The Code Book: The Science of Secrecy from Ancient Egypt to Quantum Cryptography
Page 23
Table 15 Champollion’s decipherment of and the cartouches of Ptolemaios and Cleopatra from the Bankes obelisk.
Then, on September 14, 1822, Champollion received reliefs from the temple of Abu Simbel, containing cartouches that predated the period of Graeco-Roman domination. The significance of these cartouches was that they were old enough to contain traditional Egyptian names, yet they were still spelled out-clear evidence against the theory that spelling was used only for foreign names. Champollion concentrated on a cartouche containing just four hieroglyphs: . The first two symbols were unknown, but the repeated pair at the end, , were known from the cartouche of Alexander (alksentrs) to both represent the letter s. This meant that the cartouche represented (?-?-s-s). At this point, Champollion brought to bear his vast linguistic knowledge. Although Coptic, the direct descendant of the ancient Egyptian language, had ceased to be a living language in the eleventh century A.D., it still existed in a fossilized form in the liturgy of the Christian Coptic Church. Champollion had learned Coptic as a teenager, and was so fluent that he used it to record entries in his journal. However, until this moment, he had never considered that Coptic might also be the language of hieroglyphics.
Table 16 Champollion’s decipherment of , the cartouche of Alksentrs (Alexander).
Champollion wondered whether the first sign in the cartouche, , might be a semagram representing the sun, i.e., a picture of the sun was the symbol for the word “sun.” Then, in an act of intuitive genius, he assumed the sound value of the semagram to be that of the Coptic word for sun, ra. This gave him the sequence (ra-?-s-s). Only one pharaonic name seemed to fit. Allowing for the irritating omission of vowels, and assuming that the missing letter was m, then surely this had to be the name of Rameses, one of the greatest pharaohs, and one of the most ancient. The spell was broken. Even ancient traditional names were phonetically spelled. Champollion dashed into his brother’s office and proclaimed “Je tiens l’affaire!” (“I’ve got it!”), but once again his intense passion for hieroglyphics got the better of him. He promptly collapsed, and was bedridden for the next five days.
Champollion had demonstrated that the scribes sometimes exploited the rebus principle. In a rebus, still found in children’s puzzles, long words are broken into their phonetic components, which are then represented by semagrams. For example, the word “belief” can be broken down into two syllables, be-lief, which can then be rewritten as bee-leaf. Instead of writing the word alphabetically, it can be represented by the image of a bee followed by the image of a leaf. In the example discovered by Champollion, only the first syllable (ra) is represented by a rebus image, a picture of the sun, while the remainder of the word is spelled more conventionally.
The significance of the sun semagram in the Rameses cartouche is enormous, because it clearly restricts the possibilities for the language spoken by the scribes. For example, the scribes could not have spoken Greek, because this would have meant that the cartouche would be pronounced “helios-meses.” The cartouche makes sense only if the scribes spoke a form of Coptic, because the cartouche would then be pronounced “ra-meses.”
Although this was just one more cartouche, its decipherment clearly demonstrated the four fundamental principles of hieroglyphics. First, the language of the script is at least related to Coptic, and, indeed, examination of other hieroglyphics showed that it was Coptic pure and simple. Second, semagrams are used to represent some words, e.g., the word “sun” is represented by a simple picture of the sun. Third, some long words are built wholly or partly using the rebus principle. Finally, for most of their writing, the ancient scribes relied on using a relatively conventional phonetic alphabet. This final point is the most important one, and Champollion called phonetics the “soul” of hieroglyphics.
Using his deep knowledge of Coptic, Champollion began an unhindered and prolific decipherment of hieroglyphics beyond the cartouches. Within two years he identified phonetic values for the majority of hieroglyphs, and discovered that some of them represented combinations of two or even three consonants. This sometimes gave scribes the option of spelling a word using several simple hieroglyphs or with just a few multiconsonant hieroglyphs.
Champollion sent his initial results in a letter to Monsieur Dacier, the permanent secretary of the French Académie des Inscriptions. Then, in 1824, at the age of thirty-four, Champollion published all his achievements in a book entitled Précis du système hiéroglyphique. For the first time in fourteen centuries it was possible to read the history of the pharaohs, as written by their scribes. For linguists, here was an opportunity to study the evolution of a language and a script across a period of over three thousand years. Hieroglyphics could be understood and traced from the third millennium B.C. through to the fourth century A.D. Furthermore, the evolution of hieroglyphics could be compared to the scripts of hieratic and demotic, which could now also be deciphered.
For several years, politics and envy prevented Champollion’s magnificent achievement from being universally accepted. Thomas Young was a particularly bitter critic. On some occasions Young denied that hieroglyphics could be largely phonetic; at other times he accepted the argument, but complained that he himself had reached this conclusion before Champollion, and that the Frenchman had merely filled in the gaps. Much of Young’s hostility resulted from Champollion’s failure to give him any credit, even though it is likely that Young’s initial breakthrough provided the inspiration for the full decipherment.
In July 1828 Champollion embarked on his first expedition to Egypt, which lasted eighteen months. It was a remarkable opportunity for him to see at firsthand the inscriptions he had previously seen only in drawings or lithographs. Thirty years earlier, Napoleon’s expedition had guessed wildly at the meaning of the hieroglyphs which adorned the temples, but now Champollion could simply read them character by character and reinterpret them correctly. His visit came just in time. Three years later, having written up the notes, drawings and translations from his Egyptian expedition, he suffered a severe stroke. The fainting spells he had suffered throughout his life were perhaps symptomatic of a more serious illness, exacerbated by his obsessive and intense study. He died on March 4, 1832, aged forty-one.
The Mystery of Linear B
In the two centuries since Champollion’s breakthrough, Egyptologists have continued to improve their understanding of the intricacies of hieroglyphics. Their level of comprehension is now so high that scholars are able to unravel encrypted hieroglyphics, which are among the world’s most ancient ciphertexts. Some of the inscriptions to be found on the tombs of the pharaohs were encrypted using a variety of techniques, including the substitution cipher. Sometimes fabricated symbols would be used in place of the established hieroglyph, and on other occasions a phonetically different but visually similar hieroglyph would be used instead of the correct one. For example, the horned asp hieroglyph, which usually represents f, might be used in place of the serpent, which represents z. Usually these encrypted epitaphs were not intended to be unbreakable, but rather they acted as cryptic puzzles to arouse the curiosity of passersby, who would thus be tempted to linger at a tomb rather than moving on.
Having conquered hieroglyphics, archaeologists went on to decipher many other ancient scripts, including the cuneiform texts of Babylon, the Kök-Turki runes of Turkey and the Brahmi alphabet of India. However, the good news for budding Champollions is that there are several outstanding scripts waiting to be solved, such as the Etruscan and Indus scripts (see Appendix I). The great difficulty in deciphering the remaining scripts is that there are no cribs, nothing which allows the codebreaker to prize open the meanings of these ancient texts. With Egyptian hieroglyphics it was the cartouches that acted as cribs, giving Young and Champollion their first taste of the underlying phonetic foundation. Without cribs, the decipherment of an ancient script might seem to be hopeless, yet there is one notable example of a script that was unraveled without the aid of a crib. Linear B, a Cretan script dating back to the Br
onze Age, was deciphered without any helpful clues bequeathed by ancient scribes. It was solved by a combination of logic and inspiration, a potent example of pure cryptanalysis. Indeed, the decipherment of Linear B is generally regarded as the greatest of all archaeological decipherments.
The story of Linear B begins with excavations by Sir Arthur Evans, one of the most eminent archaeologists at the turn of the century. Evans was interested in the period of Greek history described by Homer in his twin epics, the Iliad and the Odyssey. Homer recounts the history of the Trojan War, the Greek victory at Troy and the ensuing exploits of the conquering hero Odysseus, events which supposedly occurred in the twelfth century B.C. Some nineteenth-century scholars had dismissed Homer’s epics as nothing more than legends, but in 1872 the German archaeologist Heinrich Schliemann uncovered the site of Troy itself, close to the western coast of Turkey, and suddenly Homer’s myths became history. Between 1872 and 1900, archaeologists uncovered further evidence to suggest a rich period of pre-Hellenic history, predating the Greek classical age of Pythagoras, Plato and Aristotle by some six hundred years. The pre-Hellenic period lasted from 2800 to 1100 B.C., and it was during the last four centuries of this period that the civilization reached its peak. On the Greek mainland it was centered around Mycenae, where archaeologists uncovered a vast array of artifacts and treasures. However, Sir Arthur Evans had become perplexed by the failure of archaeologists to uncover any form of writing. He could not accept that such a sophisticated society could have been completely illiterate, and became determined to prove that the Mycenaean civilization had some form of writing.
After meeting various Athenian dealers in antiquities, Sir Arthur eventually came across some engraved stones, which were apparently seals dating from the pre-Hellenic era. The signs on the seals seemed to be emblematic rather than genuine writing, similar to the symbolism used in heraldry. Yet this discovery gave him the impetus to continue his quest. The seals were said to originate from the island of Crete, and in particular Knossos, where legend told of the palace of King Minos, the center of an empire that dominated the Aegean. Sir Arthur set out for Crete and began excavating in March 1900. The results were as spectacular as they were rapid. He uncovered the remains of a luxurious palace, riddled with an intricate network of passageways and adorned with frescoes of young men leaping over ferocious bulls. Evans speculated that the sport of bull jumping was somehow linked to the legend of the Minotaur, the bull-headed monster that fed on youths, and he suggested that the complexity of the palace passages had inspired the story of the Minotaur’s labyrinth.
Figure 57 Ancient sites around the Aegean Sea. Having uncovered treasures at Mycenae on mainland Greece, Sir Arthur Evans went in search of inscribed tablets. The first Linear B tablets were discovered on the island of Crete, the center of the Minoan empire.
On March 31, Sir Arthur began unearthing the treasure that he had desired most of all. Initially he discovered a single clay tablet with an inscription, then a few days later a wooden chest full of them, and then stockpiles of written material beyond all his expectations. All these clay tablets had originally been allowed to dry in the sun, rather than being fired, so that they could be recycled simply by adding water. Over the centuries, rain should have dissolved the tablets, and they should have been lost forever. However, it appeared that the palace at Knossos had been destroyed by fire, baking the tablets and helping to preserve them for three thousand years. Their condition was so good that it was still possible to discern the fingerprints of the scribes.
The tablets fell into three categories. The first set of tablets, dating from 2000 to 1650 B.C., consisted merely of drawings, probably semagrams, apparently related to the symbols on the seals that Sir Arthur Evans had bought from dealers in Athens. The second set of tablets, dating from 1750 to 1450 B.C., were inscribed with characters that consisted of simple lines, and hence the script was dubbed Linear A. The third set of tablets, dating from 1450 to 1375 B.C., bore a script which seemed to be a refinement of Linear A, and hence called Linear B. Because most of the tablets were Linear B, and because it was the most recent script, Sir Arthur and other archaeologists believed that Linear B gave them their best chance of decipherment.
Many of the tablets seemed to contain inventories. With so many columns of numerical characters it was relatively easy to work out the counting system, but the phonetic characters were far more puzzling. They looked like a meaningless collection of arbitrary doodles. The historian David Kahn described some of the individual characters as “a Gothic arch enclosing a vertical line, a ladder, a heart with a stem running through it, a bent trident with a barb, a three-legged dinosaur looking behind him, an A with an extra horizontal bar running through it, a backward S, a tall beer glass, half full, with a bow tied on its rim; dozens look like nothing at all.” Only two useful facts could be established about Linear B. First, the direction of the writing was clearly from left to right, as any gap at the end of a line was generally on the right. Second, there were 90 distinct characters, which implied that the writing was almost certainly syllabic. Purely alphabetic scripts tend to have between 20 and 40 characters (Russian, for example, has 36 signs, and Arabic has 28). At the other extreme, scripts that rely on semagrams tend to have hundreds or even thousands of signs (Chinese has over 5,000). Syllabic scripts occupy the middle ground, with between 50 and 100 syllabic characters. Beyond these two facts, Linear B was an unfathomable mystery.
The fundamental problem was that nobody could be sure what language Linear B was written in. Initially, there was speculation that Linear B was a written form of Greek, because seven of the characters bore a close resemblance to characters in the classical Cypriot script, which was known to be a form of Greek script used between 600 and 200 B.C. But doubts began to appear. The most common final consonant in Greek is s, and consequently the commonest final character in the Cypriot script is , which represents the syllable se—because the characters are syllabic, a lone consonant has to be represented by a consonant-vowel combination, the vowel remaining silent. This same character also appears in Linear B, but it is rarely found at the end of a word, indicating that Linear B could not be Greek. The general consensus was that Linear B, the older script, represented an unknown and extinct language. When this language died out, the writing remained and evolved over the centuries into the Cypriot script, which was used to write Greek. Therefore, the two scripts looked similar but expressed totally different languages.
Sir Arthur Evans was a great supporter of the theory that Linear B was not a written form of Greek, and instead believed that it represented a native Cretan language. He was convinced that there was strong archaeological evidence to back up his argument. For example, his discoveries on the island of Crete suggested that the empire of King Minos, known as the Minoan empire, was far more advanced than the Mycenaean civilization on the mainland. The Minoan Empire was not a dominion of the Mycenaean empire, but rather a rival, possibly even the dominant power. The myth of the Minotaur supported this position. The legend described how King Minos would demand that the Athenians send him groups of youths and maidens to be sacrificed to the Minotaur. In short, Evans concluded that the Minoans were so successful that they would have retained their native language, rather than adopting Greek, the language of their rivals.
Figure 58 A Linear B tablet, c. 1400 B.C. (photo credit 5.4)
Although it became widely accepted that the Minoans spoke their own non-Greek language (and Linear B represented this language), there were one or two heretics who argued that the Minoans spoke and wrote Greek. Sir Arthur did not take such dissent lightly, and used his influence to punish those who disagreed with him. When A.J.B. Wace, Professor of Archaeology at the University of Cambridge, spoke in favor of the theory that Linear B represented Greek, Sir Arthur excluded him from all excavations, and forced him to retire from the British School in Athens.
In 1939, the “Greek vs. non-Greek” controversy grew when Carl Blegen of the University of Cincinnati d
iscovered a new batch of Linear B tablets at the palace of Nestor at Pylos. This was extraordinary because Pylos is on the Greek mainland, and would have been part of the Mycenaean Empire, not the Minoan. The minority of archaeologists who believed that Linear B was Greek argued that this favored their hypothesis: Linear B was found on the mainland where they spoke Greek, therefore Linear B represents Greek; Linear B is also found on Crete, so the Minoans also spoke Greek. The Evans camp ran the argument in reverse: the Minoans of Crete spoke the Minoan language; Linear B is found on Crete, therefore Linear B represents the Minoan language; Linear B is also found on the mainland, so they also spoke Minoan on the mainland. Sir Arthur was emphatic: “There is no place at Mycenae for Greek-speaking dynasts … the culture, like the language, was still Minoan to the core.”