The Riddle of the Labyrinth (11 page)

Each sign-group, Evans knew, was almost certainly a word of the Cretan language.

BY THE END of the second millennium B.C., there was no one left who held the key to Linear B. There is historical evidence that long before Evans dug them up, the tablets had already begun to tantalize. In A.D. 66, during the reign of the Roman emperor Nero, a violent earthquake at Knossos flung to the surface what appeared to be a tin chest. The shepherds who found the chest looked eagerly inside for treasure but discovered, Evans wrote, only “documents of ‘lime-bark,' inscribed with ‘unintelligible letters.'” The documents were brought to the emperor, who, supposing them to be Phoenician, summoned Semitic-language experts to interpret them. The experts were doubtless under pressure—pressure, most likely, of being allowed to continue breathing—to give the emperor the answer he wanted. After studying the tablets, they determined that they were indeed written in Phoenician and represented, as Evans wrote, “the journal of one of the ancients—the Knossian Diktys, companion of Idomeneus, who had been present at the Trojan War.”

In fact, Evans wrote, these ancient “lime-bark” records might well have been clay tablets from the Knossos palace, thrust aboveground by an earthquake more than a millennium after they were written. The Knossos scribes were known to have stored some of their file boxes in hollowed-out stone chests, known as cists, which were lined with lead sheeting to protect the contents. While the wooden file box within would long since have decayed, the lead-walled cist itself might well be mistaken for a tin chest. Furthermore, he wrote, “The brown, half-burnt tablets of the Palace themselves bear a distinct resemblance to old or rotten wood”—hence the reference to “lime-bark” in the Classical sources describing the find.

The curious documents Emperor Nero held in his hands may have been the records of the Palace of Knossos, written in Linear B by court scribes two centuries before the Trojan War. Nero was reported to have ordered the documents translated from “Phoenician” into Greek and to have filed the translation away in his personal library. Then the tablets were forgotten again, for more than eighteen centuries.

I
T TOOK FIRE TO GIVE us Linear B. In about 1400 B.C., the final conflagration at Knossos destroyed most of the palace and its contents, marking the end of the great civilization that had been rooted there for centuries. But the blaze had one completely beneficial effect: It preserved for future generations the clay tablets that recorded the palace's final year.

Cretan scribes never fired their work. In a warm climate like the Aegean there was no need: Inscribed, the damp clay was simply left to dry in the sun, and that was usually enough. Evans discovered this practice to his dismay after he had painstakingly dug up a cluster of tablets from a less fire-damaged area of the palace. Carrying them back to his rented Cretan house for safekeeping, he unwittingly placed them under a rotten spot in the thatched roof. That night the rains came, and he awoke in the morning to find the precious unbaked records reduced to mud.

In other parts of the palace, however, where the flames had burned hotter, tablets were baked to a permanent hardness. “In this way fire—so fatal elsewhere to historic libraries!—has acted as a preservative of these earlier records,” Evans wrote after the first season's dig.

But baking also made the tablets dry and brittle, and whenever Evans unearthed a cache of inscribed clay shards, his first job was to fit them back together again like the pieces of a jigsaw puzzle. Often, fragments of a single tablet were widely separated, scattered by animals and earthquakes. (Parts of some tablets were so far-flung they would not be reunited for decades.) Making the job even harder was the fact that pieces of a single tablet might bake at different temperatures—some wound up closer to the fire's center, others farther away. When this happened, they shrank at different rates. The result was a tablet whose disjointed parts no longer looked as though they had ever fit together.

Once Evans had enough complete tablets to work with, he could start to compare the two linear scripts, A and B. He had unearthed just a few Linear A tablets at Knossos. (As it turned out, they would be the only significant examples of the script ever found there.) But in 1902, a colleague, the Italian archaeologist Federico Halbherr, uncovered a cache of Linear A tablets at Hagia Triada, a site in southern Crete. With these, Evans had a sufficient sample for comparison.

The A and B scripts had in common more than fifty characters that were identical or nearly so. Both scripts had been used by Cretan scribes to record commodities vital to the local economy: grain, oil, livestock, wine, and the like. (There were some additional Linear A inscriptions, found on ceremonial artifacts, that contained what appeared to be religious dedications.) In both scripts, text was written from left to right. Where Linear B almost always indicated word breaks (usually by means of the vertical tick marks), Linear A did so only sometimes, using a variety of devices, like little dots between groups of signs. Unlike the Linear B tablets, Linear A tablets were nearly always unruled, and in general the A script looked cruder. It was quite reasonable to assume, as Evans did, that the B script was a later, more refined outgrowth of the A, and indeed this turned out to be the case.

Before the decipherment of either script could begin, Evans needed to answer an essential question: Did they record the same language? He thought they did—they were too similar-looking for it to be otherwise. “The conclusion has been drawn,” he wrote, “that the language itself was practically identical and that the differences visible in B must be rather due to dynastic than to racial causes.”

Besides sharing individual characters, the two scripts shared a means of writing numbers. In unearthing the Knossos palace, Evans had uncovered the first European bureaucracy, and the tablets, he knew, were the palace's account books. With civilization comes stuff, and with stuff comes the need to keep track of it. Not surprisingly, the palace scribes were great enumerators. The tablets filed neatly away by subject counted everything in the Cretan kingdom from sheep, horses, and swine to footstools, bathtubs, chariot wheels (intact), and chariot wheels (broken). Other tablets contained what appeared to Evans to be census data about the kingdom's human inhabitants—who ran the gamut from monarchs to slaves—including their tax records. As a result, Linear B tablets teemed with numbers.

Evans was able to work out the numerical system quickly. The Cretans used a base-10 system, like our own decimal system. Unlike ours, it was notated by means of only five symbols:

To write the number five, for instance, the Cretan scribe repeated the “1” sign five times:
. To write the number fifty, he drew five “10” signs:
. Fifty-five was written this way:
. And so on, for 555
; 5,555
; and 55,555
. The system went up to 99,999, or
. There was no sign for zero.