacts. The only Egyptian hieroglyphs are inscriptions on scarabs which were apparently brought back as decorative souvenirs – the writing on them is incidental.

  The people who kept their records in Linear B – the last of the Mycenaean Greeks before troubled times in the Mediterranean region destroyed their cities – were the people that Homer describes in the Iliad and the Odyssey. It is natural to hope for some reference to Homeric individuals or events in the tablets, but there is none. Names of people occur, but the name Nestor, king of Pylos, does not occur on the tablets of Pylos, nor is Minos recorded at Knossos, or Agamemnon at Mycenae. Homer calls Nestor a charioteer, and indeed the Pylian tablets include an inventory of chariot wheels (though the inventory of the chariots themselves is still missing). This is as close to Homer’s Nestor as we can get.

  Instead of literature, the Mycenaean Greeks, like the Vikings after them, possessed a rich tradition of oral epic poetry – stories, myths, and legends that told them who they were, where they had come from, and what their place in the world was. These were not private stories to be enjoyed alone, curled up with a good book. They were shared stories that affirmed and nurtured a community.

  This is speculation, to some extent, because an oral tradition leaves no archaeological imprint. Yet when the curtain of history lifted again, and the Greeks again took up writing around 800 BC, the last of the oral poets were still practicing their craft. One of the greatest of these must have been Homer. His work is in the traditional oral style, yet with a polish and magnitude that suggests the influence of writing. Although later tradition describes him as blind, there is no real evidence for this; he may well have recorded the Iliad and the Odyssey himself, as a compilation and refinement of his best work. Like other oral poets, he took traditional material and shaped it to his own purposes. Each telling would have been unique, at least until it was written down.

  Writing set the stories in stone, fixed in the new medium. Was the fixing intentional or unintentional? Possibly it was unintentional, and Homer recorded his stories fully expecting later poets to rewrite them. I like to think, however, that it was intentional, that Homer was widely recognized in his day as a poet of unrivalled ability, and that as he grew older the impending loss of this great poet began to weigh heavily on his community. Was there any way to evade this loss? The newly introduced alphabet served just the right purpose.

  Since Schliemann revealed a historical core to Homer’s epics, it is tempting to look to them to put a face on the inhabitants of the ancient cities, the people who wrote in Linear B. But the oral tradition does not have the same kind of strict accuracy that writing does, and that writing was invented for. Some details may survive the centuries with startling accuracy, but others will be seriously distorted and still others forgotten and replaced by anachronisms. We will probably never fully know which details in Homer fall into which category; the decipherment of Linear B has been only marginally useful in this respect. Neither the epics of Homer nor the Linear B texts give us a clear picture of Mycenaean times.

  Yet it is best not to dismiss individual details too hastily. Had Evans taken Homer more seriously, he might have been more willing to accept the presence of Greeks on Crete, and he might have believed his own reading of polos, garnering for himself the achievement of deciphering Linear B. On Crete, Homer says in the Odyssey, “First come the Achaeans [Greeks], then the native Cretans... Central to all their cities is magnificent Cnossos.” Greeks at Knossos should have been no surprise.

  We owe the works of Homer to renewed Greek literacy, as it was only with the reintroduction of writing into Greece that Homer’s genius was preserved for posterity. On the other hand, Homer’s epics were born of the great Greek oral tradition, whose roots stretch back to the Mycenaean age. We owe them, therefore, to Greek illiteracy, to the value the early Greeks placed on the spoken word to the neglect of the written. We owe them, in other words, to the death of Linear B.

  7

  Japanese: Three Scripts are Better than One

  In Japan, as elsewhere, the history of writing is marked by both invention and staunch conservatism. Yet the Japanese have managed a unique balance between the two, on the one hand creating for themselves not just one but two syllabaries, while on the other hand continuing throughout their literary history to favor the Chinese logograms with which they first learned to write. The resulting syncretism of three scripts used simultaneously qualifies as the most complex writing system in modern use.

  According to Japanese tradition, Chinese characters came to Japan with the arrival of Wani, a Korean scholar, at the court of Emperor jin. Wani brought with him the Analects of Confucius and became tutor to one of jin’s sons. It is not clear how accurate the legend is, or when exactly jin reigned, but somewhere between the late third century and the early fifth century AD, the Chinese written language came to Japan.

  The Japanese court’s first reaction to the new technology was not to adapt the system to Japanese, but rather to learn Chinese, just as the Koreans had before them and as the Akkadians had once learned Sumerian. Such foreign language study is in fact a common response to the introduction of writing – a response justified not only by the difficulty of adapting an existing writing system to a very different language, but also by the fact that one of the important uses of written language is not actually writing, but reading. There was as yet nothing to read in Japanese. The Chinese, by contrast, had been literate for well over a thousand years and had produced great works such as the Confucian classics and the translations of and commentaries on the Buddhist scriptures.

  By the seventh century the Japanese had begun to write for themselves. At first the only way they could do so was to write in Chinese, using Chinese characters arranged in Chinese syntax – in other words, with the words arranged in the order found in Classical Chinese. In a Chinese sentence, the verb is placed between the subject and the object, as it is in English. In Japanese, however, the verb always comes at the end, after the subject and the object. In this respect and in many others, Japanese is very different from Chinese. In fact, despite numerous worthy efforts, Japanese has not convincingly been shown to
be genetically related to any other language, though Korean is generally considered the best candidate. Unlike Chinese, Japanese is rife with inflections and follows each noun with a particle that indicates its function in the sentence. The Chinese writing system contains no equivalents for these morphemes, and so Chinese characters could not straightforwardly be adapted to Japanese. Despite the fact that few Japanese could speak Chinese, writing in Classical Chinese became the official, educated style of writing prose and remained so for centuries.

  It soon became clear, however, that not everything that Japanese writers wanted to say could be written in Chinese. For one thing, how should they write their names? Names are more than their meanings: they are inextricably bound to their pronunciations, which in this case were in Japanese. The written form of a name therefore had to reflect its pronunciation. There was a precedent for a solution to this problem, in the way the Chinese wrote foreign names or the untranslatable Sanskrit terms of Buddhism. They used characters for their phono-logical value, divorced from their meaning. This rebus-style use of characters always remained marginal in Chinese, but it was to become widespread in Japan and lead eventually to the development of its two native syllabaries.