An intriguing possibility, if no more than that: “Making Sense of “Nonsense” Inscriptions: Non-Greek Words Associated with Amazons and Scythians on Ancient Greek Vases” (pdf), by Adrienne Mayor (Stanford University), John Colarusso (McMaster University), and David Saunders (J. Paul Getty Museum), asks “whether some nonsense inscriptions and non-Greek words associated with figures of Scythians and Amazons represent meaningful sounds (phonemes) in foreign languages spoken in ‘Scythia’ (Black Sea-Caucasus region)”:

We analyze the linguistic patterns of nonsense inscriptions and non-Greek words on thirteen vases featuring Scythians and Amazons by otherwise literate vase painters (550-450 BC). Our results reveal that for the first time in more than two millennia, some puzzling inscriptions next to Scythians and Amazons can be deciphered as appropriate names and words in ancient forms of Iranian, Abkhazian, Circassian, Ubykh, and Georgian. These examples appear to be the earliest attestations of Caucasian and other “barbarian” tongues. This new linguistic approach to so-called nonsense inscriptions sheds light on Greco-Scythian relations, literacy, bilingualism, iconography, and ethnicity; it also raises questions for further study.

The question is a perfectly reasonable one; the authors push their proposed answers more than seems to me compatible with what I would consider appropriate scholarly reticence, but as long as you pay close attention to words like “can,” “may,” “plausible,” and “possible,” it’s very enjoyable, especially if (like me) you have an affectionate interest in both ancient Greek and the languages of the Caucasus. My verdict: not proven and probably not provable, but fun to think about. Below the cut are some quotes from the paper to give you the general idea; for the nitty-gritty, the actual examples with Colarusso’s analyses (he’s an expert on Caucasian languages; I wrote about his Nart book here), go to the paper itself. (Thanks for the link, Alex!)

The Hellenized term amazones may have had multiple sources from related Eurasian languages. One likely source was a-maz-ah-na, Northwest Caucasian for “Forest (or Moon) Mother.” Amezan (a-mez-a-ne) was the name of a heroic horsewoman-warrior-queen of the Nart sagas, oral traditions that combine ancient Indo-European myths and North Caucasus folklore. The Circassian form is pronounced amazan (last a long), the same as in ancient Greek. The word probably entered the Greek language, along with stories about fighting women of the East, through the Black Sea trading ports where ancient Caucasian, Iranian, and other languages were spoken. … The non-Greek origins of the word amazon and the ethnonyms and toponyms preserved by Herodotus and others raise an intriguing possibility. Did some of the names assigned by Greek writers and artists to individual Amazons also originate in the languages spoken by people of the Caucasus, Black Sea region, and Scythia? Most names assigned to Amazons are etymologically Greek. It is possible, but not provable, that some of these women’s names were originally foreign and translated into Greek. In other cases, non- or pre-Greek names might have been “rationalized,” that is, made to look as if they were derived from Greek roots, as with the folk etymologies of “Aphrodite” and “amazon.” … As noted above, the Hellenized word amazones appears to have a Caucasian source, and in ancient Greek thought, Amazons were understood as related to Scythians. Therefore, it seems reasonable to ask whether any of the strange-sounding “nonsense” inscriptions associated with Amazons and with Scythians on ancient vases could have been intended to represent genuine non-Greek words from the Caucasus and neighboring regions. …

Our methodology: Co-authors Adrienne Mayor and David Saunders selected about twenty vases depicting Scythians and Amazons accompanied by “nonsense” inscriptions. We transmitted the inscriptions in Greek letters to Colarusso and asked him whether the sound patterns matched any known language forms. This was essentially a blind experiment. Colarusso knew only that the project involved strange words inscribed on ancient Greek vases that showed people in Scythian costume, but he was not shown photos of the vases until the end of the project. …

Because we are working with what might be words from unwritten foreign languages that were transliterated into Greek letters by artists who illustrated vases more than two millennia ago, our conclusions cannot yet be verified scientifically. We can only offer plausible interpretations and impressionistic guesses. Given the multiplicity of languages to be considered and the brevity of single or incomplete “nonsense” words, there is always the risk of false positive results. Yet a majority of the so-called meaningless inscriptions that we selected produced suggestive results. …

These preliminary findings suggest that at least some unfamiliar strings of letters on Attic vases may not be meaningless after all. By seeking interpretations of these “nonsense” inscriptions in terms of Circassian, Abkhazian, or Ubykh, with some Iranian (Ossetianlike) and Georgian forms, we show that what appear to be incomprehensible words in Greek can be deciphered as names or descriptions of figures of Scythians and Amazons. …

If these are, as they appear to be, the oldest attested examples of Caucasian languages, the overall picture that emerges from these case studies gives linguists an astonishing glimpse into the evolution of those languages. …

To conclude, our brief study has resulted what we regard as plausible translations of non-Greek languages on Attic vases that were long thought to be meaningless scribbles. Our linguistic analysis recovers several new names of Scythians and Amazons and descriptive words in ancient Scythian languages, words that have remained undeciphered for 2,500 years. If our linguistic impressions and speculations are on the right track, uncovering ancient traces of spoken “barbarian” languages is an exciting and historic discovery. Not only do these vases speak again, showing ancient Greek relations with cultures to the East, but “nonsense” inscriptions might contain the earliest written examples of ancient forms of Northwest Caucasian and other languages spoken by “barbarians.”


  1. thirteen vases … shed light on Greco-Scythian relations, literacy, bilingualism, iconography, and ethnicity: that’s five illuminations from thirteen objects. Good going.
    It reminds me of reading Dark Ages history: one author will discuss whether King Arthur even existed, the next will argue the case for his coming from Kelso. You can enjoy them both without being under any necessity to believe either.

  2. Exactly!

  3. Trond Engen says

    I’m not through the paper yet, but, preliminary speaking, I’m annoyed with how they use of the words dialect and root.

  4. “You can enjoy them both without being under any necessity to believe either.”
    or one can enjoy them both believing both too, cz as K.Prutkov said famously nel’zya ob”yat’ neob”yatnoe and maybe different people perceive just as if like different sides of a historical or whatever fact, incidence etc i mean truth

  5. “Note that the name for the state of California originated from Latin for an
    Amazon queen Califa, cali-forn- hot-copulation”
    Is this true? The Wikipedia article doesn’t mention this derivation.

  6. Uh, that sounds pretty dubious to me. As I say, they’re not real picky about their etymological connections. Use only as directed!

  7. Amazon queen Califa
    Must have been the first Muslim religious leader. 😉

  8. California was the name given to a mythical island populated only by beautiful Amazon warriors using gold tools and weapons in the popular early 16th-century romance novel Las Sergas de Esplandián (The Adventures of Esplandián) by Spanish author Garci Rodríguez de Montalvo. This popular Spanish novel was printed in several editions with the earliest surviving edition published about 1510. The island was ruled by Queen Calafia.
    Montalvo may have taken the name from the Arabic “caliph”. There is an otherwise unknown placename Califerne in the Chanson de Roland. Or several other suggestions have been made, or he could have just made it up out of thin air.
    The name was bestowed by Hernando de Alarcón in 1540 to what is now called Baja California, specifically in reference to Montalvo’s novel.

  9. I could add that Las Sergas de Esplandián belonged to the genre that was so soundly mocked by Cervantes in Don Quixote.
    James Phelan, a wealthy banker who served as Senator from California, built an elegant estate in Saratoga, California, which he named Villa Montalvo. It is now an arts center.

  10. What on earth is “sergas”? It’s not even in the Real Academia’s Diccionario de la Lengua Española!

  11. Diccionario Porrua de la Lengua Española (1979) defines “sergas” (plural only) as “hechos, proezas, hazañas”. A similar definition can be found in Thesaurus gran Sopena de sinónimos y asociación de ideas (1984). But it is not found in several other dictionaries I looked at.
    Looking at
    Las sergas de Esplandián (Las hazañas de Esplandián) es el quinto de la serie española de libros de caballerías iniciada con el Amadís de Gaula. Su autor fue Garci Rodríguez de Montalvo, quien también escribió el libro cuarto del Amadís.”
    I presume “sergas” is obsolete in modern Spanish.

  12. I’ve found it in my Larousse Diccionario usual, which gives the same definition as your Diccionario Porrua (“hechos, proezas, hazañas”) and then says “únicamente empleado en el título de un célebre libro de caballerías, Las sergas de Esplandián“; maybe Montalvo invented it?

  13. One dictionary explains it as from the Greek
    “erga” given a Spanish plural and with the initial “s” coming from the article (IOW the author heard “las ergas” as “las sergas.”)

  14. I mean, “obsolete” doesn’t begin to cover it—it doesn’t seem ever to have been used except in that title.

    [But see Alon Lischinsky’s comment below.]

  15. I’ve seen the “hot-copulation” thing before, but only as a joke; the fortune file on my computer includes the following:

    California, n.:
    From Latin “calor”, meaning “heat” (as in English “calorie” or Spanish “caliente”); and “fornia'” for “sexual intercourse” or “fornication.” Hence: Tierra de California, “the land of hot sex.”
    — Ed Moran

    If Mayor & Saunders took something like this for a true etymology, well, it does not inspire confidence.

  16. You must have a Berkeley-style fortune file! A fellow slackware-user? (My new laptop, though, likes not Slackware — by crashing every time when I try to leave X — and I switched to the much less BSDish Ubuntu)
    (apparently B-S-D-s-t-y-l-e is harmonized by Hat’s engine, does that have anything to do with the Bosnian-Serbian-Dalmatian-Montenegrin thing?)

  17. If they’d gotten two ancient Caucasian language experts (are there two out there?) to independently agree on the languages and translations, and mixed up the inscriptions that were suspected to be meaningful and gibberish instead of handing them to the translator in separate batches, they might have had something. After all, the premise that there might be semi-literate foreign artisans working in ancient pottery factories isn’t absurd. It’s a pity they didn’t run their experiment sensibly. (It looks to me like the derivation of “California” is their own sillyness, and they left the actual linguistics to the experts.)

  18. On second thought, I might be thinking about this too hard. And I sort of doubt there were enough literate Scythian potters to leave more than the occasional potsherds in transliterated Circassian behind.

  19. Please note this is a WORKING paper: we welcome constructive criticism and suggestions for improvement.

  20. John Emerson says

    Don’t have time to read the article, but these might be Greek garbling of those languages or parodies rather the real thing.

  21. I should have known that the authors would be reading the comments here, so Mayor, I’m sorry for being so snide. To rephrase what I said more constructively, if you could run the inscriptions by a second caucasian language expert and get something like the same translations, I think it would be pretty convincing. And don’t experiments involving humans usually mix up the control samples (here, the inscriptions that you expect to be gibberish) and the actual test material (the scythian inscriptions) so the testee doesn’t pick up on the pattern? But it’s a really cool idea!

  22. We have had positive reactions so far from linguists. If you can recommend another expert in North West Caucasian, I’d be delighted.
    All the inscriptions were emailed to Colarusso in batches of 3-10 words at a time. There was no discernible pattern and we had no idea whether individual word-strings would turn out to be gibberish or not.
    We appreciate everyone’s helpful critiques and questions and will acknowledge this list in our final paper.

  23. Surely if anyone can recommend another expert in North West Caucasian it would be Dr. Colarusso himself. Specializing in such an esoteric area of scholarship, one would suppose he would know every expert in the field at least by reputation, if not personally.

  24. Exactly. We have already received many useful comments from relevant linguists known to Dr Colarusso.

  25. SpammerI: The next time I read a blog, I hope that it doesnt disappoint me as much as this one. I mean, I know it was my choice to read, but I actually thought youd have something interesting to say. All I hear is a bunch of whining about something that you coul??
    Spammer II: I??ve read much confusing information on this topic and have been disappointed in the content. Your article makes this information clear and easy to understand.
    This post started a ground-shaking debate within the spam community. I trust we’ll be kept informed on any new development.

  26. This paper has now been published.

    I don’t see any major changes from the draft. It’s a pity there is no contribution from an additional Caucasian language specialist. If a second expert had reached similar conclusions as Dr. Colarusso it would have very much strengthened their argument. Still a fun exercise and worthy of further study but I think to be filed as “not proven”.

  27. Thanks for the update, and I agree.

  28. Alon Lischinsky says


    What on earth is “sergas”? It’s not even in the Real Academia’s Diccionario de la Lengua Española!

    That’s not really saying much; the Diccionario Usual, as its name indicates, is supposed to list only current terms in general use (although the Academy often honours this point more in the breach than in the observance). The Diccionario Histórico, however, is only available in a very partial form.

    A quick look through CORDE suggests that the term has been very rare historically, but (pace your Larousse) is attested outside that particular title, e.g.:

    Lagartijillo, poniendo cátedra, narraba ante una copa de macharnudo bizarras sergas del toreo (González Anaya, Salvador. 1944/1929. La oración de la Tarde. Madrid: Biblioteca Nueva)

  29. Thanks!

Speak Your Mind