09 August 2014

Latin aqua 'water' (updated)

Latin aqua 'water' < Paleo-European *akʷā (f.) is an interesting word with correspondences in Germanic *áxwō 'river' (Gothic ahwa, Old High German aha) and the Old European Hydronymy (OEH), a corpus first studied by Hans Krahe and more recently by Francisco Villar and which represents an older form of IE than the historically attested languages1. On the other hand, Celtiberian akua is a substrate loanword whose meaning is probably 'river' (cfr. tar akuai 'across the river' in Botorrita III2).

Contrarily to what many scholars think, the original meaning of *akʷā isn't 'water' but 'river'3, being a derivated noun from the adjective *aku- 'quick, fast', found in Latin acupedius 'swift of foot' and accipiter 'hawk', the latter with cognates in Greek ōkýpteros, literally meaning 'swift flyer' (ōkýs 'swift', pterón 'wing'), and Sanskrit āśu-pátvan- 'flying swiftly'4. These correspondences suggest Paleo-European *aku- would correspond to late IE *ōu- 'quick' > Latin ōcior.

On the other hand, the geminate stop in accipiter would be the result of a sound shift -kʷ- > -kk- which I call Kretschmer's Law and of which are examples in Italoid (e.g. the Lusitanian theonym Iccona) and Celtic (e.g. *sukko- 'pig'). There're also traces of it in Vulgar Latin 'water', as reflected in the Appendix Probi (aqua non acqua).
More specifically, its agglutinative morphology would correspond to the "IE I (pre-flexional)" stage proposed by Francisco Rodríguez Adrados. Of course, Vennemann's proposal of OEH being related to Basque must be rejected.
P. de Bernardo Stempel (2007): Water in the Botorrita Bronzes and Other Inscriptions, in Palaeohispanica 7, pp. 55-69.
Despite so, it has been compared it to other 'water' words, both within the IE family and oustide, even up to the point of building upon it a "global etymology" (Ruhlen).
M. de Vaan (2008): Etymological Dictionary of Latin and the Other Italic Languages, p. 21.


  1. Hi, Octavia!

    I found a confirmation of my previous statements from Vaclav Blažek. Below is his etymology. I only found this yesterday.

    Semitic √ʕ-q-q: Arabic: ʕaqq “to make the cloud to rain”, √ʕ-q-y Arabic: ʕaqa “to
    give to drink”
    (Steingass 1988, 710, 714) |||?Egyptian place name ʕq3 ‘Pehhu-waters’ (WPS 292) |||
    Cushitic (Central): *ʕaqº “water” > Bilin: ʔ/ʕaḳº, Khamtanga: aqº, Kemant: axº; Awngi: aγu || (East)
    haqa, D’irayta: haḳa id. ||| Omotic: (North) Yemsa: akà id. (Appleyard 2006, 144)

    IE: *H2ekG- “water”: Latin: aqua “water”; Celtiberian: akua in tar akuai “through water” (de
    Bernardo Stempel 2007, 58); Gothic: aƕa “Fluss, Gewässer”, Old Norse: ǫ́, Old English: ēa, Old
    Saxon, Old High German:
    aha, German: aha id. (Pokorny 1959, 23).
    Lit.: Trombetti apud Dolgopolsky 1964, 8; Illič-Svityč I, #139: AA+IE.

    3 = Egyptological Aleph which could represent /a/, /l/, or /ɹ/

  2. Thank you for the quote. Now comes the answer.

    The IE etymology is nuts, but the Celtiberian form is interesting because it's a foanword from a substrate language spoken in the Iberian Peninsula.

  3. Hi, Octavia!

    Correction: IE: *H2ekG- should be IE: *H2ekº

    This substrate word you speak of in Celtiberian can not be Basque, Iberian, or "Tartessian". I am convinced that Tartessian is related more to Basque and Iberian than to anything IE (lots of linguistic research and learning since we last spoke on that topic).

    But then again it could be Basque, the closest I could find is ugarre "torrent" and ugaitz "torrential river" both of which are derived from the Basque root ur "water". Interestingly, there are similar words in Afroasiatic to Basque: ur "water".

    Pokorny gives this etymology:
    aku̯ā-, more accurately əku̯ā, ēku̯- 'aqua, water, river'

    Notes: the /u̯ / should be superscripted

  4. Hola, Octavia!

    Hahahaha, I am sure you are cuter. :)

    Correction: Pokorny’s etyma

    This is Jesus Rodriquez Ramos’ entire quote about Sorothaptic (I hope he does not mind…if he does you may have to remove it or I delete it…however that works):

    “On Sorothaptic. It is better to forget this name. It has a very odd story and is a rather unfortunate idea. Corominas use to explain the palatalization of initial L- in Catalan (so lluna < luna, llac < lacus, llum < lumen) through the influence of some X people. He invented the name of "Sorotàptic" as equivalent to Urn-field... and so on.... as L to LL is a normal linguistic process (as r- to strong r-) it is rather a phantom people ((Let me explain that even if Coromines make some serious mistakes, he, who is seen as a God of linguistics in Catalonia, by common Catalans, and so saying that I make some kind of heressy, when I say he make serious mistakes the real question is that ha studied in Spain were the Linguistics traditions was very sub-standard, so he make what he could with so few resources)). The most similar to it would be Ligurs from South France, but again... what is Ligur?. It is not really clear. Probably an Indoeuropean, probably a non Celtic Indo-European, but it is not clear (BTW Avienus said that Tartessos was in the "Ligustinian" lake, not in the Celtic Lake; although it is clear that Avienus was more a Roman dilettante that a real researcher but.... neither Herodotus had a good knowledge on Iberian and was very kind of believing phantasy histories).”

    Personally, Octavia, I have seen the opposite. All the books and papers I have read most of the authors recognize difficult words and speculate a possible substrate. In Afroasiatic linguistics, authors don’t shy away from the speculation possible unknown substrates, I guess because of the well-known linguistic situation of Africa. They also acknowledge the “weirdness” that is Afroasiatic, hahaha. In Afroasiatic, as you may very well know, sometimes, the languages quite don’t follow rules like many linguists think they should. Lots of regular and irregular alternations in phonemes, like /m ~ b/ (possibly indicative of an Afroasiatic prenasalized stop /mb/) and /k ~ g ~ q/ and /b ~ v ~ w ~ p ~ f/ and the problematic Afroasiatic voiced velar fricative /ɣ/ which can be derived from or give rise to a number of phonemes. One of the most common Afroasiatic alternations is /r ~ l ~ n/ and, sometimes, /d/ is included in that group. In Chadic, /d ~ r/ is common. Also, /ʔ ~ h ~ w/ happens in Afroasiatic. Also rather common regular and irregular alternations between emphatic consonants and regular consonants such as /ḳ ~ k/ and in Egyptian /k ~ ṯ/, /d ~ ṯ /. There are so many more than mentioned here, rarer, but they occur. Don’t get me started on Berber, hahaha. In Afroasiatic it is difficult to determine substrates unless they are straightforward because of the “weirdness” that Afroasiatic can sometimes be and because of the great age of the proto language and its daughters. Lots of differences can develop over those great time periods. Sometimes, if you find a word common in contact areas like the ones between Afroasiatic and Nilo-Saharan…it is difficult to determine from which way they derive...for instance, from Nilo-Saharan > Afroasiatic or from Afroasiatic > Nilo-Saharan.


  5. (part 2)

    I agree with you about amateur linguists. I had that discussion with Mr. Valerio. I told him that amateurs often pick up what “pros” miss or refuse to acknowledge. Another linguists and I got a heated response from Claude Rilly, a well-renowned Egyptologist and Meroiticist. We pointed out some errors in his handling of Meroitic vocabulary (Wikipedia Meroitic Language Talk Page). Most of the words he claims are Nilo-Saharan are, in fact, Afroasiatic. To me, and many others Afroasiatic is the most valid classification of Meroitic…where in Afroasiatic is another issue. Though many are still pushing Nilo-Saharan. We know much much more about Afroasiatic now than in the 1950’s, 60’s, and 70’s. So I am fully in agreement with you about what you say about IE-ists. The same phenomenon occurs in the realm of Afroasiatic linguistics. Proto-Afroasiatic looks like it was in two places at once or at least had a massive extent. The urheimat is generally agreed to be somewhere within Northeastern Africa, but linguistic examination would say the Near East (A. Militarev). So I tend to agree, in part, that PIE has different areas of contact with other languages which are difficult, but not impossible, to explain. For instance, contact with Uralic on hand, Kartvelian, Hurrian, Afroasiatic and then later Semitic (Afroasiatic) on the other. Uralic being well north of the Caucasus and the latter 3 being well south of them. Also some speculation of IE words in Sumerian and vice versa. There is a similar problem with the Uralic urheimat.

    The way you describe the expansions or waves of IE sounds like the “Peopling of the Americas” theory, hahaha. You say Mesolithic, but most linguists say Middle Neolithic. How do you reconcile that?

    Yes, I know there were several IE etyma there….it was intended to demonstrate the "IE-ness" of akua.

    As for “Tartessian” I will refer you to the "Tartessian" language Talk page on Wikipedia…see my comments.

    Also, I was absent from the web for a while, I came back online a few months ago. I needed a break from the virtual world. So I was not aware of many of your posts. I was in Uni (still in Uni, Master’s program) and proceeding with my career so that left little time for much else. I have been dedicating much time to researching Meroitic and a likely or possible Afroasiatic relationship.