Spanish as a Pivot Language for Third Language Learning in the United States

Street art from beneath the Coronado Bridge, Chicano Park, Barrio Logan, San Diego, California. Source: Flickr.

Abstract: On college campuses throughout the United States, classes for decades have been offered to help Spanish speakers learn Portuguese, augmented recently by similar courses in Italian, French, and Catalan. Though there exist numerous commonalities between these innovative course offerings, the pivotal role of Spanish stands out. This essay therefore introduces the concept of a pivot language: a widely spoken second language leveraged through explicit instruction to facilitate the learning of a related third language (L3). It is suggested that in an increasingly multilingual country, this trend of taking into account students’ prior linguistic knowledge points to where language teaching is headed.

Keywords: Catalan for Spanish Speakers/catalán para hispanohablantes, cognate languages/lenguas afines, French for Spanish Speakers/francés para hispanohablantes, Italian for Spanish Speakers/italiano para hispanohablantes, pivot language/lengua de enlace, Portuguese for Spanish Speakers/portugués para hispanohablantes, Spanish as a heritage language/español como lengua de herencia, third language acquisition/adquisición de tercera lengua

As the first European language spoken in what eventually became the United States, Castilian Spanish has from early on enjoyed a special place among languages not indigenous to the Americas. Boasting the largest number of native speakers, Spanish today predominates in over half the countries of North and South America, and is by far the language new immigrants to the United States are most likely to speak (Krogstad and Keegan 2014). There are even certain parts of the country (e.g. Puerto Rico, New Mexico) where as a first language (L1) it has a history dating back centuries. The fact, then, that Spanish has long since been the most widely spoken language other than English (LOTE) in the United States—and is projected to remain so throughout the foreseeable future—should be of no surprise.

Considering this, it was only a matter of time before the growing number of post-secondary students in the United States already proficient in Spanish began to influence course offerings in related foreign languages. That it hasn’t happened sooner is perhaps more surprising, due likely to a reluctance on the part of institutions to adapt to change and a hesitance among linguists to recognize the acquisition of a third language (L3) as a phenomenon distinct from that of a second (L2). Thankfully there have been those who for decades have been arguing for this distinction (e.g. Thomas 1985). And while L3 scholars have oftentimes had to prioritize advocating for the very existence of their field, many have moved beyond this basic issue to that of how best an L3 can be taught and learned. The current essay fits squarely within this latter line of research.

This essay will unfold by first looking at the primary role Spanish has assumed, both among LOTEs in the United States, as well as in classes explicitly aiming to teach US college students a third language. After a brief history of the teaching of these classes, the essay will explore some of their common characteristics, particularly the reliance upon a pivot language as the point of departure for learning a related L3. Finally, potential implications will be considered for the field of foreign language (FL) teaching. In the end it is argued simply that it no longer makes sense not to offer language classes expressly to Spanish-speaking bilinguals, with a focus on precisely those issues shown to be salient to L3 learners, reducing time to proficiency and facilitating ultimate attainment.

The Central Role of Spanish among LOTEs in the United States

According to figures from the US Census Bureau, the percentage of people in the United States speaking Spanish at home has risen steadily since 1980, the first year in which that statistic was measured, from just over 5% of the US population to 12% in 2010. While in 1980 Spanish speakers represented 48% of all those speaking a LOTE at home, thirty years later that figure had grown to 62% (Ryan 2013: 5–7). This outsize role that Spanish has played in the linguistic makeup of the United States is only expected to increase as time goes on, with it projected to remain the country’s most widely spoken LOTE for years to come (Ortman and Shin 2011).

Paralleling the above statistics, Spanish is also several times over the most widely taught L2 in the United States, having unseated French for this distinction in 1969. The next benchmark was in the mid-1990s, when enrollments in Spanish eclipsed those in all other FLs combined. According to the Modern Language Association, in 2013 Spanish counted four times as many enrollments as French, itself still the second most popular FL learned in the United States (Goldberg et al. 2015: 27).

With so many speaking Spanish at home, others who have studied it in school, and a significant international student population from Latin America and Spain, the amount of Spanish speakers on US college campuses has become quite consequential. Accordingly, while official figures have never been compiled, an informal survey of university course catalogs reveals that more and more L3 classes are being offered, aiming to take explicit advantage of the fact that so many students are already Spanish proficient. It is with the evolution of this trend that the next section is primarily concerned.

Spanish-based L3 Instruction within US Post-secondary Education

Any thorough history of instructed L3 acquisition in the United States is obliged to point to Sicilian émigré Pietro Bachi as the forefather of teaching related languages. Hired to teach Italian at Harvard University in 1826, within a couple of years Bachi was teaching Spanish to speakers of Italian, and a few years after that, Portuguese to speakers of Spanish, publishing a groundbreaking textbook to accompany each class (Marraro 1944: 568–9).

Although there appear not to have been any such experiments for the next hundred or so years, starting in the 1940s various articles in Hispania began calling for contrastive materials to be absorbed into the teaching of Portuguese in order to help Spanish speakers access the language with greater ease (Percas 1948; Holton 1954). Thanks to the efforts of Jack Ulsh (1971), the Foreign Service Institute soon began doing just that. While it remains difficult to pin down the first time Portuguese for Spanish Speakers (PSS) was taught on college campuses, from the publication of the first PSS textbook (Simões 1991) together with articles from the early 1990s (e.g. Jordan 1991; Milleret 1992), we know that this practice was not all that uncommon. Nowa- days PSS classes are even more widespread, and as a subfield in its own right there have so far been five SEPFE (1) conferences, where PSS and L3 learning are regularly given their academic due.

Over the past two decades, classes in Italian, Catalan, and most recently French for Spanish Speakers have started to emulate the success of PSS at an increasing number of colleges, universities, and even high schools, many times cropping up out of sheer pragmatic will. These classes are often designed by language teachers with little formal training in linguistics who seem to innately grasp the benefits, from raising enrollment and retention rates to boosting morale among heritage speakers of Spanish, who themselves quickly come to view their heritage language as an asset rather than an obstacle (Carvalho 2013). It is up to language researchers and university administrators, then, to embrace the reality of these courses, along with their potential both for language departments and for those studying how foreign languages are learned.

Dynamics and Terminology of Instructed L3 Acquisition

Key to the development of any new field is the vocabulary associated with it. Although there may be a tendency to view a term like “L3 acquisition” (or the occasionally seen variant “tertiary language learning”) as somewhat problematic in its specificity, it seems no more inaccurate than the now well-established term “second language acquisition” (SLA), which is generally used to refer to the acquisition of any non-primary language. Sidestepping issues of individual chronology, instead of relying on futile attempts to encompass a multitude of linguistic backgrounds, the intention here is to identify characteristics—not of the students themselves, but rather those more universally set by educators and administrators.

Complicating the task, dynamics between at least three languages are at play within any L3 classroom. There may be wide variation, for instance, among students’ L1, even within the same class. Students’ L2 may vary just as widely, since L2 for some may be L1 for others. Then what we often refer to as L3 may or may not be the actual third language a student is learning, making the term “L3 acquisition” a potential misnomer. And so it becomes necessary, if we hope to achieve uniformity or accuracy with respect to labeling, to identify characteristics common to the course itself which will hold for all students.

Instead of L3, for example, the term ‘target language’ (TL) will always be accurate in reference to whichever language students are there to learn. This term has been in parlance for years in SLA. Next we consider what might commonly be thought of as L1. While this label will not always refer to the same language across all students, in any FL classroom it will be possible to identify the majority language of the school in which the class is being taught. For many this will likely coincide with L1. Those learners for whom this is not their native language will at least enjoy a strong enough command over it to be able to enroll in a school where, again, this is the dominant medium of instruction. An objective label like ‘majority language’ may therefore be preferable to the more widespread yet far more subjective term ‘source language,’ which in a multilingual setting could be interpreted ambiguously.

The existence, then, of that language in the middle between the majority language and the target language, is what serves to distinguish the L3 classroom and curriculum from all others. Commonly thought of as L2, this language may also very well fail to correspond to the number associated with it, for instance when classes include heritage speakers. Yet because of its pivotal role in facilitating TL acquisition and the increasing amount of L3 classes being offered, in terms of serious research potential, this area will prove more and more fruitful as time goes on. It is with this in mind that the term “pivot language” is proposed, since these classes are designed for students to take their knowledge of a certain language—in this case Spanish—and then turn, or pivot, towards a related foreign language. In the broadest possible terms, therefore, a pivot language can be defined as any widely spoken L2 leveraged through explicit instruction to facilitate the learning of a related L3.

Within this definition lie a few key characteristics. First, a pivot language must be spoken widely enough relative to a certain community (e.g., Spanish), or be of sufficient commercial, strategic, or cultural interest (e.g., Mandarin, Arabic, German), to be able to assemble a class full of proficient speakers. Instruction will typically be somewhat explicit, as this is believed to encourage positive transfer by drawing on correspondences between pivot and target language (Carvalho 2013; Jordan 1991), as well as on the increased metalinguistic awareness that the bilingual (and in particular, biliterate) language learner is often seen to exhibit (Sanz 2000; Thomas 1985). Such explicit instruction may also be effective at reducing interference, or negative transfer, that occurs as a result of the final requirement, that the two languages be related, or in the words of Ulsh (1971), “close enough to each other to enable us to use the word ‘conversion’ when describing what the speaker of one language does in order to achieve command of the other” (viii). In this way, access to the target language via the related pivot language will be virtu- ally impossible to avoid, and whatever the order in which the latter was acquired with respect to other background languages, its relationship to the TL gives it special status (e.g. Rothman 2011). In other words, if the majority language is L1, then the student uses the more proximal pivot language (L2) to gain quicker access to the L3. If the majority language was instead learned as L2, then in this case the pivot language (i.e., the more proximal L1) arguably plays an even bigger role. In either case, the concept is applied consistently and strikes a useful distinction for educators and researchers alike.

While the term ‘pivot language’ has been used in translation work for decades, from simultaneous interpretation to machine translation, its use in FL instruction appears novel. In the world of translation, a pivot language is used to facilitate indirect translation between two other languages whenever direct translation is impossible. Although this is not exactly the sense in which this essay promotes the term’s adoption, the translation world also features terminology such as ‘target language’ and ‘source language’ with meanings that differ slightly from their use in SLA, so this type of borrowing is not without precedent.

In the literature on intercomprehension, a similar concept is often referred to as a ‘bridge language.’ First gaining currency in Europe and now featured within certain L3 classes in the United States, intercomprehension strategies draw upon multiple related languages to promote TL acquisition, where “every language is at the same time source, target, and bridge language” (Donato and Escudé 2013). Because of the unique, multidirectional nature of these strategies, the flexibility of the term ‘bridge language’ seems to strip itself of any exclusive status, and consequently the need remains for something more specific. Put differently, if any language can be used as a bridge, the privileged status of Spanish in these L3 classes would seem to warrant its own term. This distinction serves to highlight one of the major differences between European multilingualism, marked by regional diversity yet relative local uniformity, and multilingualism in the United States, where LOTEs are often spread out geographically, with Spanish the most widely spoken by far. (2) Within such different linguistic environments it should be expected for some terminology to develop independently, and for each community of researchers to establish their own complementary yet varying approaches to multilingualism and L3 acquisition.

Beyond mere terminology, there are precedents for the concept on both sides of the Atlantic. Within Slavic language departments in the United States, for instance, incoming graduate students are often expected to be proficient in Russian. Rather organically in the 1980s, certain programs began using Russian as a pivot language to help students learn Czech or Bulgarian (Gribble 2013; Townsend and Komar 2000). But while that trend seems to have largely died out, in Europe a new one has potentially emerged. In an increasing number of European countries English has not only become the most widely taught L2, but also the most common language for recent immigrants to have previously learned. During a large-scale study of multilinguals learning L3 Dutch in the Netherlands, 68% listed English as their most proficient FL (Schepens et al. 2016). This broad facility with English is rightly being leveraged by European applied linguists, leading to the creation of materials on tertiary language didactics (e.g. Neuner et al. 2009), and an accompanying focus on English as a pivot language in the teaching of additional, related FLs such as German (Hufeisen 2000).

The phenomenon being teased out is tangible, therefore, and the intention of the present essay has been to help garner for it the recognition it deserves. The goal now becomes to probe deeper into the nature of L3 learning itself in contexts where a pivot language prominently features.


Outlined here so far have been the characteristics of a pivot language, as well as some of the reasoning behind identifying it as such. Its exposition brings up further questions, however. For example, is a pivot language always necessary? And if so, how proficient in this language do students, and instructors for that matter, need to be? As these classes very often attract a diverse learner population, the differences in prior linguistic experience, particularly among heritage speakers, can sometimes be rather stark (Johnson 2004). How then should educators handle students with varying degrees of familiarity with pivot-language grammar? In the online era, it is now more possible than ever to adapt curricula to students’ needs, and therein may lie the solution. As time goes on we will likely see more of a reliance upon online tools as supplementary language-learning resources, with those able to successfully incorporate students’ prior linguistic experience helping the most.

Other questions involve the creation of educational materials and L3 curricula in general. For instance, are special L3 textbooks desirable? And if so, what is their proper role? Although many US students of Portuguese speak Spanish, the vast majority of introductory textbooks are still designed with the monolingual English speaker in mind. Additionally, it remains unclear how many PSS classes adopt a substantially different curriculum from that used to teach non-PSS Portuguese (Bateman and de Almeida Oliveira 2014: 276–7). Yet, there exist very clear differences in how L3 classes can be approached compared to the traditional FL model. From the very first day of class, spoken and written fluency in the pivot language allows students to understand much more of what they hear and read than their monolingual counterparts, enabling an earlier reliance upon receptive communication and authentic texts than would otherwise be possible (Carvalho et al. 2010). Furthermore, specialists in FL instruction would do well to consider calls to deviate from the standard communicative model at various points in the retooled L3 curriculum (Carvalho 2013; Jordan 1991).

It may well be, then, that these visionary courses are the harbingers of a significant new trend in FL teaching, taking into account prior linguistic knowledge to an extent that has never been quite as possible, nor as necessary. If we want other L3 Romance classes to follow the trail blazed by PSS, most helpful would be the creation of an overarching group to bring together scholars and instructors under what has emerged as a common thread: L3 instruction most typically featuring Spanish as a pivot language. This seems to the author like the most efficient way to be able to answer key initial questions, such as which classes are being taught at which schools, for how long have they been offered, how much more quickly do students reach proficiency, and by how much have these classes helped boost enrollment. One hopes this type of project will soon be underway, which will then help us answer the deeper questions posed within the preceding paragraphs.

In conclusion, this essay aims to fit in with what needs to be a series of studies identifying common factors in Spanish-based L3 instruction. These efforts will then be able to contribute to the larger conversation, going well beyond any one language or language pair. If these stud- ies can help uncover language-learning universals for L3, the benefit to the field of linguistics would be enormous. Throughout the history of L3 acquisition—especially in terms of teaching related languages and embracing a pivot language in class—both Spanish and Portuguese have played unique roles, which are certain to be more widely acknowledged as time goes on. These forward-thinking pedagogical interventions may well become what help keep language instruction relevant in the twenty-first century, borne of the particular nature of American multilingualism, and fostering in turn a more multilingual US student population better able to handle the challenges of an increasingly interconnected world.


  1. Simpósio sobre Ensino de Português para Falantes de Espanhol
  2. The next most widely spoken language in the United States, Chinese, has over 13 times fewer speakers than Spanish, according to figures from the 2010 Census (Ryan 2013).


  • Bateman, Blaire, and Desirée de Almeida Oliveira. (2014). “Students’ Motivations for Choosing (or Not) to Study Portuguese: A Survey of Beginning-level University Classes.” Hispania 97.2: 264–80. Print.
  • Carvalho, Ana Maria. (2013). “The Field of Portuguese for Spanish Speakers in the US.” Portuguese Language Journal 7. Web. 1 Dec. 2016.
  • Carvalho, Ana Maria, Juliana Luna Freire, and Antonio J. B. da Silva. (2010). “Teaching Portuguese to Spanish Speakers: A Case for Trilingualism.” Hispania 93.1: 70–75. Print.
  • Donato, Clorinda, and Pierre Escudé. (2013). “Mutual Understanding.” Language Magazine 13.1: 20–22. Print.
  • Goldberg, David, Dennis Looney, and Natalia Lusin. (2015). “Enrollments in Languages Other Than English in United States Institutions of Higher Education, Fall 2013.” Modern Language Association. Web. 1 Dec. 2016.
  • Gribble, Charles E. (2013). Reading Bulgarian Through Russian. 2nd ed. Bloomington: Slavica. Print.
  • Holton, James. (1954). “Portuguese for Spanish Speakers.” Hispania 37.4: 446–52. Print.
  • Hufeisen, Britta. (2000). “A European Perspective: Tertiary Languages with a Focus on German as L3.” Handbook of Undergraduate Second Language Education. Ed. Judith W. Rosenthal. Mahwah: Erlbaum. 209–29. Print.
  • Johnson, Keith. (2004). “What is a Spanish Speaker?” Português para falantes de espanhol. Ed. Antônio R. M. Simões, Ana Maria Carvalho, and Lyris Wiedemann. 49–65. Campinas, Brazil: Pontes. Print.
  • Jordan, Isolde J. (1991). “Portuguese for Spanish Speakers: A Case for Contrastive Analysis.” Hispania 74.3: 788–92. Print.
  • Krogstad, Jens Manuel, and Michael Keegan. (2014). “From Germany to Mexico: How America’s Source of Immigrants Has Changed over a Century.” Pew Research Center. Web. 1 Dec. 2016.
  • Marraro, Howard R. (1944). “Pioneer Italian Teachers of Italian in the United States.” Modern Language Journal 28.7: 555–82. Print.
  • Milleret, Margot. (1992). “Cooperative Learning in the Portuguese-for-Spanish-Speakers Classroom.” Foreign Language Annals 25.5: 435–40. Print.
  • Neuner, Gerhard, Britta Hufeisen, Anta Kursiša, Nicole Marx, Ute Koithan, and Sabine Erlenwein. (2009). Deutsch als zweite Fremdsprache. Berlin: Langenscheidt. Print.
  • Ortman, Jennifer M., and Hyon B. Shin. (2011). “Language Projections: 2010 to 2020.” United States Census Bureau. Web. 1 Dec. 2016.
  • Percas, Nicolás. (1948). “What Is Wrong with Portuguese?” Hispania 31.4: 468–71. Print.
  • Rothman, Jason. (2011). “L3 Syntactic Transfer Selectivity and Typological Determinacy: The Typological Primacy Model.” Second Language Research 27.1: 107–27. Print.
  • Ryan, Camille. (2013). “Language Use in the United States: 2011.” United States Census Bureau. Web. 1 Dec. 2016.
  • Sanz, Cristina. (2000). “Bilingual Education Enhances Third Language Acquisition: Evidence from Catalonia.” Applied Psycholinguistics 21: 23–44. Print.
  • Schepens, Job J., Frans van der Slik, and Roeland van Hout. (2016). “L1 and L2 Distance Effects in Learning L3 Dutch.” Language Learning 66.1: 224–56. Print.
  • Simões, Antônio R. M. (1991). Com Licença! Brazilian Portuguese for Spanish Speakers. Austin: University of Texas Press. Print.
  • Thomas, Jacqueline. (1985). “The Role Played by Prior Linguistic Experience in Second and Third Language Learning.” Ed. R. Hall, Jr. The 11th Linguistic Association of Canada and the United States Forum 1984. Columbia: Hornbeam Press. Print.
  • Townsend, Charles E., and Eric S. Komar. (2000). Czech Through Russian. 2nd ed. Bloomington: Slavica. Print.
  • Ulsh, Jack L. (1971). From Spanish to Portuguese. Arlington: Foreign Service Institute. Print.

This essay originally appeared in the special Centenary Issue of the journal Hispania, Vol. 100, Number 5 (December 2017), pp. 279-284. DOI:10.1353/hpn.2018.0068