History without a readable voice
For most ancient societies, historians can lean on texts in a known language. Bronze Age Crete is different. The islanders did write, and thousands of signs on clay survive, but the main script of the palace period is still unread. As a result, the island’s story has to be told largely through objects and buildings, with the scripts offering clues rather than a narrative. This is a long way from the situation of Egypt, where the Egyptian hieroglyphic script became readable after the Rosetta Stone helped scholars crack it.
Three scripts, one island
Arthur Evans, who began digging at Knossos in 1900, recognised that Crete had used more than one writing system and gave the scripts the names still in use.
- Cretan hieroglyphs. Pictorial signs, found on seals and clay documents, from roughly 2000 to 1700 BCE. They are few in number and poorly understood.
- Linear A. A script of simple line-drawn signs, used from roughly 1800 to 1450 BCE, mainly for administrative records on clay tablets, though it also appears on pottery, stone offering tables and metal objects.
- Linear B. A related script that appears at Knossos and on the mainland, and records an early form of Greek.
Linear A is the most important for understanding Minoan Crete, and Linear B is the key to reading it, at least in part.
What Linear A allows us to read
Researchers have not deciphered Linear A, meaning that the language it records has not been identified and the texts cannot be translated. Even so, some aspects are understood. The number system is clear: it is decimal, with different signs for units, tens, hundreds and so on, and fractions. Some logograms for goods such as wine, grain and figs can be recognised from context. Because many Linear A signs resemble Linear B signs, scholars have tentatively transferred sound values from Linear B to Linear A. This allows signs to be pronounced, but it does not reveal meaning, because the underlying language is probably not Greek.
A few words appear frequently in the tablets, including one often read as “ku-ro” that seems to mark a total, and it appears with lists of entries that add up. That kind of internal evidence shows the tablets are accounts. Beyond this, little can be said with confidence about vocabulary, grammar or the names of people and gods.
Why it has not been solved
Several obstacles stand in the way:
- The surviving corpus is small, with a limited number of short texts.
- There is no bilingual inscription pairing Linear A with a known language to anchor it.
- The language has no known relatives. Proposals linking it to Anatolian, Semitic or other families have been made, but none has gained general acceptance.
- Some inscriptions are on objects with ritual functions, whose formulas may be hard to parse.
The best-known single object is the Phaistos Disc, a clay disc stamped with signs on both sides, found in the palace of Phaistos and usually dated to about the seventeenth century BCE. It is unique, its signs do not match Linear A, and there is no agreed reading. Claims to have deciphered it have appeared repeatedly and should be treated with caution, as one inscription is not enough to test a claim.
Linear B and a new ruling language
Linear B was deciphered in 1952 by Michael Ventris, who showed it to be Greek, in a story that rested on earlier work by others, notably Alice Kober. The result was surprising, because Evans had believed the Knossos tablets recorded the Minoan language. Instead, Knossos in its last palace phase was run by Greek speakers.
The date of the Knossos Linear B tablets is itself debated. The traditional view places them around 1450 to 1400 BCE, while some scholars have argued for a later date near 1200 BCE. The question matters because it affects how we understand the final centuries of the Cretan palaces. On the mainland, Linear B tablets from Pylos, Mycenae and Thebes belong to the thirteenth century BCE.
The palaces’ collapse
Around 1450 BCE, many Cretan sites show destruction by fire, and Linear A seems to disappear. Knossos survives, now using Linear B. Possible explanations include the long-term effects of the Thera eruption, earthquakes, internal conflict and Mycenaean takeover. The change of script and language at Knossos is the strongest sign that people from the Greek mainland had taken a leading role. Knossos itself suffered destruction later, with dates ranging from about 1375 to 1200 BCE depending on the interpretation of the evidence.
After the palaces, Crete did not become empty. People moved to defensible sites in the hills, and Cretan communities continued into the Iron Age. Our knowledge of the later centuries is thinner.
A last puzzle: Eteocretan
In the first millennium BCE, a few inscriptions from eastern Crete, at places such as Praisos, are written in the Greek alphabet but in a language that is not Greek. The language is called Eteocretan, meaning “true Cretan”, and some wonder whether it descended from Minoan. Because the texts are brief and the connection is unproven, this is a possibility and nothing more. Still, it is a reminder that Crete had speakers of a non-Greek language long after the palaces fell.
What the silence means
The undeciphered script is the main reason that questions about Minoan politics, religion and identity remain open. It is also a reminder of the limits of archaeology. A script that records only inventories may never tell us what people believed. Even a successful decipherment of Linear A might reveal mostly accounts. Until then, the history of Crete in the Bronze Age remains a story written in objects, with only a few words we can read.