Friday, August 19, 2016

Infant Translators: the Salamanca Twins

In the collection quoted from in the preceding post there is another article that deserves to be saved from drowning in the flood of literature about translation that sweeps past us these days, It's by a team of linguistic researchers, Esther Álvarez de la Fuente and Raquel Fernández Fuertes, of the Language Acquisition Lab at the University of Valladolid, Spain. The data for it was harvested from a pair of English/Spanish bilingual twins at the old university city of Salamanca, not far from Valladolid. (It's unusual to have twins as subjects, but Esther and Raquel don't follow up that aspect of the situation.) The data was videotaped and is deposited in a computerised corpus, the FerFulice Corpus, which is incorporated in the CHILDES database.

The article analyses the spontaneous and elicited translating in the speech of the boys, named Simon and Leo, from the age of 1 year 11 months to 6 years 3months; an impressive total of 178 sessions were video-recorded at regular intervals. Their mother was American, their father Spanish. Each parent always spoke to the children in her or his or own language, following the OPOL (one parent one language) principle for avoiding confusion between languages. The recordings were made in natural settings while the boys were engaged in normal play activities.

For the analysis, a matrix of variables devised by Esther was used (and modified slightly here): COMPLETENESS (complete, incomplete, null), STIMULUS (requested, spontaneous), DIRECTION (towards English, towards Spanish), ORIGIN (self-translation, translating what was said by others, situational), MAPPING BETWEEN LANGUAGES (equivalent with communicative function, equivalent without communicative function, expanded, reduced). Other researchers may care to use it. What is very desirable is to arrive at a commonly accepted set of variables in order to facilitate comparison between studies.

So let's look at some examples.
1) Mother: Can you say water?
Mother, holding up the cup of water: What is this?
Leo, reaching for the cup: Ahi! [There!]
Mother: Water?
Leo: Agua.
(Age 1 year 2 months)
This was remarkably young, indeed before the age of speaking in sentences; but it replicates Jules Ronjat's observation made a hundred years ago (see References) that his son Louis composed French/German bilingual word pairs at that age. Obviously the situation helped in the present case.
2) Simon, trying to get his toy to make a noise: Está loto [a mispronunciation of roto].
Mother, not paying attention to Simon: How about…?
Simon: B(r)eak mommy b(r)eak.
(Age 2 years 3 months)
Now he was at the stage both of sentences and of communicative intent. Still remarkably young. Furthermore he differentiates between his languages and understands the language need of his interlocutor.
But not all the children's translation attempts are successful. That would be too good to be true. Thus:
3) Mother, pointing to an elephant: Look, look, show me that animal.
Mother: What's it called?
Leo: Elefante [Spanish for elephant].
Mother: Can you say that in English?
Leo, with a trace of tears in his voice: No, elefante.
(Age 2 years 7 months)
There are several things to note in this example. First that Leo's translating – like all Natural Translation – is limited by his proficiency in the two languages. Natural Translators don't use dictionaries. Secondly that he wants to translate and feels frustrated at not being able to do so. And thirdly that his mother doesn't ask him to translate (a word that was probably not yet in his vocabulary) but to say it in English.

Without a doubt this study ranks in importance, by its length and thoroughness, with the earlier studies by Harris, Swain and others right back to Ronjat. (For more about them, enter their names in the Search box on the right.) It's one of only a handful of such studies.

One final piece of good news is that you no longer need to fork out 50 euros to buy the book in which the article appears, because Esther has posted a collection of that and other related articles in the invaluable repository and you can download free them by clicking here or from Some of the articles are in English some in Spanish.

The children involved in child language brokering are usually of school age and socialised beyond the family. For much younger translators who are still confined to the family, like the ones cited above, I propose infant translator. Hence the title of this post.

Esther Álvarez de la Fuente and Raquel Fernández Fuertes. How two English/Spanish children translate: in search of bilingual competence through natural interpretation. In M.A. Jiménez Ivars and M.J. Blasco Mayor (eds.), Interpreting Brian Harris: Recent Developments in Translatology, Bern, Lang, 2012, pp. 95-116.

The address of the Language Acquisition Lab is\uvalal.

Jules Ronjat. Le développement du langage observé chez un enfant bilingue [Language development in a bilingual child]. In French. Paris: Champion, 1913. 155 p. Available online by clicking here or at

Left: Esther Álvarez de la Fuente. Right: Raquel Fernández Fuertes.


  1. This comment has been removed by a blog administrator.

  2. After looking into this blog on your site, I really like your technique of writing a blog. I bookmarked it to my bookmarking site list and will be checking back it for my future reference. Take a look at my website as well and let me know your opinion.
    Android training in chennai>