Sunday 9 October 2011

An Italian-Spanish phrase book in IPA

I’ve just come back from a one-week trip to the Spanish island of Mallorca. As you know, the island is famous for its breathtaking coastal scenery, which is dominated by fine white sandy beaches, caves, and rocky coves. 

From a linguistic point of view, Majorcans are usually bilingual in both Catalan and Spanish. This is reflected in the way they pronounce the name of their island: in Spanish it’s maˈjorka, -ʎor-, whereas in Catalan it’s generally məˈjɔrkə, -cə. (Notice the type of pre- and post-stressed vowel used in the transcriptions: a schwa in Catalan but a full vowel a in Spanish.) 

Amongst the highlights of our tour were the so-called Cuevas del Drach (in Catalan Coves del Drach), four great caves which extend to a depth of 25 m and which contain an underground lake, called Martel Lake, where classical music concerts are held on a daily basis. On the left you can see a picture of the lake with three little boats on it in which the musicians perform their tunes.

One of our guides also took us to a spectacular beach, Es Trenc, which is really a crystalline turquoise water paradise for divers. Have a look at this picture on the right.

On my way to Mallorca I was considering myself as on summer leave from all things phonetics, but I just couldn’t resist taking a look at an Italian-Spanish phrase book which my friend Rossella kindly showed me on the second day of our trip. You’ll be surprised to hear that her phrase book wasn’t just an ordinary phrase book, but one with IPA transcriptions of both Spanish words in isolation and in connected speech. You can see a picture of pages 12-13 of her book Spagnolo per viaggiare (1999; Giunti) below. 

Notice, for instance, how the phrase hace dos días (‘two days ago’) is transcribed: ˈaθe ðɔs ˈðias. As my readers will know, in Spanish the voiced plosive d has the fricative allophone ð when intervocalic, and this is correctly shown in the transcription. (Ok, the symbol used in the book is not exactly ð, but no one’s perfect!). This process is also visible in antes de final de mes (‘before the end of the month’), in which (most) Italians would simply use a plain dental

On page 7 of the book, phonetics editor Leonardo Lavacchi explains how b d ɡ are pronounced in Spanish:

β, ð, ɣ sono le varianti fricative delle occlusive b, d, ɡ e si pronunciano con minor tensione lasciando che l’aria esca dalla bocca provocando una leggera frizione; [...] χ fricativa velare sorda [sic] si pronuncia con una frizione fra il dorso della lingua e il velo palatale”. 

(‘β, ð, ɣ are the allophonic variants of the plosives b, d, ɡ, and are pronounced with less tension in the oral cavity, causing the air to become turbulent on escaping from the mouth. [...] The voiceless velar fricative χ [sic] is produced by raising the back of the tongue towards the velum, thus causing friction.’)   

Apart from the incorrect use of the symbol for the voiceless velar fricative x (χ represents the  voiceless uvular fricative which is, by the way, also found in Spanish before u as an allophone of x), I find it absolutely amazing that an apparently ‘run-of-the-mill’ phrase book can contain IPA transcriptions for most of the words and utterances it lists. 


Congratulations, Giunti, and thank you very much, Rossella, for sharing the book!


  1. Brilliant, I agree. It's quite moving because nobody seems to be interested in these matters.
    And you might be interested to know that most native speakers of Spanish are not at all aware of the existence of the fricative allophones of our b, d, g -it may take you some effort to convice them that they do utter these allophones when speaking in a natural way. I understand things like that also happen in other places: Gimson recommended not to ask lay English speakers about their actual pronunciation of, say, weak forms of words.

  2. Been to Mallorca quite a few years ago. It was a blazing summer. It's a beautiful island. Except for a corner of one place, which I renamed El Urinal - disgusting!

  3. Very interesting Blog, congratulations. One observation to your post. Both variants [χ] uvular and [x] velar exist in Spanish dialects . The uvular version appears in Castilian Spanish.
    Emma Miliani

  4. Mi dispiace ma non sono in grado di esprimermi correttamente in inglese per parlare di fonetica. A dir vero, nemmeno in italiano, poichè sono ingegnere,non filologo.
    Vorrei fare soltanto qualche piccola correzione.
    La pronuncia in spagnolo della "ll" in Mallorca non è laterale, ma aprossimante palatale (simile all'italiano iere o ione).
    La "d" di "hace dos días" è più aprossimante che fricativa.

    D'altra parte sono d'accordo con Beatrice. Lo spagnolo medio è convinto che parla come scrive e che i fonemi li pronuncia sempre allo stesso modo independentemente della posizione nella parola. Ma insisto che l'allofono che viene pronunciato non è sempre quello fricativo ma quello aprossimante.

    Se ci fosse qualche modo di registrare la mia voce, vi porrei qualche esempio. Penso alla parola "caballo", cavallo in italiano, horse in inglese.
    Io pronuncio caballo con una aprossimante bilabiale invece di una fricativa bilabiale sonora. Mentre lo spagnolo madrelingua crede di fare una occlusiva bilabiale sonora.

    Facundo Agüero

    1. Facundo, grazie per aver commentato! Per la scelta dei suoni e relative descrizioni cf. pp.221-224 del testo "Practical Phonetics and Phonology" (2013; 3rd edition; Routledge).