A very short introduction to the undeciphered Aegean writing systems

The Phaistos Disc – see below

Every so often a news article will make the rounds of the internet – or, for that matter, a paper will be published in an academic journal – presenting a new ‘decipherment’ of an undeciphered ancient writing system. Obviously, such decipherments have taken place in the past – probably most famously that of Egyptian hieroglyphs – and it’s certainly possible that more will take place in the future; but when it comes to the undeciphered writing systems of the Bronze Age Aegean, at least, there’s good reason to be extremely sceptical about any such claims of decipherment. This post is a quick guide to some key facts about the various related writing systems found in Bronze Age Crete and mainland Greece, starting with the one deciphered writing system, Linear B, and then surveying the undeciphered ones roughly in order of how much we know about them, looking very briefly at where and when they’re from, what kinds of documents exist, and how much (if anything) we know about the writing system or the language it represents.

Linear B

Linear B tablet from Mycenae

Date: 14th-13th centuries BCE.

Location: Crete; mainland Greece.

Number of known documents: c.5,000.

Document types: clay tablets/sealings/labels; some painted inscriptions on vases.

Use: administrative.

Deciphered? Yes, in 1952 by Michael Ventris, based on previous work by scholars including Emmett L. Bennett Jr. and Alice Kober, and published by Ventris and John Chadwick.

Language: Greek, specifically a dialect known as ‘Mycenaean’ which, although older than any of the classical Greek dialects, is not the direct ancestor of any of them.

So, we understand everything about Linear B, then? Unfortunately not! Of the 87 syllabic signs that make up the writing system, 14 are still ‘undeciphered’ (which can mean anything from ‘we have a pretty good idea what sound this sign represents’ to ‘it could be almost anything’; mostly the latter rather than the former. For the very keen, see my PhD thesis). Plus of course there are still many words whose meaning we don’t know, tablets whose contents or context we don’t understand, etc. etc. – plenty of work still to be done!

Linear A

Linear A inscription on stone vase

Date: mid-18th-15th century BCE.

Location: mostly Crete; a few finds elsewhere e.g. Kythera, Thera/Santorini.

Number of known documents: c.1500.

Document types: clay tablets/sealings; stone vases; other clay/stone/metal objects.

Use: administrative; religious.

Deciphered? Short version: no.

Longer version: it’s generally agreed that at least some Linear A signs, and quite plausibly the majority of them, can be ‘read’, since they are likely to have had similar sound-values to their Linear B equivalents (Linear B was adapted directly from Linear A in order to write in Greek); but it’s still not possible to identify the language involved or to understand any of its grammatical features, the meanings of most words, etc. As an example, the word AB81-02, or KU-RO if transliterated using Linear B sound-values, is one of the few words whose meaning we do know: it appears at the end of lists next to the sum of all the listed numerals, and so clearly means ‘total’. But we still don’t actually know how to pronounce this word, or what part of speech it is, and we can’t identify it with any similar words in any known languages.

The most promising set of inscriptions for analysing linguistic features is the so-called ‘libation formula’ – texts found on stone vases used in religious rituals (‘libation tables’), which are probably dedications (so probably say something like “Person X gives/dedicates/offers this object/offering to Deity Y”), and across which similar elements often recur in the same position in the text. In principle, having a ‘formula’ of this kind should let us identify grammatical elements via the slight variations between texts – e.g. if a particular variation in one word seemed to correlate with the number of dedicators listed, we might be able to infer that that was a verb with singular or plural marking. Unfortunately, there simply aren’t enough examples of these texts to establish this kind of linguistic detail – every analysis conducted so far has identified a different element as being the name of the donor, the name of the deity, the verb of offering, etc., so it’s still not possible to draw any certain conclusions from this ‘formula’.

Language: “Minoan”, aka an unknown Cretan language.

So can Linear A be (fully) deciphered? At the moment, no. In the future – we might be able to make more progress if we find a LOT more inscriptions, and in particular more examples of the ‘libation formula’.

Cretan Hieroglyphic

Cretan Hieroglyphic inscription from Knossos

Date: 18th-17th centuries BCE (i.e. overlapping with Linear A, but falling out of use earlier).

Location: central/eastern Crete.

Number of known documents: c.500.

Document types: sealstones; clay tablets/roundels/crescents/etc; vases.

Use: administrative; other?

Deciphered? No. Assigning potential sound-values to Cretan Hieroglyphic signs based on a comparison to Linear B is much more difficult/problematic than for Linear A, since a) relatively few Cretan Hieroglyphic signs can be securely identified as corresponding to particular Linear B signs, and b) Cretan Hieroglyphic is less directly related to Linear B than Linear A is. We’re not even sure if Cretan Hieroglyphic and Linear A represent the same language or two different languages.

Language: Unknown, see above.

Can Cretan Hieroglyphic be deciphered?  No. We have even fewer Cretan Hieroglyphic inscriptions than we do for Linear A, and no really long ones; we don’t even always understand fully whether certain signs are part of an inscription or are providing decoration (or both). Plus, we would really need Linear A to be deciphered first in order to have this, rather than the more distantly related Linear B, as a comparison.

Archanes Script

Archanes seal

Date: late 3rd/early 2nd millennium BCE

Location: Archanes (and perhaps some other locations), Crete.

Number of known documents: >20 (exact number debated).

Document types: seals made of ivory/bone or stone.

Use: administrative? religious?

Deciphered? No.

Language: Unknown.

Can it be deciphered? No. This writing system is very poorly understood, and it’s debated whether it even constitutes a separate writing system or should instead be seen as an early form of Cretan Hieroglyphic or Linear A; scholars don’t even agree on precisely how many inscriptions should be assigned to the ‘Archanes Script’. We would need a significant number of new finds, a much better understanding of what this writing system consists of and how it relates to Cretan Hieroglyphic and Linear A, not to mention a prior decipherment of one or both of those.

Arkalochori axe

Arkalochori axe

Date: 17th/16th century BCE

Location: Arkalochori cave, Crete.

Number of known documents: 1.

Document types: axe.

Use: religious?

Deciphered? No.

Language: Unknown.

Can it be deciphered? No. This writing system is attested on only a single inscription, and like the ‘Archanes Script’ its relationship to the other Cretan writing systems is unclear.

Phaistos Disc

Phaistos Disc

Date: 18th/17th century BCE.

Location: Phaistos, south-central Crete.

Number of known documents: 1.

Document type: Clay disc, with signs stamped in a spiral on both sides.

Use: Unknown.

Deciphered? No.

Language: Unknown.

Is it a fake? Although the Phaistos Disc has been suggested to be a forgery, similarities of some signs to documents in other writing systems (particularly the Arkalochori axe) discovered later than the Disc suggest it is likely to be real.

Can it be deciphered? No. The Disc is a completely unique inscription whose function (and therefore likely content) we don’t know, and although some signs do bear resemblances to other Cretan writing systems, its exact relationship to these (also undeciphered and poorly understood) writing systems is very unclear.

Summary

Apart from Linear B, Linear A is the best-understood Bronze Age Aegean writing system, and although we can to some extent “read” its signs through comparison with Linear B, we still can’t actually read the texts in the sense of understanding the language they’re written in. Without significant further discoveries, we’re extremely unlikely to be able to do so. The other Cretan writing systems are even more poorly attested and understood; when all we have to go on is a handful of inscriptions or even just a single one, whose contents we can’t even make an educated guess at because we don’t know what their purpose was, and whose closest points of comparison are themselves poorly understood and undeciphered…a decipherment is, effectively, impossible. Even if somebody by pure chance managed to guess all the right values for the signs and all the right meanings for the words on, say, the Phaistos Disc, in the absence of other evidence there would be literally no way of proving this to be the correct reading.

That doesn’t mean that we can’t make any progress at all – there’s a lot of really interesting work being done on these writing systems, and by combining analyses of the texts themselves, of the objects they’re written on, and of their archaeological contexts, it’s possible to make significant progress in understanding them even without being able/trying to decipher them. To give a personal example, what I consider to be the most interesting/useful parts of my PhD thesis are not the parts where I talk about the possible values of the undeciphered Linear B signs – most of which there is again simply not enough evidence to get anywhere near deciphering them – but the parts where I assume/accept the fact that these signs aren’t decipherable and instead explore what else can be done with them to understand more about the Linear B writing system and the ways in which it was used by the people who wrote it. Having a realistic view of the prospects, or lack of prospects, of deciphering most of these writing systems without significant further evidence is the best basis for finding other ways of improving our understanding of the inscriptions and their context.

Selected further reading/references

John Chadwick, The Decipherment of Linear B (Cambridge University Press, revised 2nd edn 1992): Google Books preview

John Chadwick, Linear B and Related Scripts (British Museum, 1987)

Anna P. Judson, ‘The decipherment: people, process, challenges’, in Yannis Galanakis, Anastasia Christophilopoulou, and James Grime (eds.), Codebreakers and Groundbreakers (Fitzwilliam Museum, Cambridge, 2017): online here

Philippa Steele and Torsten Meissner, ‘From Linear B to Linear A: the problem of the backward projection of sound values’, in Philippa Steele (ed.), Understanding Relations Between Scripts: The Aegean Writing Systems, 93-110 (Oxbow, 2017): online here

Philippa Steele, ‘Other pre-alphabetic scripts of Crete and Cyprus’, in Yannis Galanakis, Anastasia Christophilopoulou, and James Grime (eds.), Codebreakers and Groundbreakers (Fitzwilliam Museum, Cambridge, 2017), 47-53: on academia.edu

John Younger’s Linear A and Cretan Hieroglyphic websites

Yves Duhoux, ‘Pre-Hellenic Language(s) of Crete’, Journal of Indo-European Studies 26 1-39 (1998)

Yves Duhoux, Thomas Palaima, and John Bennet (eds.), Problems in Decipherment (Peeters, 1989)

Robrecht Decorte, ‘The first European writing: redefining the Archanes Script’, Oxford Journal of Archaeology 37:4, 341-372 (2018)

Georgia Flouda, ‘Materiality and script: constructing a narrative on the Minoan inscribed axe from the Arkalochori cave’, SMEA Nuova Serie 1, 43-56 (2015): on academia.edu

Yves Duhoux, Le disque de Phaestos (Peeters, 1977)

Brent Davis, ‘The Phaistos Disc: A new way of viewing the language behind the script’, Oxford Journal of Archaeology 37:4, 373-410 (2018)

All photos taken by me of objects on display in the National Archaeological Museum, Athens (Linear B tablet) and the Heraklion Archaeological Museum (all others).

Update: I forgot to include Yves Duhoux’s ‘How not to decipher the Phaistos Disc’, American Journal of Archaeology 104:3, 597-600 (2000): online here. Much of what he says is generally applicable to any decipherment attempt!