Reading The Recognitions (see this post) involves encountering a whole lot of allusions, and one of them was to a Saint Olalla. Wanting to make sure I was pronouncing that right (/oˈlayə/, in Americanized form), I looked it up and discovered it was a by-form of Eulalia, which made sense. But then I noticed that the Galician form was Baia, which didn’t: “O nome Baia (Olalla) é a evolución galega do nome culto Eulalia do grego Ευλαλια.” Can anyone explain how you get Baia from Eulalia?


  1. Giacomo Ponzetto says

    In Catalan you get from Eulàlia to Laia. Originally as a nickname, unsurprisingly since Catalan nicknames traditionally form by apheresis. Nowadays as the much more common form of the name: Spain has 14 thousand Eulalias, with an average age of 63.5, and 36 thousand Laias, with an average age of 16.4.

    I have no idea how Galician got from L to B. But then Baia doesn’t seem to be a current form of the name. Hardly any Spaniards officially bear it (less than 20, and possibly none). The represented alternatives are Olaya, Olalla and Olaia, which get 3 thousand Spaniards each with an average age somewhere in the early 20s.

  2. David Eddyshaw says

    Spode’s Achilles’ heel!

  3. Giacomo Ponzetto: Thanks! It seems odd, then, that Galician Wikipedia has the article under Baia.

  4. The entry about the name says, “A variante máis común en Galicia (por ser patroa en máis parroquias) é Baia, sendo a variante culta Eulalia. Existen ademais as variantes Olalla (moi frecuente hoxe en día), Olaia, Oalla, Alla, Balla e Oballa.”

    So it looks like the first /l/ is lenited away, and you’re left with something like [oaja] > [oβ̞aja] > [β̞aja]. The approximant is an allophone of /b/, represented by that letter.

  5. Ah, makes complete sense — thanks very much!

  6. Is that also how we have Basque / Euskara?

  7. They may be unrelated. Basque comes from a Latin name for a local tribe, the Vascones. For Euskera, this study says:

    The suffix -(k)ara / -(k)era applied to euskera/euskara, the basque word for basque language, which in the case of erdara/erdera «non basque language» appears without the -k-, comes, according to the author of this work, from the feminine romanic forms aira > -era, both an evolution of the latin suffix -arius, -a, -um.

    Considering on the one hand the XVI century form enusquera, by Esteban de Garibay from Mondragon, which would habe been pronounced ê-ûskera, as a nasal vowel, and on the other hand some parallel forms like jazkera, «Way of dressing» from jantzi «to dress» and euskera, «way of holding» from eutsi, to hold, a form that would never have had nasality. The author proposes *enau(t)si, «to say» as a basis for the forms euskera/euskara. This participle would have given verb flexions such as diñost/diost, with a Biscayan variant diñaust «he says it to me» meaning originally «to say».

    (N.B. “habe”. This is the English abstract.)

  8. There was also an Aquitanian tribe, the Ausci. Their name has been suggested as te source for the present-day language name, but that doesn’t explain the -n- in Garibay’s form.
    Whether their name is related to that of the Vascones, I don’t know, and I think nobody else does.

  9. That Irigoyen quote is really interesting, thanks!

  10. Athel Cornish-Bowden says

    How do you get from “Mary” to “Polly”?

  11. The initial step from Mary /mɑri/ to Molly /mɑli/ is straightforward at least.

    My guess probably would’ve been for Baia to come from the clipping of a more Koine-ish Εβλαλια, which even is indeed the same mechanism (rounded vowel yields /w/ > /v/).

  12. Also see Santa Vaia, a Galician toponym and synonym for December (the feast of Saint Eulalia of Mérida is on December 10). And this from the “Glosario da poesía medieval profana galego-portuguesa”:

    Santa Vaia top. ‘prob. topónimo do norte portugués, derivado do lat. Sancta Eulalia, con resultados fonéticos variados (Santa Vaia, Santa Ovaia, Santalha…)’

    They quote from an old cantiga:

    pois me vou de Santa Vaia,
    morarei cabo da Maia,
    en Doir’, entr’o Port’e Gaia.

    The narrator goes away from Santa Vaia somewhere in Galicia down south to the Douro valley (my conjecture), between (O)porto and Vila Nova de Gaia. Which makes sense because Porto’s historical center is on the northern side of the Douro and Gaia (now part of Porto) is on the southern side.

  13. CuConnacht says

    Molly to Polly makes as much sense as Meg to Peg.

  14. And William to Bill, and Robert to Bob…

  15. It’s just a hypocoristic rhyme, not an obscure sound change: see Dolly (< Dolores) to Polly, and (says internet) Josie (< Josephine) to Posie.

    (And that got me finding the vast and wild rabbit hole that is The Dictionary of Medieval Names from European Sources. Maybe worth a post, Hat?)

  16. I posted it back in 2015, and you made the lone comment!

  17. I initially assumed you meant the Medieval Names Archive, which for some reason aroused considerably more interest.

  18. Ah, and I see the 2015 post was made because of Sara’s comment at the end of that thread.

  19. @Y: That “habe” looks like a typo influenced by interference from a different language. I wouldn’t have commented on it, except that just as I was reading that, I got a request to referee a physics paper by a pair of Russian authors. The paper is a confusing mess, for all the reasons that usually happens; however, in this case, there is the additional problem that the running text abruptly switches back to Russian in several places. (Clearly, proofreading of the manuscript was minimal to nonexistent.)

    @Athel Cornish-Bowden, CuConnacht, languagehat & Y: I noticed, since the last extensive discussion I remember of nicknames with initial sound substitutions, that the best known such changes in English (William to Will to Bill, Richard to Rick to Dick, Edward to Ed to Ted, Mary to Molly to Polly, Margaret to Meg to Peg) have something notable in common. They all change the initial sound, which is a sustainable consonant (or a vowel in Ed), to a stop. I don’t know what significance, if any, this fact has, but it seems interesting to me at least.

  20. Brett, there’s also N in Ned, Nell, and Nancy, which is not a stop, but maybe that’s because the names start with vowels. Maybe it’s a version of the Michael Jackson sound change.

  21. John Cowan says

    There is also Richard > Rick > Hick > hick, Robert > Rob > Hob, Roger > Rog > Hodge (the name of Dr. Johnson’s cat). The OED reports Hick in 1565. I wonder if this has to do with (still non-standard) [ʁ] in French being heard as English [h].

  22. @Keith Ivey: Of course, you are right; there are also names that go from an initial vowel to N. I meant to say something about that, but I forgot what I was going to say and then forgot that I had forgotten. I guess that means that with a name like Ed[ward], there are two paradigms it can follow for a nickname with an initial consonant. I guess the question is why there is not the same alternation in the girls’ names. (I think Gan exists as a derivative of Ann[e], but it is rare.) Maybe it is just highly atypical to have more than one nickname with just a clipping and an altered onset, and the N versions happened to win out for Elanor and Ann.

    And, naturally, I also forgot to include Robert to Rob to Bob.

    @John Cowan: Obviously, those nicknames with H are no longer commonplace, but they do provide interesting further historical data.

  23. January First-of-May says

    I think Gan exists as a derivative of Ann[e], but it is rare.

    I wonder whether Hanna(h) ~ Ann(e) were perceived as variants of the same name historically.

  24. Roger > Rog > Hodge (the name of Dr. Johnson’s cat)

    And also > Dodge (the source of my surname, as well as Lewis Carroll’s).

  25. By the way, the first sentence of this post made me think of Stevenson’s Olalla, and it wasn’t a wasted thought: as A Reader’s Guide points out, there are at least two quotations from the novella in The Recognitions. I wonder how Stevenson meant Olalla to be pronounced. “I had seen her—Olalla! And the stone crags answered, Olalla! and the dumb, unfathomable azure answered, Olalla!” But what is the reader supposed to hear?

  26. @Alex K.: Olalla (or, in the current orthography, Olaia) is the Basque form of Eulalia. I don’t know how Stevenson would have pronounced it, but the original is roughly /oˈla.ja/

  27. @Alon Lischinsky: This form is also thought to be Galician, Asturian and/or Leonese in origin, although it made it as far south as Andalusia. The problem is, yeísmo was not nearly as widespread in Stevenson’s time as it is now, and even less so during the Peninsular Wars. Moreover, it spread among urban residents first while lleísmo persisted among rural speakers. Stevenson’s Olalla would have pronounced her name with an /ʎ/, most likely. A more interesting question is how the author intended it to sound in English. I’d also love to know if Stevenson was familiar with Poe’s Eulalie.

  28. @Alex: Stevenson’s Olalla would certainly not have used /ʎ/ if she were Basque, and not even /ʝ/. The second consonant in that name is /j/

    Were she Leonese~Asturian~Galician (or high-valley Aragonese), there’s a fair chance her speech still made distinction between /ʝ/ and /ʎ/, but that would be increasingly unlikely anywhere else. Yeísmo had been consolidating for at least three centuries by then

Speak Your Mind