Hey there
Scholar-for-life who enjoys reading books and writing articles (humanities/hist/rel/orient but that extends to lots of other topics such as archeo/art hist/linguistics) is looking for like-minded people on the tech side of life. The idea is to conceive dig hum projects for universities and museums. I’m based in FR and NL and usually prefer to stay low profile but the foss aubergist can give my contact.
I admire people who are genuinely academic.
That is not going to be easy. They used to exist, like in University Computing Science Departments, but the modern Technology practitioners are pragmatic, not academic. , and their writing skills are almost nonexistant.
Good Luck. I can think of a couple of forum members who may be interested
Here’s one thing I’d like to see resolved before I die - decipherment of the Linnear B text from Crete… They deciphered Linnear A, which proved to be Mycenean Greek… It’s assumed Linnear B is the Minoan language…
That should be possible with modern software.
Does anyone know what happened to Archie?? It seems that Utrecht had built a perfectly autonomous environment when it was suddenly stopped on the order of the authorities? Or did I miss anything?
That sounds fascinating! Really cool thread.
Hi Thomas,
Welcome. I hope you find the forum interesting.
When you reply to an existing reply , can you please indicate to whom you are replying… it can be difficult, in a long topic, to trace the threads, and discourse does not do it automatically.
There are several ways
Regards
Neville
@nevj yes sure, sorry about that. Thanks for the tip!
i’m ready for the challenge; can anyone find the texts online and apply a few bots to it?
I think you need some considerable language scholar skills… or would that bias it?
there are certainly some publications online with earlier attemps; we can feed them to a bot and see what comes out. searching in “www.academia.edu” is always the best start
How do we judge success?
Do we even know what the alphabet is with Linear B?.. or is that part of the problem?
By all means have an attempt.
Nobody’s sure - there’s no “Rosetta Stone” for it.
Some of the Linnear A stuff is used as a potential hint at phonetics, but I seem to remember - Linnear A wasn’t 100% phonetic.
Linnear A was archaic Mycenean Greek… During/after the Bronze Age collapse of the Eastern Mediterranean civilzations - and the advent of the Iron Age, the Greeks adopted the Phoenican alphabet (the first two letters of which are aleph and beth - i.e. similar to Hebrew and Aramaic).
That’s where we get the word “phonetic” from - however - Phoenican was a right to left writing system, and NONE of the letters are vowels - so “aleph” isn’t a vowel! In modern Hebrew, and Arabic - “vowels” are implied by the context of the consonants…
Oops! Postscript after writing the above : The Mycenean Bronze Age Greek is “Linear B” - the undeciphered text, probably Minoan, is “Linear A”…
Linear “A” : Minoan Cretan :
Linear “B” : Mycenean Greek :
Note : the Linear A and B script is a “syllibary” - i.e. like Arabic, Aramaic, or Hebrew - each character is a syllable - not a phoenetic character… The decipherer of Linear B - was able use Greek syllables to decipher it - but - those known syllables, are not know for the Minoan language…
Why I’m so interested in this is - I’m curious whether Minoan was an Indo-European language - or an “isolate” like Etruscan, or Basque (BTW - nobody’s truly deciphered Etruscan either - but linguists have determined a number of words in Latin have a non Indo-European etymology - like the word “populi”, i.e. from Etruscan).
Here’s another thing I’d love to see found or uncovered, the 1st Century AD Roman Emperor, Claudius - wrote a long text (effectively a dictonary) on the Etruscan language - but it was lost / destroyed or whatever… I think most linguists of ancient languages, accept that Estruscan was a language “isolate” that predated the arrival of Indo-European languages (like Latin, Greek, Celtic, Germanic)…
Just to clarify… it is Linear A that is undeciphered… you had it switched with B
[quote=“Dan Tripp, post:13, topic:14977, username:daniel.m.tripp”]
So modern languages like English are not syllabic, they are phonetic… each character is a sound, but in English especially the sounds are not consistent. German and Italian are more phonetic than English. A syllable might be several characters in a phonetic language.
eg danger… dan-ger … 2 syllables
Dear me, my Grammar is poorly recollected.
As a non-techie, this is the methodology i would hope to see applied:
1)get all articles on linear a and b from academia.edu
2)feed the bot a model based on deciphered ugaritic and elamite (some of the most recent success stories)
3)compare results
You want to train an AI model using successful deciphering attempts, then ask it to work on linear A.
and , of course you would consult current work on linear A and B.
Hmm. It sounds Like a reasonable overall plan, but I bet there are mountains of fine details to be addreased .
I have never built a modern language based AI model. I built one once for analysing images but it was entirely statistical…no language model. I do know that it requires a lot of comprehensive trainig data. I spent 8 years on that project.
Why are you calling it a ‘bot’ ?
Bots are internet tools that automate things. It is not that.?
And I was wrong - Arabic and Hebrew (and Aramiac) aren’t syllabic - Arabic is classed as an “abjad” - vowels are mostly implied or indicated with diacritics… So - it’s an alphabet with hardly any vowels…