Guillermo Carvajal writes for La Brújula Verde:
A team of archaeologists has discovered in Knossos, on the Greek island of Crete, the longest Linear A inscription found to date. The script appears on a circular ivory object with an attached handle, discovered in a context of clear religious significance within a Neopalatial building. Besides providing the longest inscription in this yet-to-be-deciphered system, the find offers new perspectives on the use of Minoan writing in ceremonial contexts.
The object was uncovered in one of the rooms of a Neopalatial-phase building (1700-1450 BCE) located in the Anetaki plot, an area of Knossos characterized by its religious use from the Protopalatial period through later times.
The inscription was found in the so-called Ivory Deposit, a semi-subterranean structure where fragments of pottery, bronze, and other offering-related items were also discovered. The ivory object, provisionally identified as a scepter or ritual staff, features an engraved inscription in the Linear A script on all its surfaces. According to specialists who have studied the find, the total length of the text reaches approximately 119 signs, surpassing in length all other known inscriptions in this writing system. […]
The text on the ring is distributed across four faces, each containing inscriptions carved with an unusually high degree of precision for Linear A writing. It is distinguished by its calligraphic refinement, which suggests a ceremonial rather than administrative use, unlike most clay tablets inscribed in this system. […]
A particularly significant aspect is the absence of numerals, a feature that distinguishes this inscription from administrative texts on clay tablets. This reinforces the hypothesis that Minoan writing was not limited to accounting but also played a role in religious and ceremonial contexts.
The paper is Kanta, A., Nakassis, D., Palaima, T. G., & Perna, M. (2025), “An archaeological and epigraphical overview of some inscriptions found in the Cult Center of the city of Knossos (Anetaki plot)” (Ariadne, 27–43, doi.org/10.26248/ariadne.vi.1841). Thanks, Trevor!
I wasn’t sure what “brújula” (in general, when not part of a proper name) might mean in Spanish, and when I looked it up I was cheered by the last sentence of this etymology: ‘From Italian bùssola (compare French boussole), from Vulgar Latin *buxula, a diminutive form of Late Latin buxis (“box”), from Ancient Greek πυξίς (puxís, “boxwood box, box”), from πύξος (púxos, “boxwood”). Apparently influenced by Spanish bruja (“witch”).’
We discussed ‘compass’ back in 2007, and I mentioned the Vulgar Latin word and its offspring here — but not, sadly, the Spanish word.
Unfortunately, the paper says very little; it refers to a forthcoming paper. However, it says that the longest known inscription so far has 105 signs, of which 40% are numerals and separation signs, which are practically absent in the new one (the sign for 10 does show up once, at least). In other words, the new one is a text, not a spreadsheet as usual. That’s going to help.
According to that old thread, the specific Russian word derived (maybe via French) from that Vulg. Latin one was used only to mean “surveyor’s compass,” with different and etymologically unrelated Russian words referring other things all called “compass” in English. I wasn’t immediately sure why surveyors would use different compasses than anyone else for figuring out which way was north etc. but wikipedia advises me that “surveyor’s compass” in English refers or at least use to refer to a gizmo also known by the intriguing name of “circumferentor.” Which you don’t hear about very much these days because it was superseded in the 19th century by the improved gizmo known as the theodolite, which is a word I am vaguely aware of although kind of fuzzy on the details beyond “um, a sort of gizmo that surveyors use.” I assume the Spanish word has a wide enough semantic scope to cover anything that tells you which way north is? (And/or what any particular direction from where you are “is” as quantified in azimuth terms.)
I assume the Spanish word has a wide enough semantic scope to cover anything that tells you which way north is?
Not only that. There is a magnetic and calamitous synonym (the RAE up to its old tricks again). calamita or caramida from Greek via Arabic. A magnet affixed to a length of reed floating in water.
Too easy to confuse the compass “calamita” with the other “calamita” apparently meaning “small, green [female] toad with a yellow line along her back.”
One of the longest Gaulish inscriptions and then the longest in Linear A, in the course of a few months. That’s pretty amazing!
I was surprised to learn that an inscription of 119 signs would be set a new record. That’s a tough corpus to make progress on.
the other “calamita” apparently meaning “small, green [female] toad with a yellow line along her back.”
What do you mean, “other” ? They all line up facing north when they’re in heat.
“the specific Russian word”
And I guess Grin’s Ассоль
I love that novel!
I was surprised to learn that an inscription of 119 signs would be set a new record. That’s a tough corpus to make progress on.
Yes, but it’s a little less tough now. And the very specific context may be helpful.
Epidalea calamita is the natterjack toad.
Hence German Büchse “small box; (obsolete) gun”. The tree is Buchsbaum, so the ü does not come directly from the original y.
@LH, thanks! That’s informative: I read it when I was 7, forgot everything and did not know if it is good:)
“My brother went to Knossos and all I got was this stupid circular ivory object.”
Here is the scholarly journal that Brujula Verde based their article on:
https://ejournals.lib.uoc.gr/Ariadne/article/view/1841/1751
Though they do report 119 signs, they break it down as 84 “completely or partially preserved” and 35 “in small traces or probably present.” Less promising than it seemed. I’m not going to be able to digest this tonight, but a quick partial read also suggests that the text is more similar to previous Linear A texts than I expected from the magazine article. It may have a religious context, but it may still be a sort of listing of or mnemonic for the objects used in a ceremony, rather than a narrative sentence or something like that.
There is also a second, much shorter inscription on the handle of the object. This text does include numbers and seems to match more closely the common run of past Linear A finds.
I missed the Gaulish inscription news. Where is this discussed?
One place is here starting on March 10th. I had previously raised it on a thread here back in January but haven’t re-discovered that discussion yet.
Another aspect of the text on this scepter is that at least 9 of the signs are single syllabograms ligatured to ideograms for vases, which the paper’s authors interpret as helping characterize the vases “on the acrophonic principle.” It’s tough to be sure given the schematic way they’ve written this up, but it’s also possible that the ideograms are included among the 84 legible signs.
Every new inscription is a bit more data. But this inscription may not offer nearly as much information as initially hoped.
The recent date of the article seems to be real, but this scepter was apparently found sometime during excavations that ended in 2017, and the contents of the article were presented at a conference in 2021. It’s possible that specialists have had more details since I found a dissertation that seems to show familiarity with a draft of the paper. Still, that seems like a long time to sit on such a find without providing your interpretation of the text or decent photos in which others could try to puzzle out the signs.
From what I’ve seen, 4 years from excavation to conference and 4 years from conference to publication is certainly not lightning-fast, but neither is it a glacial pace for inscriptions.
But I think I have solved it:
There is also a second, much shorter inscription on the handle of the object. This text does include numbers
A short list with numbers and a longer description on an object found in the EU? It must be a list of ingredients and a health warning (“This scepter is not fit for consumption. Do not insert in body cavities. Keep away from children and pets”.)
I’m eager to see what knowledgeable people make of this:
Linear A deciphered
He gives one word, supposedly Semitic, and… that’s it. No pre-publication, no texts, just an announcement of “I deciphered it.” There’s a reference to a nine-page manuscript, Ya Diktu: Grammar of the Minoan Peak Sanctuary Libation Formula, which is nowhere I can find.
The supposed Semitic root is a little problematic, but we’ll let that slide.
His LinkedIn announcement:
(I’ve seen people get that excited about having deciphered Linear A/Phaistos/Rongorongo/Harappan, which they didn’t.)
Everything Y said, except that the Semitic root is not just “a little problematic,” but pretty much impossible. He identifies “*301-WA-JA” as a triconsonantal root, which doesn’t work, since it alternates with “*301-WA-E” in some inscriptions.
Sounds like just another crank. (Also, is “Shalam” some sort of ecumenical mix of Shalom and Salam?)
Well, the long -o- in shalom was originally long -a-; perhaps this is what he imagines the equivalent of shalom/salm sounded like in his Minoan Semitic.
Cyrus Gordon proclaimed a long time ago that he had demonstrated that Linear A was Semitic, magisterially declaring that, now he had done the hard part of the decipherment, others should fill in the details.
[Gordon, IIRC, reckoned that kuro “total” (probably) must represent Semitic kullu, so that proved it. Better reasoning than imaginary roots, at any rate.]
Let’s accept for the sake of argument that kuro = kullu, wouldn’t that indicate that even if Linear A was used to write a Semitic language, these Semitic speakers were not the inventors of the script? If they were, one would have expected them to distinguish /r/ and /l/. (Ok, one could also assume that these phonemes had merged in their variety of Semitic.)
Anyway, no reason to draw any conclusions until we have an actual decipherment.
At the great big conference I submit an abstract to every year, it’s explicitly forbidden to put into the abstract what you’re going to do, let alone what might result from analyses you haven’t done yet.
Also, I recently learned LinkedIn appears to have a reputation.
From his *itter account (@IdaeanDaktyl), slightly edited:
(All Hebrew misspellings and bad glosses are sic, sic, sic.)
I think we’re well into Hebrew is Greek territory here. (Vol II: And Greek is Hebrew.)
[Separately, I’m trying to think of an ethnonym that genuinely has the literal meaning “brothers”, but drawing a blank.]
@Y: As I was reading that, I found myself wondering, What about the obviously non-Semitic ΔANAOI?. But then he addresses that question, in Sidney Harris style.
https://www.jstor.org/stable/40466876
Toth is reborn!
I’m reminded of D.E.’s comment in the thread that discussed Gaulish, that most ancient tribal names in Celtic languages aren’t understood today, not even some that got quite significant discussion from their contemporaries, and despite the breadth of IE languages that might help color them in. It would be surprising if we could confidently resurrect the meanings of multiple Minoan-era terms for ethnic groupings.
Anyway, I got to Di Mino via a long diversion into Mycenaean facts and factoids. One of the links led me to the etymology of basileos — a Linear B word that is sometimes transliterated qa-si-re-u, and is thought to have been pronounced *gʷasileus. And that *gʷ > b is apparently regular, perhaps only before a.
Routinely here and elsewhere, when the p and q Celtic divide is brought up, someone will mention that the Italic branch seems to have gone through a similar development, and/or that the shift may be more reversible than one might naively expect. But I don’t remember anyone mentioning that Greek had a similar shift.
I get that such changes aren’t completely uncommon or unexpected. It still seems surprising across two related branches and a third neighboring but thought to be relatively unrelated. Is there any theory that suggests that there is any common factor (maybe substrate?) between the Italic, Celtic and Greek shifts?
@DE: [Separately, I’m trying to think of an ethnonym that genuinely has the literal meaning “brothers”, but drawing a blank.]
I always wondered about the Ichwan Bedwine, who had that name even before they got to Arrakis, as I recall. Fictionalily shouldn’t disqualify them.
But I don’t remember anyone mentioning that Greek had a similar shift
It’s not very similar …
The Greeks, never ones to do things by halves, had different developments of the labialised velars in different dialects – as well as before different vowels.
In Attic, they managed to shift them to alveolars before front vowels: hence τίς τέ corresponding to Larin quis, que.
I don’t think it’s necessary to invoke substrates for the demise of lablialised velars or of labial-velars. They’re not very popular.
Mooré and Farefare have both simplified their labial-velars (popular in West and Central Africa, if not so much elsewhere) to plain velars, which looks all very Sprachbundy, given that they’re near neighbours (though not contiguous); you can fairly easily prove that it can’t be an inherited common innovation, and Bisa, the Mande language that intervenes geographically between them, hasn’t got any labial-velars. A lot of former Bisa speakers now only speak Mooré, but that seems to be a fairly recent development, and I don’t know of any evidence that any forebears of the Farefare were Bisa.
Hardly any Oti-Volta ethnonyms are analysable; only Batammariba “Builders” for sure, and perhaps Nõoteba “Farmers” (though the tone is wrong, if so); conceivably “Mossi” was “Red People”, but that one’s pretty much just my own guess.
Unanalysability seems to be very much the norm in those parts. Not even “Hausa” has anything but a frankly speculative origin. A fair number of folk etymologies are about, but the very existence of folk etymologies tells its own story: nobody actually knows what the ethnonyms originally meant.
…sure, let’s spell -/jawa/- as -WI-JA- for shits & giggles…
Oh, look, a whole new phoneme for Proto-Semitic, demurely hidden in parentheses.
Thought to have gone *[kʷi] > *[kᶣi] > *[tɕʷi] > *[tʃi] > [ti], which makes phonetic sense. The last step looks very un-European, but such things happen routinely elsewhere, and Greek developed quite the allergy to affricates in general. In particular, *-[ts]- became -tt- in Attic (but -ss- in most other dialects, even Ionic).
(Also, Central Bavarian turned [kx] into [k] just a few centuries ago.)
Those people are possibly attested in the Egyptian list of Sea Peoples, and they do have three consonants there… d-n-n.
Never mind that d-y-n would be in the wrong order anyway; if -wi-ja- can mean -/jawa/-, anything goes.
Leaving aside all the other objections to Cyrus Gordon’s confident equation of Linear A kuro “total” with Semitic kull- (like, no explanation of the second vowel, -r- handwaved as equivalent to /l/), it occurs to me that these documents were (if Linear B is anything to go by), not drawn up by budding Homers, nor by mystical adherents of Mother Goddesses, but by, not to put too fine a point on it, suits. Accountants.* Just the sort of people who might use a nice foreign jargon term to mark a “total” (a word which English has taken from the poetic language of Roman devotional works; and “sum”, of course, is borrowed from the title of Thomas Aquinas’ magnum opus.)
In relation to kull-, I’ve often also been struck by how its Arabic avatar has been remarkably widely borrowed in West African languages (e.g. Humburi Senni Songhay kul “all.”) I think “all” is one of those words that gets into Swadesh lists (and the like) because scholars familiar with Indo-European and Semitic think of it as particularly stable, but it’s so far from being stable in Oti-Volta that I ditched it from my lists, on the grounds that even closely related languages often had completely unrelated words. Putting it into the lexicostatistics just increased the noise-to-signal ratio. (I was only using it for hints to relative closeness of languages, not attempted demonstrations of relatedness, still less anything chronological.)
* Of course, accountants can be mystical adherents of Mother Goddesses too. But usually they tend not to express this much in their spreadsheets/tablets.
That recent paper/“element”/whatever of Kate Kelley’s on proto-Elamite was the (properly scholarly) antithesis to this kind of thing: we can’t decipher the script, and it may not even be something that can be deciphered in the usual (linguistic) sense, but we can still make reasonable deductions about genres and about the place of these inscriptions within that society by looking at the overall patterns.
Proto-Semitic *dyn is ‘to judge’ or so. So Δαναοί is ‘judges’? The middle y is dropped because it’s practically a vowel, I guess. The suffix is Greek, because.
Timmy, Dana, and Donna are frantic.
/r/ and /l/ aren’t distinguished in Linear B, and evidently not in Linear A either, so that part isn’t surprising; but this does rather strongly suggest the script* was not developed for a language that was Semitic As We Know It.
And yes, the tablets (there are a few other inscriptions… like the one the OP is about) bear spreadsheets, not texts in any stricter sense. They’re standardized enough that they are pretty well understood despite the continuing dearth of understanding of the language(s).
* “Cretan hieroglyphics” at first.
I was recently informed on Social Media that the Δαναοί were Israelites, viz. the tribe of Dan. Even though the usual timeline for the Trojan War is hard to reconcile here with the usual timeline for the Lost Tribes becoming Lost.
David Marjanović: Hence German Büchse “small box; (obsolete) gun”. The tree is Buchsbaum, so the ü does not come directly from the original y.
From which FYLOSC got ‘piksla’ (ashtray).
Recently seen: “I’m not posting any Odyssey memes, because I fear the GIFs bearing Greeks.”
Heh. (And it evinces the Only Correct Pronunciation of GIF.)
I have a vague memory of seeing something claiming that Greek is a Semitic language. Does anyone know if there has ever been such a fringe theory?
This was from about forty years ago. The friend of a friend of my parents had a library with a small number of what I guess were monographs. They were slim (1cm thick or so), slightly taller than a paperback (so maybe octavo size?), perfect bound, with pale green covers of heavy paper stock. The subjects of the other monographs were not linguistics, I think. Or at least, they did not make particularly outlandish claims that I can recall.
(While trying to search for evidence that such a work existed, I stumbled upon a different fringe theory: The Phoenician Origin Of Britons, Scots & Anglo-Saxons, by Laurence Waddell. Indeed.)
https://archive.org/details/josephyahudahebrewisgreek/mode/1up
We’ve discussed it here before, needless to say. The author means exactly what he says: not Hebrew is related to Greek, or Hebrew is influenced by Greek …
I actually saw this for sale, new, in my youth. I nearly wish I’d bought it.
Sadly, my own monograph, Mooré is Swedish, was less successful, owing to the machinations of Big Linguistics and the Scandi Noir lobby.
I sometimes wonder whether the key insight of comparative linguistics (and perhaps of the whole scientific method) might not be “pure coincidences actually happen all the time.”
Or, in the words of the Master:
https://www.nga.gov/artworks/165729-mr-natural
As Kostas Georgakopoulos attests, “Crete is the only place in the Aegean world where a word connected to Ahhiyawa appears.” (2012). That word is A-KA-WI-JA-DE, recorded on a Linear B tablet from Knossos
FWIW, Georgakopoulos’s paper, “Minoan-Anatolian relations and the Ahhiyawa Question: a re-assessment of the evidence”, is here. P. 147 is where the inscription is mentioned. That inscription, C914, may be seen here. An annotated key to all of the Linear B vocabulary is here.
“Hebrew is Greek”
to him.
It’s Chinese to most people, though.
“This object’s media is not available for download.”
It may not be legally available for download, but, c’mon, this is ridiculous. You can just right-click on the picture and download it. Or open it in a new tab. There’s not even a transparent GIF on top of it!
…unfortunately with all spaces rendered as slightly negative, a bizarre thing academia.edu sometimes does.
So is the annotated Linear B vocabulary – it’s housed on kinezika.com!
Look, the word “is” has been in English for a long time; since the time of the Phoenicians, even. Clearly it has gained enormous heavy lifting power over that time from being exercised so much.
(Pos lete “monomania” b’ivrit?)
(GT says it’s just borrowed, but I would guess something like shiga’on yichidi; a singular madness)
Somewhat more seriously, thanks for the reference, which is certainly the first time I’ve seen it. But it can’t be the work I am faintly remembering. There are too many pages; the monographs must have been in the range of 100-150 pages or so.
@DM: Here is the paper at the source, without the a.e nonsense.
Thanks! Have downloaded, will read.
In suggesting that ahhijawa and A-KA-WI-JA have syllables and vowels reversed, I fear we’re getting misled by our term Achaean. I don’t think ahhija- is a root here.
What Georgakopoulos writes is that “Achaiwia” appears on a Linear B tablet meaning Achaean. He doesn’t mention A-KA-WI-JA-DE though that version is found in other academic writing online. I have neither Greek nor a university affiliation to track that through journals very far, but I see a lot of references. This isn’t Georgakapoulos’ main point. He seems to be accepting someone else’s proposal, and it seems
I think the root (or should I say the “stem”?) as used in the Linear B must relate to Greek Achaio- (not Achai-) with the suffix -ija, found in other Linear B words creating an adjective or substantive noun (ie of ethnicity) out of a place term, so to show the parsing better, something like A-KA-O(I-JA). Along with -de, mentioned as an allative.
There’s something a little wonky there — using an allative with something that has transformed a placename into something else. Like talking about going to Russiania. But maybe -ija can also create a substantive noun of place out of a root of ethnicity.
The Mycenaean word would presumably be independent of the Hittite word, so I don’t think it’s so weird to see a different arrangement of syllables when the syllables that differ are just suffixes on the root. The HIttites heard the root as Ahhi- (as did the English, apparently), but the Cretans knew it as Achaio-.
That doesn’t add up to evidence for a Hebrew origin, of course, but I’m trying to understand what is being proposed.
What Georgakopoulos is ultimately trying to suggest is that Achaean is parallel to Pelasgian and that both have a word for sea in their etymology. He relates Achaean to the aqua etymon and Pelasgian to pelagos. In short, he wants both words to mean Sea Peoples.
That’s creative and fun. Again, I wonder whether anyone here finds it plausible.
He relates Achaean to the aqua etymon and Pelasgian to Pelagos. In short, he wants both words to mean Sea Peoples.
I thought that looked janky, but was hoping David Marjanović would comment on it first.
The “aqua” etymon itself seems to be confined to Italic and Germanic.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/h%E2%82%82%C3%A9k%CA%B7eh%E2%82%82
Also, I don’t think *kʷa would turn up as ka in Mycenian Greek, where the labialised velars were still alive and well. And it wouldn’t give *kha in later Greek either.
It’s just not possible as an intra-Greek development. But presumably it’s supposed to come from a handy imaginary “substrate”?
I can see why the Egyptians might lump together some of these groups as “Sea Peoples”, but why would any of them have used that as a self-designation?
Busy with reviewing a large manuscript with a very large and very sloppily written list of references. – I’ll still read the paper once I find the time, but at this point it looks like I don’t need to comment anymore. Finding “aqua” in there does not work at all whatsoever, and our knowledge of the Pelasgians is almost literally limited to their name, so… pseudoscience.
>I thought that looked janky
Going back and re-reading it, hoo-boy:
>The Indo-European root Ach- (Wait, what root?) is connected with water. Many names of rivers and lakes, such as Inachos- Ἰναχος, Acheloos- Ἀχελώος, Lake Acherousia- Ἀχερουσία, and others have this root (Sakellarakis/Sapouma-Sakellarakis 1997, 47). Even in the Hittite language the word akw-anzi means ‘they drink’ (Gurney 1990, 99), while the Luwian word aku- means ‘to drink’
Maybe it would be better to read Achaeans as drunkards. And the greatest drunkard was Ach-
illes.
There is some scholarly discussion that Hittite eku- / aku- could relate to the Italic and Germanic words. There’s still no “I.E. ach root”. And nothing to push such an etymon into Greek but a small set of river names that have other proposed etymologies and no known sound change pathway.
It doesn’t; if it did, it would consistently be spelled ḫakku-, never e- or even a-, and never -k-. This is briefly discussed in the blog post I linked to.
As also presented there, the Anatolian word does seem to have a Latin and a Greek cognate, but no Germanic one, and they have meanings related to “drinking”, not “river” or “water”.
The “Sea People” were called both “the Water People” (Achaeans) and “the Shore (?) People” (Pelasgians), but never “the Sea People”. Hm.
(“The Sea People” being a concept created by later historians to describe the invaders who came to Egypt and Palestine from beyond the horizon. In a place like Crete, everyone and their neighbors were Sea People.)
“Aqua” is obviously from the Dagara a kʋ̃ɔ “the water.”