This thematic issue of SSLA comprises six articles from one research program, High-Level Proficiency in Second Language Use. This program has involved approximately 30 researchers from four language departments at the Faculty of Humanities, Stockholm University.Footnote ¹ In a national competition with 15 other proposals within the social sciences and humanities, the program received long-term funding (years 2006–2012) from Riksbankens Jubileumsfond (the Bank of Sweden Tercentenary Foundation).Footnote ² In this thematic issue of SSLA, selected empirical findings from the first four years (i.e., 2006–2009) are reported. During its first four years, the program has consisted of nine subprojects investigating advanced, near-native, or even nativelike levels of proficiency in a second language (L2). The goal has been to take a decisive step forward in the understanding of the cognitive and psycholinguistic, structural-linguistic, sociopsychological, and societal conditions that, on the one hand, allow language learners to attain such outstanding levels in a L2 and, on the other hand, prevent them from going all the way to become entirely like first language (L1) speakers. In brief, the focus of the research program has been on L2 users who approach, brush against, or, indeed, reach the borderland where the two categories of L1 and L2 users of a language cease to be distinguishable. The focus on high-level proficiencies in a L2 is strongly motivated by current developments in SLA theory. As will be shown, the last phase of successful L2 acquisition is actually the one that is least well understood; however, the searchlight has begun to be directed toward the issues implicated at this level.

THEORETICAL BACKGROUND

The bulk of the early descriptive and groundbreaking SLA research in the 1970s and 1980s concerned early or middle stages of acquisition with a focus on the identification of developmental sequences in grammar, phonology, or lexicon. These sequences were alternatively interpreted within nativist, universalist, crosslinguistic, functional, or processing-oriented approaches to the theory of L2 acquisition (see Doughty & Long, Reference Doughty, Long, Doughty and Long2003b). Even though it was generally acknowledged that L2 acquisition takes place in a social context—with few exceptions, such as Schumann’s acculturation model, which intended specifically to account for early fossilization in a L2 on the basis of social and psychological distance (Schumann, Reference Schumann1978), and early L2 studies of migrant workers in Germany (Heidelberger Forschungsprojekt: Pidgin-Deutsch, 1975; Meisel, Clahsen, & Pienemann, Reference Meisel, Clahsen and Pienemann1981), relating L2 development, or lack of such development, to the social conditions in the learning context—the field basically developed a linguistic and cognitive rather than social focus.

The development of the field over the 1990s and early years of the twenty-first century has involved a diversification in many directions, with certain strongholds. Among the central directions are (a) the characterization of learners’ early varieties (Perdue, Reference Perdue1993); (b) processability theory (Pienemann, Reference Pienemann1998); (c) the universal grammar (UG) framework for the study of L2 competence, with its more recent development into studies of interfaces (summarized in White, Reference White, Doughty and Long2003, Reference White, Ritchie and Bahtia2009, respectively); (d) processes of L2 acquisition such as the mechanisms of incidental and intentional learning (Hulstijn, Reference Hulstijn, Doughty and Long2003); and (e) the declarative and procedural determinants of L2 acquisition and implicit and explicit knowledge and learning (DeKeyser, Reference DeKeyser, Doughty and Long2003; Paradis, Reference Paradis2009). Additionally, research has seriously scrutinized factors that may explain individual differences such as age of acquisition (see DeKeyser & Larson-Hall, Reference DeKeyser, Larson-Hall, Kroll and De Groot2005; Hyltenstam & Abrahamsson, Reference Hyltenstam, Abrahamsson, Doughty and Long2003), motivation and aptitude (see DeKeyser, Reference DeKeyser2000; Dörnyei & Skehan, Reference Dörnyei, Skehan, Doughty and Long2003; Robinson, Reference Robinson2005), and identification (see Block, Reference Block2007). SLA research framed on recent directions in social theory such as social constructivism, socioculturalism, and poststructuralism is also taking on a significant role in the field (Block, Reference Block2003; Ortega, Reference Ortega2009). The continuing proliferation of theoretical approaches, it must be added, is, of course, also a weakness (Long, Reference Long2007).

After more than 40 years (Gass, Reference Gass, Ritchie and Bhatia2009) of empirical research in this field, the accumulated body of knowledge about L2 acquisition is indeed impressive (see Doughty & Long, Reference Long, Doughty and Long2003a; Ritchie & Bhatia, Reference Ritchie and Bahtia2009). However, that knowledge is less elaborated about stages in which the actual continued learning of the L2 becomes less obvious and the communicative use of it takes precedence. Even though learning is clearly still going on during these phases, speakers may appear as if they have stopped developing their L2 proficiency. The focused study of advanced levels of language proficiency only started in the 1990s, yet it is possible to identify three areas in which high-level proficiency in L2s has been researched. Two of these directions have become centrally integrated into the theoretical development in SLA; the third one has not—at this point—done so, although, in our view, it has definite potential to contribute significantly to SLA theory.

COGNITIVE AND PSYCHO-NEUROLINGUISTIC PERSPECTIVES

Research in the first area has been based on questions and hypotheses generated by the notion of a critical period for language acquisition (for overviews, see Abrahamsson, in press; Birdsong, Reference Abrahamsson and Robinson2009; Hyltenstam, Reference Birdsong, Ritchie and Bhatia2010; Hyltenstam & Abrahamsson, Reference Hyltenstam and Hogan2003). Among the questions that have been asked in this area, the most salient ones have concerned (a) a characterization of the ultimate level of attainment, given an onset of L2 acquisition at different ages (Johnson & Newport, Reference Johnson and Newport1989); (b) the existence of L2 users who started their L2 acquisition after puberty and yet reached a nativelike level of attainment (Abrahamsson & Hyltenstam, Reference Abrahamsson and Hyltenstam2008, Reference Abrahamsson and Hyltenstam2009; Birdsong, Reference Birdsong1992; Bongaerts, Reference Bongaerts and Birdsong1999; Coppieters, Reference Coppieters1987; Ioup, Boustagui, El Tigi & Moselle, Reference Ioup, Boustagui, El Tigi and Moselle1994); and (c) the issue of whether prepubertal or childhood learners uniformly attain a nativelike level in their L2 (Harley & Hart, Reference Harley and Hart1997; Hyltenstam, Reference Hyltenstam and Harris1992). As discussed in Hyltenstam and Abrahamsson (Reference Hyltenstam, Abrahamsson, Doughty and Long2003), the traditional assumption of a critical—or sensitive—period with a successive offset around puberty must be considered too simplistic in light of current available data. Although a case can be made for the existence of maturational (biological) constraints on (L2) acquisition, details on the relationship of age of onset and ultimate attainment, along with the interaction between biological factors and social, psychological, and cognitive factors, need to be further investigated.

The first two articles in this special issue are representative of a cognitive or neurolinguistic perspective on the role of age, maturational constraints, and language-learning aptitude in high-level L2 acquisition and bilingualism. First, in a study of 200 L1 Spanish speakers of L2 Swedish, Abrahamsson investigates the relationship between age of onset of L2 acquisition and (nativelike) ultimate attainment of L2 phonetic and morphosyntactic intuition as well as the question of whether children and adults approach the task of learning a new language in fundamentally different ways—that is, incidentally or implicitly and intentionally or explicitly, respectively. The article by Bylund, Abrahamsson, and Hyltenstam describes a study of the balance between L1 and L2 grammatical intuition in 30 early, highly proficient Spanish-Swedish bilinguals. The focus is set on the question of whether a nativelike L2 proficiency hampers a nativelike L1 proficiency and vice versa, as well as on what factors are the most reliable for predicting whether or not early bilinguals eventually end up as nativelike speakers of both their languages. The findings—contrary to some current suggestions (e.g., Flege, Reference Flege and Birdsong1999; MacWhinney, Reference MacWhinney, Kroll and de Groot2005; Ventureyra, Pallier, & Yoo, Reference Ventureyra, Pallier and Yoo2004)—clearly speak in fact to the position that L1 maintenance does not hamper L2 nativelikeness (see also Hyltenstam, Bylund, Abrahamsson, & Park, Reference Hyltenstam, Bylund, Abrahamsson and Park2009).

STRUCTURAL-LINGUISTIC PERSPECTIVES

The second area is closely linked to earlier studies that describe developmental stages and acquisition sequences in initial and middle phases of acquisition. The questions asked have evolved around how grammar, lexicon, and discourse develop interrelatedly at later stages of the learning process and, ultimately, around what features found in native use of a language are still lacking at the very advanced level of L2 users of the same language (Bartning, Reference Bartning1997; Lambert, Reference Lambert1997). Among the landmarks in this development are studies that deal with the interface of information structure, meaning, and linguistic form (Bardovi-Harlig, 2000; Bylund, Reference Bylund and Pavlenko2011; Carroll, Murcia-Serra, Watorek, & Bendiscioli, Reference Carroll, Murcia Serra, Watorek and Bendiscioli2000; Schmiedtová, von Stutterheim, & Carroll, Reference Schmiedtová, von Stutterheim, Carroll and Pavlenko2011). The methodological breakthrough made possible by large computerized L2 corpora allows new insights on acquisitional sequences and developmental stages at this high-proficiency level (Bartning & Schlyter, Reference Bartning and Schlyter2004). Additionally, research has focused on learner varieties at advanced stages (Perdue, Reference Perdue2000).

Thus, the third and fourth articles in the present thematic issue deal with structural-linguistic aspects of the very last stages of L2 development. In a study of 30 L1 Swedish high-level speakers of L2 French, Bartning, Forsberg Lundell, and Hancock offer a linguistic contextual explanation for morphosyntactic deviances (MSDs), and the results are discussed in relation to the ongoing debate on the constructs of complexity, accuracy, and fluency. The article by Bardel, Gudmundson, and Lindqvist reports on the design and use of a profiler for lexical sophistication created in order to assess the lexical richness of intermediate and advanced L2 French and L2 Italian learners with Swedish as L1. The frequency-based method normally used in lexical profiling is modified by recategorizing some low-frequency words that teachers consider easy, and it is argued that this method gives a more correct picture of advanced L2 lexical profiles.

Attempts have also been made at a general characterization of the linguistic features that mark near-native and nativelike proficiency (e.g., Hopp, Reference Hopp2009, Reference Hopp2010). However, another direction in this second area is represented by work in the UG framework that seeks to define what is characteristic of end-state grammars—that is, of final states or, indeed, near-nativeness (Sorace, Reference Sorace and Lambropoulou1998). Sorace (Reference Sorace, Doughty and Long2003) suggests optionality as one characterizing feature of end-state grammars, that is “the existence of two or more variants of a given construction” (p. 135). This is parallel to earlier claims about systematic variability as a general defining feature of interlanguages (Hyltenstam, Reference Hyltenstam1978). In the same vein, the theoretical discussion of the issue of completeness contributes to the identification of what distinguishes language use in the last stages of L2 acquisition from L1 use (Schachter, Reference Schachter1990).

SOCIOLINGUISTIC AND SOCIOCULTURAL PERSPECTIVES

The third area comprises research on the development of youth varieties in multicultural contexts (Källström & Lindberg, Reference Källström and Lindberg2011; Kotsinas, Reference Kotsinas, Androutsopoulos and Scholz1998; Rampton, Reference Rampton1995; Sebba, Reference Sebba1997). What is particularly interesting is that these varieties develop the way they do for intertwined reasons, only one of which is the fact that some of the speakers who use them are L2 users (Bijvoet & Fraurud, Reference Bijvoet, Fraurud, Källström and Lindberg2011). In fact, speakers who use such youth varieties may well be L1 speakers, but because some of the characteristic features are perceived as nonnative or, in fact, have their linguistic origin in other languages present in the contact situation, speakers may be perceived as L2 users. Much research in this area has focused on what significance and meanings are expressed through the use of youth varieties and how the speakers’ identification and positioning are reflected in and shaped by their experiences and social interactions (see Haglund, Reference Haglund2005; Pavlenko & Blackledge, Reference Pavlenko and Blackledge2004).

The sociolinguistic and sociocultural perspective on high-level language acquisition and use is represented by the two final articles of this thematic issue. The article by Bijvoet and Fraurud presents results from a speaker evaluation study in which Swedish speech stimuli from 12 young Stockholmers were evaluated by 343 listeners from different backgrounds. The study demonstrates how young people may divide and relate to the linguistic space of Stockholm in very different ways and that they vary in accuracy with regards to linguistic self-perception. It is suggested that the diversity of constructions of sociolinguistic variation as well as gaps between self-perception and production must be considered in the study of language development and use, especially in multilingual contexts in which a majority of the young people cannot be described easily in terms of native-nonnative speakers, L1-L2 users, or mono-bilingual speakers. The article by Eliaso Magnusson and Stroud deals with data from young people who were born and raised in Sweden, all of whom ethnically self-identify as Assyrian or Syrian, and whose repertoires are complexly multilingual. All participants are generally perceived to be native speakers of Swedish on a daily basis, but at certain moments, these young people are classified as near-native or simply not native. The article presents analyses of their narrative recounts of metalinguistic reflexivity from occasions and interactional moments when they are classified as nonstandard speakers—and therefore near-native speakers or learners. From these findings, the authors suggest revisiting notions of nativelikeness.

LABELS FOR HIGH-LEVEL L2 LEARNERS AND USERS

A number of concepts related to high-level proficiency in SLA have been introduced in the previous section. In the present research program (and, consequently, in this thematic issue), concepts describing the category of learners-users—on whom both this program and this issue focus—include advanced, near-native, nativelike, and native language users. These labels, in order of mention, are taken to characterize language users that occupy successively more final sections of the L2 developmental continuum, in which, in the end, the categories of L1 and L2 speakers merge. Starting from the end, when we discuss a native speaker of a language—without going into the complexities involved in this notion (see Cook, Reference Cook1999; Davies, Reference Davies2003)—we mean the archetypical L1 speaker, who is not necessarily monolingual but who has learned that language and continued to use it regularly throughout the life span. A nativelike speaker of the same language is someone who, in all respects, uses the language like a native speaker, in spite of the fact that the language in question is not the user’s L1. By a near-native speaker we mean someone who is perceived, in normal oral interaction, as a native speaker but who can be distinguished from native speakers in some feature when his or her L2 proficiency is analyzed in greater linguistic detail (Hyltenstam & Abrahamsson, Reference Hyltenstam, Abrahamsson, Doughty and Long2003). By an advanced L2 learner or user we mean a person whose L2 is close to that of a native speaker but whose nonnative usage is perceivable in normal oral or written interaction. As the superordinate term for these categories of language proficiency, we use the notion high-level proficiency (see the research project High Level Language Ability Research Hub, directed by Catherine Doughty at the Center for the Advanced Study of Language [CASL] at the University of Maryland).

The notions of L2 learner and L2 user have already been commented on. It is clear that learning, or acquisition, and use of a L2 are two activities that go on simultaneously and in parallel over the entire L2 development continuum. As previously noted, it is possible to discern a difference in research focus. Thus, a concentration on acquisition has been salient for early and middle phases, whereas an interest in aspects of language use has been perhaps more typical in research on high-level language proficiency. The much-debated notion of fossilization—that is, the putative cessation of target-oriented acquisition before a nativelike proficiency level has been attained (see Han, Reference Han2004)—is adequate in relation to this distinction. In light of current research, however, a total cessation of development is unlikely; rather, it is likely that the changes may not be target-oriented in an efficient way (Long, Reference Long, Doughty and Long2003).

Not only is the theoretical understanding of issues related to high-level proficiency of L2 use relevant for the theoretical development of SLA, but it also has an obvious and clear application in teaching practice and in other aspects of professional life in which high-level proficiency in a L2 is required.