Solving the riddle of an ancient script
It was one of the most captivating mysteries of the modern age, requiring three detectives and 52 years to solve. Along the way, there was magnificent obsession, bitter disappointment, world-shaking triumph and swift, unexplained death.
At the centre of the mystery lay a set of clay tablets from the ancient Aegean, inscribed more than 3000 years ago and discovered at the dawn of the 20th century amid the ruins of a lavish Bronze Age palace.
Written by royal scribes, the tablets teemed with writing: tiny pictograms in the shapes of swords, horses' heads, pots and pans, plus a set of far more cryptic characters whose meaning is still debated today.
The tablets were unearthed in 1900 by the great English archaeologist Arthur Evans. Digging at Knossos, Crete, he discovered a sprawling palace comprising grand staircases, once-bubbling fountains and hundreds of rooms linked by a network of twisting passages.
From the ruins, Evans unearthed crumbling murals, exquisitely worked gold jewellery and a massive alabaster throne. But these treasures paled beside the tablets. Set down in wet clay in about 1450BC, they were, when Evans found them, Europe's oldest written records.
Evans named the curious writing Linear Script Class B - Linear B for short. (At Knossos, he would later unearth tablets bearing a somewhat older script, still undeciphered, which he called Linear A.)
Evans had uncovered the first European bureaucracy, and the tablets, he knew, were the palace's account books. If they could be deciphered, they would illuminate a wealthy and literate civilisation that had flourished in the Aegean a full millennium before the glories of classical Athens.
Oxford-educated, vastly wealthy and used to getting his way, Evans vowed to solve the riddle of the script. Yet he was unable to do so - or even to determine what language the tablets recorded - before he died in 1941, at 90. As a result, Linear B gained a reputation as one of the most intractable puzzles in history, a locked-room mystery with almost no possibility of procuring a key.
Then along came Alice Kober. The story of Linear B has long been a British masculine triumphal narrative, bracketed by two remarkable Englishmen: Evans and Michael Ventris, the dashing young amateur who, against all odds, deciphered the script in 1952.
But at the narrative's centre there stands an equally remarkable American woman: Kober, an overworked, underpaid classics professor at Brooklyn College in New York City. For it was she, sitting night after night at her dining table, who hunted down the hidden patterns within the script that would furnish the long-sought key.
Though the full extent of her work remained unknown for decades (Kober's private writings became available only recently), scholars of the decipherment now believe that without her painstaking analysis, Linear B would never have been deciphered when it was, if ever.
The daughter of Hungarian immigrants, Alice Elizabeth Kober was born in Manhattan in 1906. On her graduation from the city's public university system in 1928, where she had taken a course in early Greek life, the 21-year-old announced that she would one day decipher the script. No one believed her, but she very nearly kept her word.
At Evans' death, only a few facts about Linear B had been established. It employed about 85 basic characters. The number of characters revealed Linear B to be a syllabic script, in which each character stood for a distinct syllable of the ancient language, like ma or pa, bo or do, tam or kam. Many of the tablets were inventories, counting everything in the palace storehouses from chariot wheels to cattle.
The Linear B numerical system was easily understood. A base-10 system like ours, it was notated by means of five characters (denoting 1, 10, 100, 1000 and 10,000), which could be used in combination.
The tablets also contained pictograms, which stood for whole words and showed the objects inventoried. Many were understandable.
And yet, when it came to the non-pictographic signs that littered Linear B, which appeared in small strings called "sign-groups", Evans was flummoxed. When he died, the meaning of only a single Linear B word was known. The word was "total", revealed by the fact that it appeared regularly at the bottom of inventories, just before the tally.
All in all, one word was not much to show for 40 years' work.
For the decipherment to advance, as Kober declared in a 1948 lecture, it would be necessary "to develop a science of graphics". It was just such a science that she set out to construct.
Alice Kober never married, nor is there evidence she ever had a romantic partner. Her life was her work, and what a great deal of work there was. Night after night, she sat at the table in the house in Brooklyn she shared with her widowed mother and pored over the strange Cretan inscriptions.
Her first order of business was frequency analysis: the creation of statistics "of the kind so successfully used in the deciphering and decoding of secret messages", as she wrote, for every character of the script.
Anyone who has solved a Sunday paper cryptogram has met frequency analysis head-on. At its simplest, it entails pure counting, with the decipherer tabulating the number of times a particular character appears in a particular text. If the text is long enough, the frequency count for each letter should mirror its statistical frequency in the language as a whole.
Kober compiled statistics on each character, tabulating its incidence at the beginnings of words, the middles, the ends, in combination with every other character and much else.
When she began her work in the early '30s, Kober kept her statistics in a series of notebooks. But during the Second World War and for years afterwards, paper was scarce. Undaunted, she scissored an immense set of "index cards" from anything she could find, including the backs of church circulars and a great many checkout slips she discreetly pinched from the Brooklyn College library.
Her dedication almost defies belief. In a letter to a colleague in 1947, Kober itemised the time it took to compile a single statistic: "You can figure out for yourself how long it will take to compare each of 78 signs with 78 other signs, at 15 minutes (with luck) for each comparison. Let's see, 78 times 77 times 15 minutes - that's about 1500 hours."
Over the years, Kober cut and annotated 180,000 cards, storing them in empty cigarette cartons, the one paper product of which, sadly, she seemed to have no shortage. Even now, to open one of her ersatz file boxes is to be met with the faint whiff of mid-century tobacco.
From the start, Kober approached the decipherment differently than other investigators. To her great disgust, most scholars persisted in looking at the problem through the wrong end of the telescope, seeking first to identify the language the Minoans spoke and only afterwards to unravel the script.
Everyone, or so it seemed, had a theory about what language the tablets recorded. Ventris was convinced it was the lost Etruscan tongue, and clung steadfastly to the idea until weeks before the decipherment.
Others held even stranger notions. "It is possible to prove, quite logically, that the Cretans spoke any language whatever known to have existed at that time - provided only that one disregards the fact that half a dozen other possibilities are equally logical and equally likely," Kober said in a 1948 lecture. "One of my correspondents maintains that they were Celts, on their way to Ireland and England, and another insists that they are related to the Polynesians of the Pacific."
Rather than speculating on the language of the script, or on how to pronounce its symbols, Kober analysed those symbols as abstract objects of pure form. She was willing to inhabit, as she evocatively wrote, a world of "form without meaning" for however long it took.
This let Kober make vital discoveries. Her first was that the language of Linear B was inflected: that is, it relied on word endings, or suffixes, to give its sentences grammar, much as Latin, German or Spanish does.
The discovery was born of her relentless search for patterns. Among those she identified were three-word sets sharing similar suffixes, which Ventris would waggishly name "Kober's triplets".
These "triplets" let Kober pinpoint critical relationships among the characters of Linear B.
Kober next drew up a grid of these related characters. As she knew, if the phonetic values of even a few characters could be determined, the interdependencies she plotted would let the whole grid fill itself in, in a domino reaction. And it was these relationships that let Ventris, after reading her published articles in the late '40s, make a crucial intuitive leap and unlock the mystery of Linear B.
It could so easily have been Kober who solved the 50-year riddle. But on May 16, 1950, Alice Kober died, aged 43. No one knows what she died from, but it seems probable, given her heavy smoking, that she had some form of cancer.
In June 1952, Ventris, just shy of his 30th birthday, solved the riddle of Linear B. Ventris was an architect who had never been to university. But he had a prodigy's gift for languages and an obsession with the tablets that dated to his youth.
Everyone knew that the tablets were the municipal documents of a Bronze Age Cretan kingdom.
What if, Ventris wondered, some of the related words in Kober's "triplets" were actually related forms of Cretan place names - forms analogous to English words such as "Britain/Briton/British"?
He began plugging phonetic values into the triplets. One word in particular reared up seductively. Ventris' analysis suggested that the first character stood for the syllable "ko", the next for "no", and the third for "so". "Ko-no-so" recalled a particular place - and not just any place, but Knossos, the chief city of Cretan antiquity.
Ventris started plugging sound values into other words on the tablets. They too yielded Cretan place names, including "a-mi-ni-so" (Amnisos) and "tu-ri-so" (Tulissos). As predicted, each correct sound value generated new ones in a chain reaction. Soon, the solution massed before his eyes.
On June 1, 1952, Ventris took the microphone at BBC Radio to announce his discovery: Linear B recorded a very early Greek dialect - which was spoken long before Hellenic peoples were known to have existed, 500 years before Homer and seven centuries before the advent of the Greek alphabet.
His great triumph would end in tragedy. Beset by self-doubt as he was invited to speak before the world's greatest learned bodies, Ventris died four years later, at 34, in a swift, strange car crash that some observers believe was suicide.
"I don't like the idea of getting paid for scholarly writing," Alice Kober had said in 1948, two years before her own untimely death. "If I wanted to make money writing, I'd write detective stories."
That, as it turns out, is precisely what she was writing: read today, her work is a forensic playbook for archaeological decipherment. And it is something even more valuable besides: the story of an unsung heroine that can, at last, be properly told.