Verified by 
Post written by Aneta Pavlenko.
As bilinguals, one of our greatest hopes is that the efforts we have invested in learning a second language (L2) will also pay off when we learn the third one (L3). The more languages we know, the easier it should be to learn them, right? A positive answer to this question can have tremendous implications for language education. And yet researchers limit themselves instead to the non-committal “it depends." Why is that? And what does our success in L3 learning depend on?
The first undeniable advantage involves strategy. If you learned your L2 as a teenager or an adult, you have managed to figure out what strategies work best for you. The more extraverted among us enjoy talking to target language speakers, while the introverts prefer to pour over textbooks and grammar books. I, for one, am an avid reader and foreign movie watcher, always hoping that new words will find their way into my memory effortlessly, in the process known as incidental learning. Research also shows that learners who are literate in both of their languages, and those who have metalinguistic knowledge, are in a particularly advantageous position because, among other things, metalinguistic awareness allows them to make comparisons and generate creative hypotheses.
Another area of potential advantage involves cross-linguistic similarities: learning a language typologically similar to the language or languages we already know allows us to utilize our prior knowledge through the process known as positive transfer or reliance on already familiar sounds, words, and grammatical categories. Unfortunately, not all of the similarities we perceive are real. Sometimes, our interlingual connections are erroneous, leading to what is known as negative transfer, which includes, for example, our reliance on false cognates, or faux amis (for a previous post on this topic, see here). It was such superficial similarity between the Spanish word embarazada (pregnant) and the English embarrassed that led the marketing people at Parker Pen to enter the Mexican market with a bold slogan: "No te embarazará chorreándose en tu bolsillo" (It won’t impregnate you by leaking in your pocket).
And here is where it becomes interesting when a third language is concerned. We know that our first language (L1) affects the process of learning other languages, but we do not always expect that our L2 also wants to play an active role in the learning of L3. Yet this is precisely what happens and it is the influence from L2 Spanish that leads English-speakers learning L3 Italian to produce forms like aiudarono (helped, a blend of L2 Spanish ayudaron and L3 Italian aiutarono) or uccido (killed, a blend of L3 Italian ucciso and an L2 Spanish past participle –ido).
But then again it shouldn't really come as a surprise that there is an interaction between typologically-related languages. My own Spanish mercilessly interferes with my Italian, while Polish and Russian emerge, uninvited, every time I try to speak Ukrainian. While inconvenient, this is hardly surprising. What is utterly counterintuitive is to see a typologically distant L2 (e.g., French) actively interfere with the L3 (e.g., German), as in the case of American linguist Larry Selinker who asked his German colleague in L3 German: Tu as mein Fax bekommen? (Did you [in french] get my fax [in German]?). The replacement of the L3 German Du hast with the similarly sounding French Tu as caught them both by surprise yet this is exactly what happens time and again—the previously learned foreign language suddenly pops out. Such interference, dubbed a foreign language effect, is so frequent that it may lead us to suspect that our brain has a separate section for all languages learned later in life.
Studies using neuroimaging techniques suggest that this is not the case. Rather, it appears that language processing is not restricted to single sites in the brain but spread across various parts. In addition, the same areas subserve different languages of multilingual speakers. The only difference is that the use of the native or the dominant language is optimized and thus more automatic, and the use of other languages requires more cognitive resources. The shared representation, in turn, means that all languages are connected in the multilingual mind. Like our second language, our native language does not remain an independent observer in the process of L3 learning. Instead, it influences it, through both positive and negative transfer, in ways that are sometimes subtle and at other times pretty visible, as we have just seen. Our L1 may even team up with the L2 in messing with our L3. And our first language is not immune from influences either, as experienced by the English speaker who upon returning from a short, two-week, trip to Germany joyfully informed his friends that He had drinken many beers (although, in this case, we may also be seeing a beer effect).
All in all, we should probably expect that the effects of the previous languages on L3 learning will be diverse. On the one hand, we come to the learning process as more experienced and better equipped learners but, on the other, our languages are not immune to playing tricks on us. The best we can do is to relax and enjoy the learning process, as well as appreciate the insights we invariably gain about our languages and ourselves.
For a full list of "Life as a bilingual" blog posts by content area, see here.
Photo of three labels on green grass from Shutterstock.
References
De Angelis, G. (2007). Third or additional language acquisition. Clevedon, UK: Multilingual Matters.
De Bot, K., & C. Jaensch (2015). What is special about L3 processing? Bilingualism: Language and Cognition, 18, 2, 130-144.
Selinker, L. & B. Baumgartner-Cohen. (1995). Multiple language acquisition: ‘Damn it, why can’t I keep these two languages apart?’. Language, Culture and Curriculum (special issue on multiple language acquisition), M. Ben-Soussan & I. Berman (eds.). 8.2, 1-7
Aneta Pavlenko's website.
