Graduate School of Education Fall 2018

English Teaching Theories

Table of Contents Section 1: Syllabus & Assessment

Overview of course and weekly plan Grading Criteria In-class activities and projects

Section 2: Readings with Questions

Reading #1: “Introduction: The Nature of Theories” by Bill VanPatten & Jessica Williams from VanPatten & Williams (2016) in Theories in Second Language Acquisition

Reading #2: “Age” by Lourdes Ortega from Ortega (2009) in Understanding Second Language Acquisition

Reading #3: “Interaction and Instructed Language Acquisition” by Shawn Loewen & Masatoshi Sato (2018) in Language Teaching, 51/3.

Reading #4: “Saying what we mean: Making a case for ‘language acquisition’ to become ‘language development’” by Diane Larsen-Freeman (2014) in Language Teaching.

Reading #5: “Second Language Learning Explained? SLA across 10 Contemporary Theories” by by Lourdes Ortega from VanPatten & Williams (2016) in Theories in Second Language Acquisition

Section 1 Syllabus & Schedule

Course: English Teaching Theories Instructor: James Brawn Email: jbrawn67@gmail.com Course Description The purpose of this class is to look at research that informs theories involving the teaching of English as a second language. The goal is to use this research to help us make informed choices about the methods, materials, strategies and techniques that we use in our language classrooms. Text The text for this course will be a packet of materials that you can pick up from 참글. Assessment

30% Attendance and Participation 20% Homework on Readings 25% Presentation 25% Final Paper Attendance & Participation Attendance is mandatory. Participants who arrive to class 10 minutes or more after the start of class will be considered late. Participants who are late 3 times will receive 1 absence. Any participant who misses ¼ or more of all class meetings WILL receive an F in the course. Homework on Reading I will also check to make sure you are doing your reading homework. You will answer questions about your assigned readings and I will collect and give feedback on those questions. Presentation Each person will research and present on a different theory of Second Language Acquisition. According to Long (2007), “the literature offers as many as 60 theories, models, hypotheses, and theoretical frameworks” (p. 4) which attempt to explain the process of second language learning and acquisition. Please choose from the list below.

1) Multidimensional Model (Meisel. Clashen & Pienemann 1981) 2) Krashen’s Monitor Theory (Krashen, 1985) 3) Schumann’s Acculturation Model (Schumann, 1986) 4) Cummins’ Linguistic Interdependence Hypothesis for bilingual proficiency (Cummins, 1991) 5) Ellis’s Integrated Theory of Instructed Second Language Acquisition (Ellis, 1990) 6) Universal Grammar (White 1996, 2003a, 2003b)

7) Optimality Theory (Eckman, 2004 original described by Prince & Smolensky) 8) Bioprogram Hypothesis (Huebner, 1983 originally described by Bickerton) 9) Functional-Typological Model (Sato, 1990 originally described by Givon, 1979) 10) Accommodation Theory (Bebe & Giles, 1984) 11) Competition Model (Kilborn & Ito, 1989; MacWhinney, 1997) 12) Connectivist models of Second Language Acquisition (Gasser, 1990) 13) Socio-Educational Model (Gardner, 1988) 14) Fundamental Difference Hypothesis (Bley-Vroman, 1990 & 1997) 15) Interaction Hypothesis (Long, 1996; Gass, 1997) 16) Cognitive-interactionist (Andersen, 1989)

Final Paper The purpose of this paper is to analyze a recent academic research paper to see how it relates to the theories of language learning we have studies this semester. I have provided 10 example articles (see web site). You are free to choose your own, but please clear it with me first.

Week Topics Homework (for the next class) 1 Class Overview & Introductions

Read #1: The Nature of Theories

2 What is a theory? Key Concepts and terms

3 Key Concepts continued Read #2: Age 4 Age and language acquisition Read #3: Interaction and

Instructed Language Acquisition5 Interaction and language acquisition 6 Intro mid-term project 7 Review and project preparation 8 Presentations 9 Presentations Read #4: Saying what we

mean: Making a case for ‘language acquisition’ to become ‘language development’

10 Acquisition vs. Development 11 Acquisition vs. Development continued Read #5: Second Language

Learning Explained? SLA across 10 Contemporary Theories

12 What does it mean for language teaching? 13 Introduce final project 14 Theory and practice 15 Review 16 Final Class: Survey Final paper due

Section 2 Readings

& Homework Questions

The Nature of Theories (Please answer in full sentences and use your own words)

1. What is a theory? What is a model? How do theories and models differ? What should a good theory be able to do? Can you give your own example?

2. What is a hypothesis? How does it relate to a theory? What is a construct? How does

it relate to a theory?

3. The author summaries 10 observations about second language learning. A good theory of second language acquisition should do what with these 10 observations?

4. How can we define implicit and explicit learning? From whose perspective are the constructs about implicit and explicit learning defined?

Almost everyone has heard of Einstein’s Theory of Relativity. People have also heard of things such as the “Theory of Evolution” and “Atomic Theory.” What is common to all these theories is that they are theories about what scientists call natural phenomena : things that we observe everyday. Theories are a fundamental staple in science, and all advances in science are, in some way or another, advances in theory development. If you asked scientists, they would tell you that the sci-ences could not proceed without theories. And if you ask applied scientists (such as those who develop medicines or attempt to solve the problem of how to travel from Earth to Mars), they would tell you that a good deal of their work is derived from theoretical insights.

Theories are also used in the social and behavioral sciences, such as psychology, sociology, and economics. As in the natural sciences, social sciences attempt to explain observed phenomena, such as why people remember some things better than others under certain conditions or why the stock market behaves the way it does.

In the field of second language acquisition (hereinafter SLA) research, theories have also come to occupy a central position. Some researchers, though by no means all, would even say that the only way SLA can advance as a research field is if it is theory driven. The purpose of the present book is to introduce the reader to certain current theories in SLA and provide a background for continued in-depth reading of the same. As a starting point, we will need to examine the nature of theories in general.

What Is a Theory?

At its most fundamental level, a theory is a set of statements about natural phe-nomena that explains why these phenomena occur the way they do. In the sci-ences, theories are used in what Kuhn (1996) calls the job of “puzzle solving.”

1 INTRODUCTION

The Nature of Theories

Bill VanPatten and Jessica Williams

2 VanPatten and Williams

By this Kuhn means that scientists look at observable phenomena as puzzles or questions to be solved. Why does the earth revolve around the sun and not fly off into space? Why are humans bipedal but gorillas knuckle-walkers? These are all questions about things that confront us every day, and it is the job of scientists to account for them.

In short, then, the first duty of a theory is to account for or explain observed phenomena. But a theory ought to do more than that. A theory also ought to make predictions about what would occur under specific conditions. Let’s look at three examples: one familiar, the other two perhaps less so. In the early part of the 19th century, scientists were already aware of the presence of microorganisms in the air and water, and they had an idea about the connection between the organ-isms and disease. However, they had no idea of how they came into existence; indeed, belief in the spontaneous generation of these organisms was widespread. Disease was thought to be caused by “bad air.” Careful experimentation by Louis Pasteur and other scientists demonstrated that microbes, though carried by air, are not created by air. Living organisms come from other living organisms. These discoveries led to the development of the germ theory of disease, which proposed that disease was caused by microorganisms. The acceptance of this theory had obvious important applications in public health, such as the development of vac-cines, hygienic practices in surgery, and the pasteurization of milk. It not only could explain the presence and spread of disease, it could also predict, for example, that doctors who delivered babies without washing their hands after perform-ing autopsies on patients who had died from childbirth fever would transmit the disease to new patients. Even more important, the same theory could be used to connect phenomena that, on the surface, appeared unrelated, such as the transmittal of disease, fermentation processes in wine and beer production, and a decline in silkworm production.

Now let’s take an example from psychology. It is an observed phenomenon that some people read and comprehend written text faster and better than oth-ers. As researchers began to explore this question, a theory of individual dif-ferences in working memory evolved. That theory says that people vary in their ability to hold information in what is called working memory (defined, roughly, as that mental processing space in which a person performs computa-tions on information at lightening speed). More specifically, the theory says that people vary in their working memory capacity : Some have greater capacity for processing incoming information compared with others, but for everyone, capacity is limited in some way. Initially used to account for individual differ-ences in reading comprehension ability in a person’s first language, the theory also accounts for a wide range of seemingly unrelated phenomena, such as why people remember certain sequences of numbers and not others, why they recall certain words that have been heard, why people vary on what parts of sentences they remember best, why certain stimuli are ignored and others attended to, and why some students are good note takers and others are not. A theory of working

Introduction 3

memory, then, allows psychologists to unify a variety of behaviors and outcomes that on the surface level do not necessarily appear to be related. There are even attempts to apply the theory of SLA to explain why some people learn faster and better than others.

Let’s take a final example, this time from language. In one theory of syntax (sentence structure), a grammar can allow movement of elements in the sentence. This is how we get two sentences that essentially mean the same thing, as in the following:

(1) Mary said what? (2) What did Mary say?

In this particular theory, the what is said to have moved from its position as an object of the verb said to occupy a place in a different part of the sentence. At the same time, this theory also says that when something moves, it leaves a hidden trace. Thus, the syntactician would write (2) like (3):

(3) What i did Mary say t i ?

In (3) the t stands for the empty spot that the what left and the i simply shows that the what and the t are “co-indexed”; that is, if there happens to be more than one thing that moves, you can tell which trace it left behind.

To add to the picture, the theory also says that t s, although hidden, are psycho-logically real and occupy the spot left behind. Thus, nothing can move into that spot and no contractions can occur across it. Armed with this, the syntactician can make a variety of predictions about grammatical and ungrammatical sentences in English. We might predict, for example, that (4) is a good sentence but (5) is bad and not allowed by English grammar:

(4) Should I have done it? (5) Should I’ve done it?

The reason for this is that should has moved from its original spot and left a t behind, as illustrated in (6):

(6) I should have done it. Should i I t i have done it?

At the same time, the syntactician would predict restrictions on the contraction of want to to wanna. Thus, (7) is fine because there is no trace intervening where a contraction wants to happen:

(7) Who i do you want to invite t i to dinner? Who do you wanna invite to dinner?

4 VanPatten and Williams

All English speakers would agree, however, that (8) is awful:

(8) *Who do you wanna take Susie to the prom next month?

You could probably work this out yourself, but the reason (8) sounds bad is that the who has moved and has left behind a t that blocks a possible contraction. Compare (7) and (8) redone here as (9) and (10):

(9) Who i do you want to invite t i to dinner? Who do you wanna invite to dinner?

(10) Who i do you want t i to take Susie to the prom next month? *Who do you wanna take Susie to the prom next month?

Be careful not to pronounce wanna like want tuh ; want tuh is not a contraction and is merely the schwawing of the vowel sound in to. Want tuh sounds OK in sentence (8) precisely because it is not a contraction.

Thus, the theory unifies constraints on contractions with modals ( should, would, will, may, might ), with auxiliaries ( do, have ), with copular verbs ( be ), with the verb want, and with pronouns ( I, you, he, and so on). It makes predictions about good and bad sentences that perhaps we have never seen or heard, some of which—like silkworms and beer—don’t seem to have much in common.

To summarize so far, a theory ought to account for and explain observed phe-nomena and also make predictions about what is possible and what is not. In addition, most theories—good ones, that is—when accounting for and predict-ing things, also tend to unify a series of generalizations about the world or unify a series of observations about the world. In the brief view we had of syntactic theory, the few generalizations made about how syntax works unify a variety of observations about contractions and not just contractions with should. All contrac-tions conform to the generalizations.

For SLA, then, we will want a theory that acts like a theory should. We will want it to account for observable phenomena (something to which we turn our attention later in this chapter). We want it to make predictions. And, ideally, we want it to unify the generalizations we make as part of the theory. In other words, we want a single theory to bring all of the observed phenomena under one umbrella. Whether this is possible at this time has yet to be determined and is something that this book will explore.

What Is a Model?

Many people confuse theories and models. A model describes processes or sets of processes of a phenomenon. A model may also show how different components of a phenomenon interact. The important word here is how. A model does not need to explain why. Whereas a theory can make predictions based on generalizations,

Introduction 5

this is not required of a model. The problem is that in the real world—and in SLA as a research discipline—this distinction is not always maintained. You will find as you read further in the field that researchers often use model and theory inter-changeably. Thus, although in principle it would be a good idea to distinguish between these two terms as they do in the natural sciences, in practice many of us in SLA do not do so.

What Is a Hypothesis?

Distinct from a theory, a hypothesis does not unify various phenomena; it is usu-ally an idea about a single phenomenon. Some people use theory and hypothesis interchangeably, but in fact, they are distinct and should be kept separate. In sci-ence, we would say that a theory can generate hypotheses that can then be tested by experimentation or observation. In psychology, for example, there are theories regarding memory. You may recall the theory about working memory and capac-ity discussed earlier. The theory says (among many other things) that working memory is limited in capacity. This means that people can pay attention to only so much information at a given time before working memory is overloaded. The theory also says that there are individual differences in working memory and how people use what they have. Some people have X amount of working memory capacity as they attend to incoming information, whereas others have more or less. A hypothesis that falls out of this, then, is that working memory differences among individuals should affect reading comprehension: Those with greater working memory capacity should be faster readers or should comprehend more. This is a testable hypothesis. We ought to add here that the only valuable hypotheses for a theory are those that are testable, meaning some kind of experiment can be run or some kinds of data can be examined to see if the hypothesis holds up. Another example of a hypothesis comes from SLA: the Critical Period Hypothesis. This is a theory in neurolinguistics that states that at an early age, the brain begins to specialize; specific brain functions become increasingly associated with specific areas of the brain. In addition, some functions may be developmentally controlled; that is, they turn on and, more important for language learning, turn off at specific points in development. The Critical Period Hypothesis is a direct consequence of this theory. It states that the ability to attain native-like proficiency in a lan-guage is related to the initial age of exposure. If language learning begins after a certain age (and there is a considerable controversy over what this age is as well as whether there even is a critical period—see the various papers in Birdsong, 1999), the learners will never reach a level of proficiency or competence comparable to a native speaker’s. A corollary to this hypothesis is that language learning ability declines with age after this point. Again, both of these are testable hypotheses. Recall that earlier we said we wanted a theory to make predictions. Predictions are actually hypotheses. When we make a prediction based on a theory, we are in effect making a hypothesis.

6 VanPatten and Williams

These definitions about theories, models, and hypotheses are important because in everyday speech, we may use the term theory in a way not intended in science. For example, one might hear in a disparaging tone that something is “just a theory.” In science, the phrase “just a theory” makes no sense, as all work is theo-retically driven. What is more, the term theory has often been politicized to deni-grate particular theories (e.g., evolution) so that “just a theory” becomes a way of dismissing something that has scientific rigor but runs against some other set of beliefs. Finally, in movies and other nonscientific situations, one often hears the term theory used to mean “an idea” or a “hypothesis.” A detective trying to solve a crime might say, “I have a theory about the killer,” when that detective means, “I have an idea about the killer.” We cannot, of course, rid everyday speech of how it uses certain words. Our point in bringing up the everyday use of theory is to make sure that the reader understands the term as it is used in this book.

Constructs

All theories have what are called constructs. Constructs are key features or mech-anisms on which the theory relies; they must be definable in the theory. In the theory about disease transmission, germ is a construct. In the theory about working memory, capacity is a construct; and in the theory about syntax, a trace is a construct.

In evaluating any theory, it is important to understand the constructs on which the theory relies; otherwise, it is easy to judge a theory one way or another—that is, as a good or bad theory—without a full understanding of the underpinnings of the theory. For example, without an understanding of the construct germ, it would have been easy to dismiss germ theory. But given that the construct germ was eas-ily definable and identifiable, dismissal of germ transmission and diseases was not so facile. To fully understand something like Relativity, one must have a thorough grasp of the constructs time, space, and others.

In SLA, we find an abundance of constructs that are in need of definitions. For example, take the term second language acquisition itself. Each word is actually a construct, and you can ask yourself, “What does second mean?” “What does language mean?” and “How do we define acquisition ?” In SLA theorizing, most people use the term second to mean any language other than one’s first language. It makes no difference what the language is, where it is learned, or how it is learned. This suggests, then, that any theorizing about SLA ought to apply equally to the person learning Egyptian Arabic in Cairo without the benefit of instruction as to the person learning French in a foreign language classroom in the United States. By defining second in an all-encompassing way, it has an effect on the scope of the theory. If the construct second were not defined this way, then it would have limited scope over the contexts of language learning. For example, some people define second language to refer to a language learned where it is spoken (e.g., immigrants learning English in this country, an American learning Japanese in Osaka), whereas foreign is used to refer to situations in which the language is not

Introduction 7

spoken outside of the classroom (e.g., German in San Diego, California). Thus, if second were defined in the more restricted way, a theory of SLA would be limited to the first context of learning.

The term language is deceptively simple as a construct, but have you ever tried to define it? Does it mean speech? Or does it mean the rules that govern speech production? Or does it mean the unconscious knowledge system that contains all the information about language (e.g., the sound system, the mental diction-ary, syntactic constraints, rules on word formation, rules on use of language in context)? Thus, any theory about SLA needs to be clear on what it means by lan-guage. Otherwise, the reader may not fully grasp what the theory claims, or worse, misinterpret it.

In summary, here are key issues discussed so far:

• Theories ought to explain observable phenomena. • Theories ought to unify explanations of various phenomena where possible. • Theories are used to generate hypotheses that can be tested empirically. • Theories may be explanations of a thing (such as language) or explanations of

how something comes to be (such as the acquisition of language). • Theories have constructs, which in turn are defined in the theory.

Why Are Theories and Models Either Good or Necessary for SLA Research?

We have explored what theories are but only obliquely addressed why they might be useful. Certainly, they help us to understand the phenomena that we observe. Consider again the Critical Period Hypothesis. It has often been observed that speak-ers who begin the process of SLA later in life usually have an accent. A theory about the loss of brain plasticity during natural maturation may help explain this phenomenon. The same theory might predict that learners who begin foreign language study in high school will be less likely to approach a native-like standard of pronunciation than those learners who have access to significant amounts of target-language input much earlier in life. These kinds of predictions have clear practical applications; for example, they suggest that foreign language learning should begin at a young age.

Let’s look at another concrete example. In one theory of SLA, producing lan-guage (usually called output ) is considered an important element in structuring linguistic knowledge and anchoring it in memory. In another theory, in contrast, output is considered unimportant in developing second language knowledge. Its role is limited to building control over knowledge that has already been acquired. These differences in theory would have clear and important consequences for sec-ond language instruction. In the first case, output practice would have a significant role in all aspects of instruction. In the second case, it would be most prominent in fluency practice.

8 VanPatten and Williams

So far we have explored the utility of theories from a practical, real-world perspective. Theories are also useful in guiding research, which may not always have immediate practical purposes related to, say, instruction. If we step back for a moment and consider the theories previously mentioned, we have looked at the following:

1. a theory that explains/predicts constraints on contraction in English 2. a theory that explains/predicts foreign accents in adult learners 3. theories that predict the role of output in the second language acquisition

process

You may notice that they are not all the same. The first is a theory of what is to be acquired, that is, the unconscious mental representation of constraints on language. It is not enough to say, for example, that learners are acquiring English, for this begs the question, “What is English? How is it different from Spanish or Chinese?” Clearly, a dictionary of the English language is not the language itself, and so memorizing a dictionary is not equivalent to acquiring English. Nor would it be sufficient to study a big grammar book and commit all its rules to memory. It is very unlikely that any grammar book includes the wanna/’ve rule that appeared earlier in this chapter, for example. And what about the sound system and constraints on syllable formation (e.g., no syllable in English can start with the cluster rw, but such a syllable-initial cluster is possible in French)? In short, English, like any other language, is complex and consists of many components. You may recall that we touched on this issue when we noted that language itself is a construct that a theory needs to define. Once the theory defines what it means by language, it can better guide the questions needed to conduct research.

The second two items on the preceding list are not really about the target of acquisition; rather, they address the factors that affect learning outcomes (e.g., the Critical Period position) or they address how learning takes place, in other words, processes learners must undergo. These processes may be internal to the learner (such as what might be happening in working memory as the learner is attempting to comprehend language and how this impacts learning) or they may be external to the learner (such as how learners and native speakers engage in con-versation and how this impacts learning). Theories regarding factors or processes are clearly different from theories about the what of acquisition, but they, too, can guide researchers conducting empirical research.

Finally, research can return the favor to theorists by evaluating competing theo-ries. For example, one theory of learning, including language learning, maintains that humans are sensitive to the frequency of events and experiences and that this sensitivity shapes their learning. Within this theory, linguistic elements are abstracted from exposure to language and from language use. What look like rules in a learner’s grammar are really just the result of repeated exposure to regularities in the input. A competing theory maintains that language learning takes place

Introduction 9

largely by the interaction of innate knowledge (i.e., human-specific and universal linguistic knowledge) and data gathered from the input. Within this theory, fre-quency may have some role in making some aspects of language more “robust,” but it is not a causal factor as it is in the first theory. Each of these two theo-ries can generate predictions, or hypotheses, about how language acquisition will take place under specific conditions. These hypotheses can then be tested against observations and the findings of empirical studies.

What Needs to Be Explained by Theories in SLA?

As we mentioned at the outset of this chapter, one of the roles of theories is to explain observed phenomena. Examples we gave from the sciences were the observation that the Earth revolves around the sun and doesn’t fly off into space and that humans are bipedal while our closest relatives are knuckle-walkers. Theo-ries in science attempt to explain these observations, that is, tell why they exist.

In the field of SLA research, a number of observations have been cataloged (e.g., Long, 1990), and what follows is a condensed list of them. At the end of the chapter are references for more detailed accounts of these observations.

Observation 1: Exposure to input is necessary for SLA. This observation means that acquisition will not happen for learners of a second language unless they are exposed to input. Input is defined as language the learner hears (or reads) and attends to for its meaning. For example, when a learner hears “Open your books to page 24” in a second language, the learner is expected to comprehend the mes-sage and open his or her book to page 24. Language the learner does not respond to for its meaning (such as language used in a mechanical drill) is not input. Although everyone agrees that input is necessary for SLA, not everyone agrees that it is sufficient.

Observation 2: A good deal of SLA happens incidentally. This captures the observa-tion that various aspects of language enter learners’ minds/brains when they are focused on communicative interaction (including reading). In other words, with incidental acquisition, the learner’s primary focus of attention is on the message contained in the input, and linguistic features are “picked up” in the process. Inci-dental acquisition can occur with any aspect of language (e.g., vocabulary, syntax, morphology [inflections], phonology).

Observation 3: Learners come to know more than what they have been exposed to in the input. Captured here is the idea that learners attain unconscious knowledge about the L2 that could not come from the input alone. For example, learners come to know what is ungrammatical in a language, such as the constraints on wanna contraction that we saw earlier in this chapter. These constraints are not taught and are not evident in the samples of language learners hear. Another kind of unconscious knowledge that learners attain involves ambiguity. Learners come to know, for example, that the sentence John told Fred that he was going to sing can mean that either John will sing or Fred will sing.

10 VanPatten and Williams

Observation 4: Learners’ output (speech) often follows predictable paths with predictable stages in the acquisition of a given structure. Learners’ speech shows evidence of what are called “developmental sequences.” One example involves the acquisition of negation in English. Learners from all language backgrounds show evidence of the following stages:

Stage 1: no + phrase: No want that. Stage 2: subject + no + phrase: He no want that. Stage 3: don’t, can’t, not may alternate with no: He can’t/don’t/not want that. Stage 4: Negation is attached to modal verbs: He can’t do that. Stage 5: Negation is attached to auxiliaries: He doesn’t want that.

In addition to developmental sequences, there are such things as “acquisition orders” for various inflections and small words. For example, in English, - ing is mastered before regular past tense, which is mastered before irregular past tense forms, which in turn are mastered before third-person (present tense) - s. These stages of development also capture the observation that learners may pass through “U-shaped” development. In such a case, the learner starts out doing something correctly then subsequently does it incorrectly and then “reacquires” the correct form. A classic example comes from the irregular past tense in which learners begin with came, went (and similar forms), then may begin to produce camed, goed/wented, and then later produce the correct went, came and other irregular forms.

Observation 5: Second language learning is variable in its outcome. Here we mean that not all learners achieve the same degree of unconscious knowledge about a second language. They may also vary on speaking ability, comprehension, and a variety of other aspects of language knowledge and use. This may happen even under the same conditions of exposure. Learners under the same conditions may be at different stages of developmental sequences or be further along than others in acquisition orders. What is more, it is a given that most learners do not achieve native-like ability in a second language.

Observation 6: Second language learning is variable across linguistic subsystems. Lan-guage is made up of a number of components that interact in different ways. For example, there is the sound system (including rules on what sound combinations are possible and impossible as well as rules on pronunciation), the lexicon (the mental dictionary along with word-specific information such as verb “X” cannot take a direct object or it requires a prepositional phrase or it can only become a noun by addition of - tion and not - ment, for example), syntax (what are possible and impossible sentences), pragmatics (knowledge of what a speaker’s intent is, say, a request versus an actual question), and others. Learners may vary in whether the syntax is more developed compared with the sound system, for example.

Observation 7: There are limits on the effects of frequency on SLA. It has long been held that frequency of occurrence of a linguistic feature in the input correlates with whether it is acquired early or late, for example. However, frequency is not

Introduction 11

an absolute predictor of when a feature is acquired. In some cases, something very frequent takes longer to acquire than something less frequent.

Observation 8: There are limits on the effect of a learner’s first language on SLA. Evidence of the effects of the first language on SLA has been around since the beginning of con-temporary SLA research (i.e., the early 1970s). It is clear, however, that the first language does not have massive effects on either processes or outcomes, as once thought. ( We will review one particular theory in Chapter 2 .) Instead, it seems that the influence of the first language is somehow selective and also varies across individual learners.

Observation 9: There are limits on the effects of instruction on SLA. Teachers and learners of languages often believe that what is taught and practiced is what gets learned. The research on instructed SLA says otherwise. First, instruction some-times has no effect on acquisition. As one example, instruction has not been shown to cause learners to skip developmental sequences or to alter acquisition orders. Second, some research has shown that instruction is detrimental and can slow down acquisition processes by causing stagnation at a given stage. On the other hand, there is also evidence that in the end, instruction may affect how fast learners progress through sequences and acquisition orders and possibly how far they get in those sequences and orders. Thus, there appear to be beneficial effects from instruction, but they are not direct and not what many people think.

Observation 10: There are limits on the effects of output (learner production) on lan-guage acquisition. Although it may seem like common sense that “practice makes perfect,” this adage is not entirely true when it comes to SLA. There is evidence that having learners produce language has an effect on acquisition, and there is evidence that it does not. What seems to be at issue, then, is that whatever role learner production (i.e., using language to speak or write) plays in acquisition, there are constraints on that role, as there on other factors, as noted earlier.

Again, the role of a theory is to explain these phenomena. It is not enough for a theory to say they exist or to predict them; it also has to provide an underlying explanation for them. For example, natural orders and stages exist. But why do they exist and why do they exist in the form they do? Why do the stages of nega-tion look the way they do? As another example, why is instruction limited? What is it about language acquisition that puts constraints on it? Why can’t stages of acquisition be skipped if instruction is provided for a structure? And if instruction can speed up processes, why can it?

As you read through the various theories in this volume, you will see that cur-rent theories in SLA may explain close to all, some, or only a few of the phenom-ena. What is more, the theories will differ in their explanations as they rely on different premises and different constructs.

The Explicit/Implicit Debate

Of concern and considerable controversy in the field of SLA are the roles of explicit and implicit learning and knowledge. These concepts are notoriously

12 VanPatten and Williams

difficult to define, in part because they rest on constructs such as consciousness and awareness, which themselves have been the subject of extended scholarly debate.

Hulstijn (2005) defines the distinction in learning as follows:

Explicit learning is input processing with the conscious intention to find out whether the input information contains regularities and, if so, to work out the concepts and rules with which these regularities can be captured. Implicit learning is input processing without such an intention, taking place unconsciously. (p. 131)

Hulstijn’s definition of explicit learning appears to include both awareness of what is to be learned and the intention to learn it. Not all researchers agree. DeKeyser (2003) counts only the former as a hallmark of explicit learning and its absence as a defining feature of implicit learning, which he calls “learning without awareness of what is being learned” (p. 314). Elsewhere, Hulstijn (2003) also provides a more fine-grained distinction, noting that whereas explicit learning involves awareness at the point of learning, intentional learning additionally involves a “deliberate attempt to commit new information to memory” (p. 360). Ellis (2009a) offers a definition of explicit learning that includes intentionality, demands on attentional resources, and awareness of what is being learned and a definition of implicit learning as learning that takes place when all of these features are absent.

What is important to note about all of these definitions is the absence of instruction; that is, they present explicit/implicit learning from the viewpoint of what the learner thinks and does, not from the perspective of what the environ-ment is doing to the learner. Thus, the issue that confronts us here is not the role of instruction (that is handled by Observation 9). Instead, the focus is on what is going on in the mind/brain of the learner when that learner is exposed to L2 input (with or without instruction). Thus, the reader is cautioned not to confuse explicit/implicit learning with explicit/implicit teaching.

As we mentioned, the relative roles (or contributions) of explicit and implicit learning are debated in SLA. Does SLA fully or largely involve explicit learning? Does it fully or largely involve implicit learning? Or does SLA somehow engage both explicit and implicit learning, and if so, how, under what conditions, and for what aspects of language? On one hand, some scholars have questioned whether learning without awareness is even possible. On the other hand, others have ques-tioned whether explicit learning can ever provide the basis for spontaneous and automatic retrieval of knowledge.

Indeed, embedded within these questions about learning is the distinction between explicit and implicit knowledge. Ellis (2009b) asserts both a behavioral and neurobiological basis for this distinction. For the first, he offers “the well-attested fact that speakers of a language may be able to use a linguistic feature accurately and fluently without any awareness of what the feature consists of and vice versa” (p. 335), and for the second, “whereas implicit knowledge involves

Introduction 13

widely divergent and diffuse neural structures . . . explicit knowledge is localized in more specific areas of the brain” (p. 335). Implicit and explicit learning and knowledge are clearly related yet distinct concepts (Schmidt, 1994). Ellis (2009a) connects them by referring to the resulting representations of the two types of learning. Specifically, he claims that implicit learning leads to subsymbolic knowl-edge representations, whereas explicit learning results in symbolic representations, allowing learners to verbalize what they have learned.

Regardless of the how one defines the two types of knowledge, the major ques-tion that has challenged researchers is the nature of any interface between them. Although most scholars agree that implicit knowledge is the goal of acquisition, how does implicit knowledge develop? Can explicit knowledge become implicit? Does explicit knowledge somehow aid the acquisition of implicit knowledge? Or are they completely separate systems, which, under most conditions of SLA, do not interact?

Because the field has not yet arrived at a consensus on these questions, and because there is conflicting evidence on the relative roles of explicit and implicit learning, we cannot offer an observation like those that have preceded this section. Therefore, we are asking the contributors to this volume to address explicit and implicit learning and knowledge in a special section in each chapter, asking them to discuss what each theory or framework would claim about the two types of learning and the development of the two types of knowledge.

About This Volume

In this volume, we have asked some of the foremost proponents of particular theories and models to describe and discuss them in an accessible manner to the beginning student of SLA theory and research. As they do so, the various authors address particular topics and questions so that the reader may compare and con-trast theories more easily:

• The Theory and Its Constructs • What Counts as Evidence for the Theory • Common Misunderstandings • An Exemplary Study • How the Theory Addresses the Observable Phenomena of SLA • The Explicit/Implicit Debate

Our own interests and areas of expertise have led us to the linguistic and cog-nitive aspects of SLA. Thus, the theories and perspectives taken in the present volume—for the most part—reflect such orientations. To be sure, there are social perspectives that can be brought to bear on SLA (see Atkinson, 2011; Block, 2003). These perspectives are often offered as “alternatives” to the linguistic and cognitive orientations that are said to dominate L2 research, but in our view, they

14 VanPatten and Williams

are simply looking at different phenomena (see, e.g., the discussion in Rothman & VanPatten, 2013). In excluding such perspectives from the present volume, we do not suggest that they are unimportant for the field of SLA research as a whole. Instead, our intention is to gather those approaches that currently compete to explain the acquisition of a linguistic system (with primary emphasis on syntax, morphology, and, to a lesser degree, the lexicon). For those who seek socially ori-ented frameworks used in L2 research, we suggest using something like Atkinson’s (2011) edited volume (or parts of it) as an accompaniment to the present volume.

Discussion Questions

1. In what ways do theories affect our everyday lives? Try to list and discuss examples from politics, education, and society.

2. Discuss a theory from the past that has been disproved. Also discuss a theory from the past that has stood the test of time. Do you notice any differences between these theories in terms of their structures? Is one simpler than the other? Does one rely on nonnatural constructs for explanation?

3. Theories are clearly useful in scientific ventures and may have practical appli-cations. They have also become useful, if not necessary, in the behavioral and social sciences. In what way is the study of SLA a scientific venture rather than, say, a humanistic one?

4. Reexamine the list of observable phenomena. Are you familiar with all of them and the empirical research behind them? You may wish to consult some basic texts on this topic listed in the “Suggested Further Reading” section (e.g., Ellis, Gass, Long).

5. Is there an observable phenomenon in particular you would like to see explained? Select one and, during the course of the readings, keep track of how each theory accounts for this phenomenon.

Suggested Further Reading

Atkinson, D. (Ed.). (2011). Alternative approaches to second language acquisition. New York, NY: Routledge. This volume presents six approaches to SLA that complement or contrast with cognitive approaches to the field. Two of the approaches are represented in this volume.

Ellis, R. (2008). The study of second language acquisition (2nd ed.). Oxford, England: Oxford University Press. This volume is a comprehensive overview of the field that continues to be an excellent resource on many topics in the field.

Gass, S. (2013). Second language acquisition: An introductory course (4th ed.). New York, NY: Routledge. This is a basic introduction to the field in a form that is accessible to readers new to the field. It includes authentic data-based problems at the end of each chapter that help readers grapple with issues typical of SLA research.

Introduction 15

Hustijn, J. (2005). Theoretical and empirical issues in the study of implicit and explicit second language learning. Studies in Second Language Acquisition, 27, 129–140. This article is the introduction to a special issue on implicit and explicit learning and knowledge in SLA. As such, it provides a good overview of the issues on this topic.

Lightbown, P., & Spada, N. (2013). How languages are learned (4th ed.). Oxford, England: Oxford University Press. This volume is aimed at teachers and focuses on language acquisition in classroom settings.

Long, M. H. (1990). The least a second language acquisition theory needs to explain. TESOL Quarterly, 24, 649–666. The observations listed in this chapter are based, in part, on this seminal article.

Rothman, J., & VanPatten, B. (2013). On multiplicity and mutual exclusivity: The case for different theories. In M. P. García Mayo, M. J. Gutierrez-Mangado, & M. Mar-tínez Adrián (Eds.), Contemporary approaches to second language acquisition (pp. 243–256). Amsterdam, Netherlands: John Benjamins. This chapter, while taking a generative perspective on language, argues that different theories exist because of the complexity of acquisition, suggesting that multiple theories may be necessary to understand acquisition in its entirety.

VanPatten, B. (2003). From input to output: A teacher’s guide to second language acquisition. New York, NY: McGraw-Hill. This is an introductory volume for teachers with little background in SLA. It focuses on how input data are processed, what the linguistic system looks like and how it changes, and how learners acquire the ability to produce language, among other aspects of acquisition.

References

Atkinson, D. (Ed.). (2011). Alternative approaches to second language acquisition. New York, NY: Routledge.

Birdsong, D. (Ed.). (1999). Second language acquisition and the critical period hypothesis. Mah-wah, NJ: Lawrence Erlbaum Associates.

Block, D. (2003). The social turn in second language acquisition. Edinburgh, Scotland: Edin-burgh University Press.

DeKeyser, R. (2003). Implicit and explicit learning. In C. Doughty & M. Long (Eds.), The handbook of second language acquisition (pp. 313–348). Cambridge, England: Cambridge University Press.

Ellis, R. (2009a). Implicit and explicit learning, knowledge and instruction. In R. Ellis, S. Loewen, C. Elder, R. Erlam, J. Philp, & H. Reinders (Eds.), Implicit and explicit knowl-edge in second language learning, testing and teaching (pp. 3–25). Bristol, England: Multilin-gual Matters.

Ellis, R. (2009b). Retrospect and prospect. In R. Ellis, S. Loewen, C. Elder, R. Erlam, J. Philp, & H. Reinders (Eds.), Implicit and explicit knowledge in second language learning, testing and teaching (pp. 335–353). Bristol, England: Multilingual Matters.

Hulstijn, J. (2003). Incidental and intentional learning. In C. Doughty & M. Long (Eds.), The handbook of second language acquisition (pp. 349–381). Cambridge, England: Cam-bridge University Press.

Hulstijn, J. (2005). Theoretical and empirical issues in the study of implicit and explicit second language learning. Studies in Second Language Learning, 27, 129–140.

16 VanPatten and Williams

Kuhn, T. S. (1996). The structure of scientific revolutions (3rd ed.). Chicago, IL: University of Chicago Press.

Rothman, J., & VanPatten, B. (2013). On multiplicity and mutual exclusivity: The case for different theories. In M. P. García Mayo, M. J. Gutierrez-Mangado, & M. Mar-tínez Adrián (Eds.), Contemporary approaches to second language acquisition (pp. 243–256). Amsterdam, Netherlands: John Benjamins.

Schmidt, R. (1994). Deconstructing consciousness: In search of useful definitions for applied linguistics. AILA Review, 11, 129–158.

Age (Please answer in full sentences and use your own words)

1. What is a critical period of development? What is a sensitive period of development? How are they similar and how are they different? Is there evidence for a critical or sensitive period in first language development? If so, what is the evidence?

2. In an EFL situation, is earlier necessarily better? Why or why not? What is the

evidence?

3. How is native like attainment assessed in most studies of second language learners? What have these studies found?

4. What factors generally account for adult learners who have reached native-like attainment in the studies mentioned in this chapter?

Age

Children acquiring their first language complete the feat within a biological window offour to six years of age. By contrast, the ages at which different L2 learners may beginlearning the new language range wildly. Thus, age emerges as a remarkable site ofdifference between L2 and L1 acquisition. Perhaps for this reason, understanding therelationship between age and L2 acquisition has been a central goal since the inceptionof the field of SLA. Two issues are hotly debated. One pertains to the possibility that abiological schedule may operate, after which the processes and outcomes of L2acquisition are fundamentally and irreversibly changed. This is also known as theCritical Period Hypothesis in L2 learning. The other issue relates to the possibility thatthere may be a ceiling to L2 learning, in the sense that it may be impossible to developlevels of L2 competence that are isomorphic to the competence all humans possess intheir own mother tongue. Although the topic of age has been investigated profusely inSLA, clear or simple answers to vital questions about the relationship between age andL2 learning have not been easy to produce. As you will see in this chapter, the accruedfindings remain difficult to reconcile and interpret, and many questions to understanduniversal age effects on L2 acquisition remain open.

2.1 CRITICAL AND SENSITIVE PERIODS FOR THE ACQUISITION OF HUMANLANGUAGE

The idea that there may be an optimal, maybe even critical, age period for theacquisition of language entered SLA research through the work in neurolinguisticsof Penfield and Roberts (1959) and Lenneberg (1967). Their ideas quickly becameinfluential in a time when the new field called SLA was emerging. These authorscontributed neurolinguistic data supporting a natural predisposition in the child’sbrain for learning the first language, together with anecdotal observations thatchildren were also adept foreign language learners, when compared to adults. Thepossible causes tentatively identified at the time were the loss of plasticityundergone by human brains by year nine of life (Penfield and Roberts, 1959) orperhaps the completion by the onset of puberty of the process of lateralization, thespecialization in all right-handed individuals of the left brain hemisphere forlanguage functions (Lenneberg, 1967). The hypothesis of a critical period for L1acquisition, and as a corollary for L2 acquisition, seemed natural in the late 1960sand continues to be considered plausible today.

2

Critical and sensitive periods for the acquisition of human language 13

Indeed, critical periods have been established for several phenomena in animalbehaviour and in the development of certain human faculties, such as vision. Thehypothesis is that there is a specific period of time early in life when the brainexhibits a special propensity to attend to certain experiences in the environment(for example, language) and learn from them. That is, the brain is pre-programmedto be shaped by that experience in dramatic ways, but only if it occurs within abiologically specified time period. To be more precise, two different kinds of age-related periods for learning are typically distinguished: critical and sensitive.Knudsen (2004) offers useful illustrations of both cases from outside the SLA field,which are summarized in Table 2.1. In much of the SLA literature, nevertheless, theterms ‘critical period’ and ‘sensitive period’ are discussed as essentiallysynonymous. This is probably because the available evidence with regard to theacquisition of an additional language is still too preliminary for SLA researchers tobe in a position to make finer distinctions between the two notions (see alsodiscussion by Harley and Wang, 1997).

Table 2.1Critical and sensitive periods in animal learning, based on Knudsen (2004)

An example of a critical period is ocular representation in the cortex of kittens. This neurological process

develops according to a narrow window of opportunity between 30 and 80 days of life. If kittens are deprived

from the experience of viewing during this time window (because one eye is forced to remain closed), they

will lose vision, simply because the closed eye and the brain failed to connect, as it were. That is, even

though the now uncovered eye is optically normal, it fails to convey the visual information to the axons in the

thalamus, which in turn cannot convey it to the neurons in cortical level IV. Another well-known example of a

critical, irreversible period is filial imprinting in the forebrain of ducks, which makes them follow the first

large moving entity between 9 and 17 hours of hatching and bond with it as the recognized parent. An

example of a sensitive period, on the other hand, is barn owls’ ability to process spatial information

auditorily (indispensable for catching mice in the dark!). Young owls develop the ability to create mental

maps of their space based on auditory cues at a young age. If either hearing or vision is impaired during this

sensitive period, auditory spatial information will not be processed normally later in life. However, problems

can be compensated for and reversed, even well past the sensitive period, if the visual or auditory

impairment is restored and rich exposure to sound input is provided

The evidence for a critical or sensitive period for first language acquisition inhumans is strong, although it remains far less well understood than the criticalperiod for the development of, for example, ocular representation in our cortex.Some of the evidence comes from research involving sadly famous cases of childrenwho, due to tragic circumstances, were deprived from regular participation inlanguage use and social interaction until about the age of puberty. This was the caseof Genie (recounted from different perspectives by Curtiss, 1977; Rymer, 1993) andof several feral children (discussed in Candland, 1993). Under such seriouslydetrimental circumstances, these adolescents could not learn the mother tongue tothe level of their peers, even after they were rescued and efforts were made to ‘teach’them language.

Age

Additional evidence comes from studies of so-called postponed first languageacquisition in the deaf population. This case, although unfortunate, accounts forabout 90 per cent of deaf babies who are born to hearing parents with knowledge ofonly an oral language. In essence, these babies grow up without experiencingregular exposure to either spoken or sign input for the first six years of life and,therefore, they learn no first language until they are sent to school, precisely ataround that age. The findings contributed over the years by Canada-basedresearcher Rachel Mayberry (2007; Mayberry and Lock, 2003) suggest that thesechildren will usually exhibit incomplete acquisition of their late-learned firstlanguage. A study in the United States by otolaryngologists Mario Svirsky andRachael Holt (2005) augments this evidence. These researchers tracked thevocabulary, grammar and speech perception abilities of 96 deaf babies who hadreceived cochlear implants at ages one through four. These babies’ spoken languagebegan developing once awareness of sound was made possible through theimplants. However, babies who received the implant after the age of 2 exhibitedslower progress and overall lower performance in vocabulary and grammar (butnot in speech perception skills), compared to babies who had their hearing restoredbefore the end of the second year of life. The researchers interpreted the evidencecautiously but suggestively as indication that a sensitive period for L1 acquisitionexists, and one that is much shorter than once thought (see also Svirsky et al.,2007). In light of the new cochlear implant evidence, we can speculate that it mayend after the first two years of life, at least for some aspects of acquisition.

For L2 acquisition, as well, it seems plausible to posit that there are sensitiveperiods for a number of language areas. But what does the record of SLA researchtell us?

2.2 JULIE, AN EXCEPTIONALLY SUCCESSFUL LATE L2 LEARNER OFARABIC

Striking SLA evidence bearing on the sensitive period hypothesis for L2 acquisitioncomes from a study conducted by Georgette Ioup and her colleagues (Ioup et al.,1994). These researchers investigated the limits of ultimate attainment achieved byJulie, an exceptionally successful L2 user. The study is unique, as you willappreciate, because Ioup et al. employed a rich case study methodology that yieldedin-depth knowledge of Julie’s L2 learning history as well as a wealth of data probingseveral areas of her L2 competence.

Julie was an L1 speaker of British English who had moved to Egypt at the age of21 due to marriage to an Egyptian. She settled in Cairo with her husband, became ateacher of English as a foreign language (EFL) and had two children. Julie had neverreceived formal instruction in the L2 and could not read or write in Arabic. Yet, shewas able to learn Egyptian Arabic entirely naturalistically and regularly passedherself off as a native speaker. In fact, her family and friends remembered she wasable to do so just after two and a half years of residence in the country. Accordingto Julie, Arabic became the dominant language at home after the third year of

14

Julie, an exceptionally successful late L2 learner of Arabic 15

residence in Cairo (although she also reports her children grew up to be competentbilinguals). Julie had been living in Egypt for 26 years at the time of the research.

In order to test whether her accent was truly indistinguishable from that of anative speaker, the researchers mixed Julie’s recorded explanation of her favouriterecipe with recordings of the same oral task by another six native and advancednon-native speakers of Arabic. Seven of 13 judges rated Julie as definitely native,whereas the other six rated her as non-native, and commented that a few smalldifferences in vowel and consonant quality and in intonation had given her away.In another task testing her speech perception abilities, Julie proved herself able topick out Egyptian from non-Egyptian accents among seven different varieties ofArabic with 100 per cent accuracy. She was a little bit less adept at discriminating aCairo-sounding Egyptian accent from two other Egyptian regional accents, but sowere six of the 11 native-speaking judges.

In order to probe her tacit knowledge of the Arabic language, Julie, 11 L1 Arabiccontrol participants and another very advanced non-native speaker of Arabic wereasked to do three other tasks that tested morphosyntactic phenomena. The firsttask was translating 12 sentences from English into Arabic. Here, once again, Juliemade very few mistakes. For example, she translated ‘went to the club’ as raahitlinnaadi, without dropping the preposition l- (Ioup et al. explain that this isnecessary in Arabic to indicate the meaning is not just ‘to’ but ‘into’; p. 82). Thesecond task involved judging the grammaticality of selected Arabic sentences.Julie’s judgements diverged from those of the majority of native speakers on onlyfive out of 37 sentences. Apparently, she preferred the unmarked word order choicefor questions and rejected variable word order alternatives that are alsogrammatical in Arabic. In the third task, Julie and the others had to answer thequestion ‘who did X’ in response to 18 recorded sentences containing cases ofanaphora, a syntactic phenomenon so subtle that it is unlikely to be learnedthrough explanations or through conscious analysis of the input. Anaphora refersto the binding of a pronoun to the right preceding noun in a sentence. For example,who does she refer to in the following sentence (p. 89)?

(1) ‘Nadia saw Mona when she entered the room’a. Nadya shaafit mona heyya daxalit il-oodab. Nadya shaafit mona lamma daxalit il-ooda

Who entered the room – Mona or Nadia? The preferred answer would be ‘Nadia’for the English sentence. In Arabic, if the pronoun heyya is used, as in (1a), thepreferred interpretation is that ‘she’ refers to Nadia, the more distant referent.Conversely, if heyya is omitted, as in (1b), the preferred interpretation for ‘she’will be Mona, the closest referent outside the embedded clause ‘when sheentered the room’. Julie was able to correctly interpret the anaphora pronouns intwo-thirds of the 18 items. She performed less well on the remaining six items,which involved relative clauses such as ‘Ahmad bought the dress for the girl thatyou went to the lady that she angered’ (ah.mad ishtara il-fustaan li-l-bint illi intaruh.t li-s-sit illi heyya’za alit-ha, p. 90). Not only Julie, but the 11 native speakers

Age

too, found it difficult to answer the question ‘who angered whom’ after theyheard this sentence! However, only Julie went on to answer in a way that wouldmean she interpreted the overt pronoun heyya to refer to the closest referent inthe sentence.

In their study, Ioup et al. also included another exceptionally successful late adultlearner, Laura, who like Julie was married to an Egyptian and typically passed off asa native speaker. Unlike Julie, however, she held a Master’s degree in modernstandard Arabic from a US university and was a teacher of standard Arabic at auniversity in Cairo at the time of the study. Laura performed by and large as well asJulie in all tasks except for the speech perception one. In other words, she was alsoexceptionally successful. However, Ioup et al. concentrate on Julie and leave Laurain the background of their report, perhaps because Laura was an instructed learnerand hence many more factors come into consideration. Julie, by being a purelynaturalistic late learner, provides a strong test case for the Critical PeriodHypothesis. Or rather, some would say, against it! In the end, it is difficult toevaluate what the small degree of variability in Julie’s L2 outstanding performancemeans, particularly given that there was definitely some variability for nativespeaker responses across all measures too. Interestingly, Ioup (2005) herselfbelieves the preponderance of evidence supports the existence of age-relatedsensitive periods for L2 learning.

2.3 ARE CHILDREN OR ADULTS BETTER L2 LEARNERS? QUESTIONS OFRATE

We all tend to think that children pick up languages speedily and effortlessly. Likemany apparently undeniable truths (e.g. the earth does look very flat to our plaineyes, after all!), this assertion was questioned once it was submitted to systematicinvestigation. When in the 1970s several SLA researchers compared children andadults’ L2 learning rate in second language environments, the findingsunexpectedly were suggestive of an advantage for adults over children. Forexample, in two oft-cited studies conducted in the Netherlands, Catherine Snowand Marian Hoefnagel-Höhle (1977, 1978) found that adults and adolescents werebetter than children in terms of what they could learn in a 25-minute instructionsession or up to a year of naturalistic exposure to L2 Dutch. Although the advantageof the older learners began diminishing after ten months or so, the findings weresurprising because they flatly contradicted assumed critical period effects.

In a seminal article, Stephen Krashen, Michael Long and Robin Scarcella (1979)put a grain of salt on these findings. They concluded that older is better initially, butthat younger is better in the long run. They based this conclusion on a review of 23studies of L2 learning in second language contexts published between 1962 and1979, comprising the available findings at the time. The 18 studies that involvedshort-term comparisons (with lengths of L2 exposure in a second languageenvironment between 25 minutes and one year, rarely up to three years) suggestedthat adult learners and older children learned at a faster pace than younger

16

Age and L2 morphosyntax: questions of ultimate attainment 17

children. This may have been in part an artefact of instruction or tests thatdemanded cognitive maturity and involved metalinguistic skills, because adultsmay be able to use cognitive and metacognitive abilities and strategies to learnmany aspects of the L2 initially faster. Findings were available also from five long-term studies, however, of which the most widely cited to date were dissertationsconducted at Harvard University and New York University, respectively, by SusanOyama (1976) and Mark Patkowski (1980). The five long-term studies revealed thatwhen accomplishments in the L2 were compared after at least five years ofresidence in the L2 environment (often after ten or 20 years, and for someparticipants up to 61 years), then young starters were clearly better than adultstarters. Long (1990) reassessed the evidence on rate and ultimate attainment adecade later and reiterated the same conclusions, arguing that the rate advantagefor adults dissipates after a little more than a year, because children eventuallyalways catch up and surpass late starters. More recent findings from a study byAoyama et al. (2008) also lend support to the same conclusion.

Age findings gleaned in foreign language contexts in the last few years, however,have complicated this picture (see studies in García Mayo and García Lecumberri,2003; Muñoz, 2006). Particularly in the context of English learned in Cataluña(Muñoz, 2006), when early starters studying English from the age of eight to 16were compared to late starters studying English from the age of 11 to 17, the latestarters actually maintained an advantage that persisted well after five years ofinstruction (seven and nine years, respectively). That is, younger starters do notappear to catch up in these foreign learning contexts, where the L2 is only availablethrough instruction. This is actually not surprising if we remember that the sametime length of five years entails an intensity and quality of exposure to the L2 thatcan be radically different in foreign versus second language learning contexts. Atthree hours a week by nine months of school a year, students enrolled in a foreignlanguage in school may experience as little as 540 hours of actual instruction and L2exposure over five years. By contrast, in the same chronological time window,learners in L2 environments may accrue about 7,000 hours of L2 exposure (if wecalculate a conservative four hours a day). (A sobering comparison is that childrenlearning their L1 may receive of the order of 14,000 hours of exposure, also basedon a conservative estimate of eight hours a day!) Thus, as Singleton (2003) suggests,in foreign language contexts considerably more than five years would be needed tocapture any lasting differences between differing starting ages. Age may exertuniversal influences on the learning of a second language, but context moderatesthese universal effects and needs to be considered carefully.

2.4 AGE AND L2 MORPHOSYNTAX: QUESTIONS OF ULTIMATEATTAINMENT

Contemporary SLA researchers interested in the question of critical or sensitiveperiods now consider it essential to take a long-term view on ultimate attainmentand to evaluate the end state of L2 acquisition (Long, 1990; Hylstenstam and

Age

Abrahamsson, 2003). Two lines of recent research have investigated this question,both focusing on the area of L2 morphosyntax.

The first line is correlational. That is, it uses statistical analyses to determine thedegree to which two sets of numbers (age and scores on some L2 test) co-vary orbehave in a similar pattern. Building on the pioneering studies by Oyama (1976) and

18

Table 2.2L2 morphosyntactic knowledge along the age of onset continuum

Studies comparing knowledge of morphosyntax associated with varying ages of onset for L2 acquisition.

• Main research question: Are age and morphosyntactic attainment systematically related?

• Researchers’ interpretation of results: Johnson and Newport interpreted their data in support of a

critical/sensitive period. Birdsong and Molis’s replication did not support all of the original findings.

Johnson and Newport (1989) 46 L1 Chinese and Korean adult speakers of L2 English who were

college educated and had been living in the US for at least five years

took a 276-item grammaticality judgement task. There was a

statistically significant negative correlation of r = –0.77 between age of

arrival (which ranged from 3 to 39) and grammaticality judgement

score. The correlation was larger when only the 3 to 15 age group was

examined (r = –0.87) and disappeared for the 17 to 39 group. The

youngest group (3 to 7 years old when they arrived in the US) scored

within the range of the NS control group, the adolescent group (who

had arrived between 8 and 16 years of age) showed scores linearly

declining with age and the group of adults (who had arrived at between

17 and 39 years of age) scored variably, without age holding any

systematic relationship with their grammaticality intuitions. One late

arrival (at age 23 upon arrival) scored 92% accurate, as high as native

speakers

Birdsong and Molis (2001) Exact replication of Johnson and Newport: 61 college-educated L1

Spanish speakers of L2 English took Newport and Johnson’s

grammaticality judgement task. The 29 early acquirers had arrived in

the US between age 3 and 16 and had a mean length of residence of

12.2 years in the L2 environment. The 32 late acquirers had arrived at

age 17 or older and had a mean length of residence of 10.5 years.

There was a statistically significant negative correlation of r = –0.77 for

the full sample between age of arrival and grammaticality judgement

score. The early arrivals exhibited no variation, as they obtained near-

perfect scores. There was a statistically significant negative correlation

of r = –0.69 between age of arrival for the late arrival speakers and

grammaticality judgement score. 13 late arrivals obtained 92% or

higher scores, 3 above 95%. (The study also examined reported

amount of L2 use. Amount of current L2 use was strongly related to

judgements)

Age and L2 morphosyntax: questions of ultimate attainment 19

Patkowski (1980), it looks at L2 learners sampled to represent a wide range of agesof arrival in the L2 environment, as early as 0 years of life and as late as 50. In mostof these studies, the target language investigated is English. The key question askedis: Are age and morphosyntactic attainment systematically related? Ultimate attain-ment is usually measured by comparing L2 learners’ responses on grammaticalityjudgement tasks along differing arrival ages and against a native speaker baseline.The accumulated findings suggest that, by and large, learners who began acquiringthe L2 before a certain age, which these studies locate to be around puberty, will tendto exhibit intuitions that are very close to those of native speakers of that language.The late learners’ intuitions, by way of contrast, are not likely to be in the nativespeaker range, and this holds true regardless of the number of years, since theyarrived in the L2 environment past puberty. In the details, however, the evidencepresents a consistent dissonance. A glimpse of this dissonance can be seen in Table2.2, which summarizes a study of this kind conducted by Jacqueline Johnson andElissa Newport that has become a classic, and a replication by David Birdsong andMichelle Molis that failed to support the original findings.

Two results in Table 2.2 are noteworthy. First, in Johnson and Newport (1989)the relationship between age and grammatical intuitions abruptly disappears afteraround puberty, whereas in Birdsong and Molis (2001) grammaticality scores keepgradually declining across all ages beyond puberty. Second, both studies turned upone or more learners who had begun to learn the L2 as adults but scored within thenative speaker range. These two patterns recur in a number of other partialreplications of Johnson and Newport. For example, DeKeyser (2000) producedfindings that resonate with those of Johnson and Newport (1989) with 57Hungarian US immigrants but Flege et al. (1999) obtained a pattern of resultssimilar to that of Birdsong and Molis (2001) with 240 Korean permanent residentsin the United States.

The second parallel line of work on age differences and L2 morphosyntacticdevelopment pertains to studies that are specifically designed to investigate theupper limits of successful late L2 learning. The focus is, like in Ioup et al. (1994), onexceptional learners who seem to have reached native-like ultimate attainment,often with L2s other than English. Even though these cases have traditionally beenconsidered purely exceptional, Birdsong speculates that they may actually accountfor as much as 5 per cent to 25 per cent of cases of learners who are given ‘a fairchance of success’ (Birdsong, 1999b, pp. 14–15). The method of choice has been tocompare their performance on grammaticality judgement tasks to that of native-speaking controls, sometimes using retrospective interviews to probe learners’explanations of their choices. The question at stake in this kind of study is: Hownative is really ‘near-native’? Table 2.3 summarizes two oft-cited empirical studiesof this kind, both with highly successful L2 French learners. R. Coppieters (1987)found strong evidence for the critical period but David Birdsong’s (1992)replication of the same study found strong evidence against it. Interestingly, severalstudies, using a similar line of evidence but slightly different designs, have alsoreplicated this pattern of conflicting findings. For example, Sorace (1993) foundevidence favouring a critical period for morphosyntax among 44 L2 Italian learners,

Age

but both White and Genesee (1996), with 89 L2 English learners, and Montrul andSlabakova (2003), with 64 L2 Spanish learners, found evidence against it.

The empirical dissonance illustrated in Tables 2.2 and 2.3 is persistent. Simplyput, doubts as to whether a critical period for L2 learning really exists will not goaway as long as studies continue to show that there is no sharp drop in grammaticalintuitions after some supposedly critical age, and as long as cases of exceptionallysuccessful late learners are discovered against the backdrop of the overwhelmingtendency for early starters to ‘succeed’ and for late starters to ‘fail’.

2.5 EVIDENCE ON L2 MORPHOSYNTAX FROM COGNITIVE NEUROSCIENCE

Knowledge about how the brain handles language is highly relevant to anydiscussions of critical periods for L2 learning. For a decade now, the new field of

20

Table 2.3 Differences between near-native and native morphosyntactic knowledge

Studies scrutinizing knowledge of morphosyntax in L2 speakers who achieve exceptionally high levels of

ultimate attainment and are identified as near-native outside the laboratory (typically, cases of L2 acquisition

after puberty)

• Main research question: Can some exceptionally successful L2 acquirers be indistinguishable from native

speakers in their morphosyntactic knowledge?

• Researchers’ interpretation of results: Coppieters: No, near-native speakers’ L2 knowledge is different from

true monolingual native speakers. Birdsong: Yes, some rare, exceptional near-native speakers cannot be

distinguished from native speakers even under tight laboratory scrutiny

Coppieters (1987) 21 L2 French speakers, all of whom were highly successful and

educated French users who had begun learning the L2 after puberty.

They did a grammaticality judgement task and were interviewed. Their

average grammatical intuitions on the task were three standard

deviations away from the average of native speaker controls. Their

rationalizations for their judgements during the interview were different

from those of native speakers. Subtle syntactic-semantic and

morphosemantic differences of knowledge distinguished nativeness

from near-nativeness

Birdsong (1992) Partial replication of Coppieters: 20 L2 French speakers all of whom

were highly successful and educated French users who had begun

learning the L2 after puberty. Their age of arrival in France was 19 to

48 and their L1 was English. Of these, 15 participants performed on a

grammaticality judgement task within the native speaker range. There

was a negative correlation of r = –0.51 between age of arrival and

scores in this late-starter-only sample

Evidence on L2 morphosyntax from cognitive neuroscience 21

cognitive neuroscience has contributed interesting evidence on the issue. The dataare elicited with neuroimaging techniques such as event-related potentials, whichoffer excellent temporal resolution and make it possible to measure in millisecondsthe activation patterns of neural networks involved in different cognitiveoperations while the brain is processing language stimuli. The converging findingsfavour a critical period interpretation for L2 morphosyntax.

Some researchers have shown that localization of language functions in the brainis less lateralized in late bilinguals (more right hemisphere activation is observed)than in early bilinguals and monolinguals. This is the conclusion supported byresearch conducted in France by neuroscientists Stanislas Dehaene and ChristophePallier (see Dehaene et al., 1997; Pallier et al., 2003). Likewise, Helen Neville and herlab in the United States have produced evidence that, when engaged in certainkinds of L2 syntactic processing, the bilingual brains of people who began learningtheir L2 later in life (eight years or older in most of these studies) show cleardifferent activation patterns from those of monolingual and early bilingual brains.Such age-related differences disappear when brain activation is inspected duringthe processing of L2 semantic stimuli. For example, Weber-Fox and Neville (2001)investigated Chinese–English bilinguals who were first immersed in the L2environment anywhere between age one and past age 15. Those bilinguals whowere exposed first to the L2 after year seven processed closed-class words (i.e.function words like with, the, some) differently from the early bilinguals and themonolingual controls, whereas open-class words (i.e. content words like nose,stored, glad) yielded no major differences in brain activation patterns acrossgroups. Germany-based researcher Anja Hahne (2001) also found that herRussian–German bilingual participants, all of whom had learned German as theirL2 after the age of 10, processed syntactically anomalous sentences of the kind DasGeschäft wurde am geschlossen (‘The shop was being on closed’) statisticallysignificantly different from monolinguals, whereas no differences were foundbetween the two groups with semantically anomalous sentences. Based on suchfindings, these and other neurocognitive researchers (e.g. Ullman, 2001) havesuggested that the learning of syntactic functions (in the L1 or the L2) isfundamentally different from the learning of semantic features. Specifically, theypropose that syntax involves computational learning mechanisms and isconstrained by a biological schedule, and that semantics draws on associativelearning mechanisms and is free from critical period constraints.

Other neurocognitive scientists, however, offer different interpretations of thefindings. For example, in the United States Lee Osterhout and colleagues(Osterhout et al., 2002) demonstrated that the different neural activation patternsuncovered for function versus content words could be also explained by thedifferential word length in both kinds of stimuli (content words are typically longer,and this naturally can affect processing). Italian researchers Daniela Perani andJubin Abutalebi (2005) suggest that it is not the age of onset but the degree of activeuse of the L2 that matters when explaining degrees of brain activation. They arguethat the neural systems serving L2 and L1 grammatical processing are the same,and that higher attained proficiency and higher daily exposure to the L2 are

Age

independently correlated to lower activation patterns. Furthermore, they report onstudies that show that, even when attained proficiency is kept constant, the brainsof L2 speakers who have less daily exposure to the L2 exhibit higher degrees ofactivation in the left prefrontal cortex. They claim this parallels the generalneurocognitive finding that increased practice leads to lower levels of neuralactivation, because with more practice the same processing task will consume lessresources (see Chapter 5, section 5.2). Along the same lines of reasoning, Osterhoutand colleagues (Osterhout et al., 2008) have initiated a research programme thatinvolves measuring the brain activity of zero-level beginning learners while theyprocess L2 stimuli, longitudinally as they progress through their regular college-level foreign language courses. They have found that brain activation patterns canchange in degree and location just after experiencing about four months or 80hours of college instruction. At least for certain L2 forms, the brain’s activationpatterns become similar to the patterns observed in fluent L1 users.

Thus, the neuroscience findings, although fascinating, should not be interpretedhastily because, at this early stage of our knowledge about brain and language, it isdifficult to evaluate what they may mean for the critical period discussion. AsMarinova-Todd et al. (2000) pointed out, given what we know about the plasticityof the brain, any age-related differences in brain location and neural activitypatterns may be as much a result of the brain’s architecture shaping howsubsequent linguistic experience is processed and used for L2 learning, as it couldbe the result of the brain having been shaped by previous experience. Evidence infavour of a critical period explanation will come only when neuroscientists canestablish beyond doubt that the former, and not the latter, is actually the case.

2.6 L2 PHONOLOGY AND AGE

Unlike subtle morphosyntactic knowledge, which may be difficult to evaluate outsidethe laboratory, foreign accents are so conspicuous that they can often be detected bythe untrained ear. Thus, we all tend to think that, if there are sensitive periods forsome areas of L2 learning but not others, then phonology must be one of those areas.

This is Tom Scovel’s (1988, 2000) position. For Scovel, speech has a special statuswhen it comes to critical periods because ‘pronunciation is the only part of languagewhich is directly “physical” and which demands neuromuscular programming’(1988, p. 62). For example, one-third of the human brain’s cortex is dedicated to con-trolling motor skills in the lower face, lips, tongue and throat, all involved in the pro-duction of speech. After reviewing a large number of early studies of foreign accentdetection in his 1988 seminal book, he concluded that, in study after study, non-nativespeaking samples were consistently and accurately detected by native-speaker judges.Likewise, the evidence accumulated since 1988 overwhelmingly shows that foreign-sounding accents are likely to develop when the L2 is first learned later in life.

The study of 240 L1 Korean speakers of L2 English by James Flege and colleagues(briefly mentioned in section 2.4) also lends support to Scovel’s suggestion thatspeech is different. Flege et al. (1999) found that the two moderating variables of

22

What causes the age effects? Biological and other explanations 23

self-reported amount of L2 use and amount of education in the L2 were morerelated to the morphosyntactic results than to the pronunciation results. Thispattern of findings suggests that acquisition of phonology may be more imperviousto non-biological influences such as L2 use and education, and therefore morestrictly tied to biological schedules, than other areas of the L2. Nevertheless, Flege(e.g. 1987, 1999) proposes an explanation for the observed age-related phonologyeffects that is remarkably different from Scovel’s.

According to Flege, phonetic categories or mental representations of speechsounds in the L1 are stabilized by age five to seven. After that point, new phoneticcontrasts will be processed through such an L1 filter, and hence it is more difficult,although not biologically impossible, to detect and produce L2 categories that arenot salient. Ironically, then, foreign accents may arise ‘not because one has lost theability to learn to pronounce, but because one has learned to pronounce the L1 sowell’ (Flege, 1999, p. 125). The older people are when they begin learning an L2, themore settled they may be in their L1 perceptions. In other words, instead of viewingneurophysiological maturational constraints as the main explanatory factor for thedevelopment of L2 phonology, as Scovel does, or as a result of neurofunctionalreorganization during development, as cognitive neuroscientists do, Flege puts theexplanatory emphasis on psychoperceptual and phonetic causes related to previousmassive experience with the mother tongue.

We have said that there is clear evidence that accents are likely to develop whenthe L2 is first learned later in life. This notwithstanding, and paralleling the findingsfor L2 morphosyntax, there is also evidence in L2 phonology of exceptional post-pubertal learners whose accents are not recognized as foreign even under closescrutiny in the laboratory. Julie and Laura (Ioup et al., 1994) were the first cases, butseveral more have emerged by now. Notably, Theo Bongaerts and his colleagues inthe Netherlands have produced a number of such studies involving very advanced,late L2 learners of English and, in subsequent replications, of French and Dutch(see Bongaerts, 1999). These exceptional learners shared two features. They had allreceived considerable amounts of high-quality L2 instruction and they all self-reported high levels of motivation and concern to sound native-like. Although herresults are less dramatic, Alene Moyer (1999) also found that judges did identify asnative the accent of one of 24 advanced L2 German users in the United States, all ofwhom had begun learning the L2 after the age of 12. In sum, in L2 phonology as inL2 morphosyntax, it is not impossible (although it is admittedly rare) to attainnative-like levels. Indeed, it is remarkable that the feat has been attested with someexceptionally successful late L2 learners for target languages as different as Arabic,Dutch, English, French and German.

2.7 WHAT CAUSES THE AGE EFFECTS? BIOLOGICAL AND OTHEREXPLANATIONS

As you can surely appreciate by now, the interpretation of the evidence on age-related effects on L2 learning is far from being settled. For one thing, the

Age

interpretation of findings from correlational data on morphosyntax is subject to anumber of methodological criticisms, particularly a discussion about whether anabrupt drop-off at around puberty or a gradual decline across ages is beingobserved in the data (see Table 2.2; see also discussion in Birdsong, 1999a, 2006).In addition, the critical period explanation does not sit well with the fact that onecan always find exceptional learners who began learning the L2 after puberty, oftenin their twenties, and who perform within the range of native-speaker controls intheir grammatical intuitions or go undetected as non-native speakers by multiplejudges.

Furthermore, it is possible to conclude that age-related differences exist in how askill or ability is learned, and to propose explanations that do not invoke pre-programmed biological changes in the brain as an underlying cause. One suchexplanation lies with previous and entrenched knowledge of the L1 and L1–L2interactions, instead of biology. In L2 phonology, as we saw, James Flege takes sucha position. In this view, ‘age is an index of the state of development of the L1 system.The more fully developed the L1 system is when L2 learning commences, the morestrongly the L1 will influence the L2’ (Piske et al., 2001, p. 196). Other SLAresearchers emphasize general socio-educational and motivational factors inconnection to age effects on L2 learning. This is the position espoused by EllenBialystok and Kenji Hakuta (1999) and by Catherine Snow, Stefka Marinova-Toddand their colleagues (Marinova-Todd et al., 2000), among others. They argue thatsocio-educational and motivational forces are so radically different in the lives ofadolescents and adults, when compared to children’s lives, that languageattainment differences can be expected, but they are probably a consequence ofexperience and socialization, and not biological or insurmountable in nature.

Other SLA researchers argue that the posited sensitive period (or periods) isindeed real (e.g. Hylstenstam and Abrahamsson, 2003). In their view, there is a notwell-understood but nevertheless biologically determined impossibility, after a cer-tain age, to continue using the implicit learning processes that are best suited fornatural language learning during the early years of human life (e.g. DeKeyser, 2003;Ioup, 2005). This kind of explanation is compatible with Robert Bley-Vroman’s(1990) well-known Fundamental Difference Hypothesis, which posits that theacquisition process undergone by children and adults is fundamentally differentbecause children posses the innate ability to intuit the L1 grammar, whereas adultshave lost this ability and thus need to resort to problem solving and conscious atten-tion to handle L2 learning (see also further discussion in Chapter 7, section 7.9).

While those scholars who favour the critical period position may turn out to beright, thus far they have been unable to produce a clear answer as to what biological,irreversible changes may cause the brain to use implicit processes when learninglanguage up until a certain age but not later. As mentioned in section 2.1,lateralization and plasticity are neurobiological processes that have beenconsidered by various researchers. Another process is myelination, or thedevelopment during the first 10 or 12 years of life of white-matter substance aroundthe brain’s nerve fibres which protects the nerves and enables faster conduction ofinformation across nerve cells (Pulvermüller and Schumann, 1994; Pujol et al.,

24

A bilingual turn in SLA thinking about age? 25

2006). An additional suggestion is pubertal increases in estrogen or testosterone(Ullman, 2004). To date, however, we lack sufficient evidence for neurological orneurochemical correlates that can support unequivocally a critical periodsexplanation (see Singleton, 2005, for more extended discussion).

In sum, it would be premature to proclaim that critical periods for L2 learningexist when so much discordant evidence keeps emerging across relatively diversebodies of research. The preponderance of evidence suggests that late and adult L2acquisition generally results in lower levels of ultimate attainment and moreindividual variability than is observed for L1 and very early L2 acquisition.However, the field is of two minds as to whether critical periods for L2 acquisitionexist. Age effects on L2 learning are pervasive and undisputed, but satisfactoryexplanations, biological or otherwise, for the observed effects are yet to beconclusively produced.

2.8 A BILINGUAL TURN IN SLA THINKING ABOUT AGE?

Matters surrounding age effects on L2 learning have become even more complex inrecent SLA discussions, as two threads of evidence have become available to theresearch community.

The first new thread of evidence pertains to the realization that age effects maybe present in additional language acquisition much earlier than previouslythought, perhaps by age four. The claim is far from conclusive but appears to bereasonably promising because it converges out of diverse research programmes.Thus, Hyltenstam and Abrahamsson (2003) noted that small but important mor-phosyntactic differences are detected in the written and spoken performance ofextremely young L2 child starters, if researchers take care to recruit participantswho began L2 learning at such early ages (e.g. below 6) and if they employ fine-grained elicitation and analysis procedures. Likewise, in L2 phonology Flege etal. (1995) found very young L2 starters who did not attain the native-like levelsof pronunciation we all assume of pre-pubertal children learning other lan-guages. In the neighbouring field of bilingual research, similar findings havebeen emerging for some time. For example, Catalan researcher Núria Sebastián-Gallés and her colleagues (Sebastián-Gallés et al., 2005) investigated lexical rep-resentations of early versus simultaneous bilinguals by asking them to tell apartCatalan L2 words and non-words on a lexical decision task. They found that par-ticipants who had started to be exposed to Catalan at age 4 or earlier, but notfrom birth, did less well on this lexical decision task than participants for whomboth Catalan and Spanish were available from birth. As you will remember froman earlier discussion in this chapter, the case of postponed first language devel-opment among deaf babies after cochlear implants (Svirsky and Holt, 2005) isconsistent with the view that there may be extremely early age effects for (firstand second) language learning. Most of the SLA findings and associated explana-tions for age effects in L2 learning have been generated under the assumption ofa much later biological window, namely around age 6 for phonology and perhaps

Age

around puberty or up to age 15 for morphosyntax. If age effects do set in asextremely early in life as age two or four, the long-held assumption that an earlystart guarantees complete and successful L2 acquisition loses much of its power.Moving the onset of age effects into the very first years of life also blurs the tra-ditional distinction between L2 and bilingual learners. Thus, a new range of the-oretical and empirical arguments in SLA may have to be considered in thefuture, and SLA researchers may need to turn to the study of bilingualism whenreassessing the evidence.

A second recent realization for which increasing evidence is mounting is that theactual relative amounts of L2 and L1 use at the time of study may be central to thetask of gauging age effects. This is the so-called issue of language activation (alsocalled language dominance in bilingual studies) (see Birdsong, 2005; Perani andAbutalebi, 2005). For example, findings offered by Flege and MacKay (2004)suggest that young starters who do not live up to expectations of complete successin L2 pronunciation may present low L2-use profiles and spend much of their timeusing the L1. The study by Sebastián-Gallés et al. (2005) also revealed a similareffect for amount of L2/L1 use, even in the simultaneous (from birth) bilingualsthey studied. Their participants did best in the language they were more activelyand consistently using in daily activities at the time of study. When the putativecritical age is pushed back to a much earlier point in life, and the age effects turn outto be entangled with language activation and practice effects, it becomes imperativeto re-evaluate the extant evidence with a new lens.

Finally, the findings on age-related L2 learning effects are grounded in the widelyheld assumption in SLA that the obvious benchmarks for evaluating L2 acquisitionand L2 competence are L1 acquisition and L1 competence. In light of the realpossibility that bilingualism and language activation and dominance effects operateacross all ages, beginning as early as age two of life, we may need to revise thisassumption in the future. Quite simply, it may be that bilingual attainment,whether in early or late bilinguals, cannot be directly compared to monolingualattainment. At least David Birdsong (2005, 2006) in the United States, Birgit Harleyand Wenxia Wang (1997) in Canada, Vivian Cook (1991, 2008) in the UnitedKingdom and David Singleton (2001, 2003) in Ireland have raised the possibilitythat such a comparison may be misguided. Cook (2008) explains the dangers ofcomparing L2 users with monolinguals eloquently:

There is no reason why one thing cannot be compared to another; it may beuseful to discover the similarities and differences between apples and pears.SLA research can use comparison with the native speaker as a tool, partlybecause so much is already known about monolingual speakers. The danger isregarding it as failure not to meet the standards of natives: apples do not makevery good pears. Comparing L2 users with monolingual native speakers canyield a useful list of similarities and differences, but never establish the uniqueaspects of second language knowledge that are not present in the monolin-gual.

(p. 19)

26

How important is age in L2 acquisition, and (why) does it matter? 27

What could, then, be taken as a fair point of comparison to gauge attainment in agestudies? Citing the work on multicompetence and bilingualism by Cook (1991) andGrosjean (1989), Singleton (2003) suggests that:

the appropriate comparison in the investigation of age effects in L2 acquisitionis not between post-pubertal L2 beginners and monoglot native speakers butbetween post-pubertal L2 beginners and those who begin to acquire an L2 inchildhood.

(p. 10)

It may well be that the existing evidence on L2 critical periods will need to bereinterpreted as pointing at a fundamentally different state of language cognitionfor monolinguals and bilinguals, and a different state of overall readiness forlanguage activation of the L1 and the L2 depending on current amount of use ofeach, rather than a fundamental difference between early monolingual and late L2acquisition. If bilingualism and language activation/dominance effects operateacross all ages, then the explanatory onus would subtly move away from biologyand on to changes in the brain and in cognitive processing that are shaped by theexperience that results from being exposed to more than one languagesimultaneously or sequentially and across varying ages. The putative impossibilityto attain nativelikeness after a certain age, if reinterpreted under a bilingual lens bySLA researchers themselves, may turn out to mean that it is impossible forbilinguals to be monolinguals. This would be inconsequential both from atheoretical and a practical viewpoint. After all, saying that L2 learners cannot reachlevels that are isomorphic with monolingual competence would be a non-issue in aworld in which bilingualism would be considered the default state of the humanlanguage faculty.

In the end, these other recent strands of research suggest that a number ofenvironmental (e.g. opportunities for exposure and use of high-quality L2 inputand amount and quality of L1 use) and socio-affective factors (e.g. motivation, L2instruction and overall education) may mutually interact and become importantpredictors of success at earlier as well as later starting ages. Thus, these additionalvariables deserve much more research attention in the future. We can also predict,or at least hope, that many more SLA researchers in the future will turn to newmethods and designs that enable them to investigate the bilingualism of early andlate L2 learners in their own right, not as deficient or deviant replicas ofmonolinguals.

2.9 HOW IMPORTANT IS AGE IN L2 ACQUISITION, AND (WHY) DOES ITMATTER?

As we have seen in this chapter, over its more than 40 years of existence the field ofSLA has contributed a wealth of research on how age universally influences L2acquisition. Having answers to the questions raised by age effects in L2 acquisition

Age

is important because it would make us go a long way in our quest for understandingthe human language faculty as a whole. In addition, the main findings about ageand L2 acquisition generated by SLA researchers to date can be used productivelyto advocate for various populations of L2 learners.

First, knowing that young children may have a slow start when acquiring an L2can be an important research-based argument against harmful attempts topromote so-called sink-or-swim educational policies that attempt to reduce or evencompletely withdraw the first and second language support that is to be provided tolanguage minority children by schools. Such policies have been dangerouslygaining ground in the United States for some time now (see Crawford, 2000).Similarly, knowing that older children and adults can have an initial L2 learningadvantage for rate over early starters, and that this advantage may last for about fiveyears in second language environments and for even longer in foreign languagesettings, can also help problematize misguided attempts to mandate public schoolsto begin foreign language instruction in the first years of elementary educationwithout first evaluating whether the local resources and conditions canappropriately sustain such efforts throughout the full length of schooling (see thediscussion of these issues in the United States in Lally, 2001). This trend isregrettably expanding, particularly in areas of the world where English is seen asthe default foreign language (e.g. Nunan, 2003). The third important age-relatedfinding to remember is that amazingly successful late learners such as Julie andLaura exist, and that perhaps they account for as much as 5 per cent to 25 per centof best-scenario learning cases (Birdsong, 1999b). This is hopeful for languageteachers and educators. Indeed, knowing that many of them are highly motivatedstudents who also enjoyed high-quality instructional experiences (Bongaerts, 1999;Moyer, 1999) is certainly good ammunition for lobbying in favour of increasinginvestment of material and human resources for the improvement of second andforeign language education.

2.10 SUMMARY

● In terms of L2 learning rate, adults and older children enjoy an initialadvantage over young children that may last over up to one year, some-times up to three years, particularly if they are tested through tasks thatdemand cognitive maturity and involve metalinguistic skills. After fiveyears, however, early starters catch up and are better than late starters insecond language contexts. In foreign language contexts, by contrast, thelagged advantage for an earlier start has not been observed, even after fiveyears.

● In terms of L2 ultimate attainment, most learners who begin acquiring theL2 before a certain age, typically before puberty, will develop levels ofmorphosyntactic and phonological competence that are very close to thoseof native speakers of that language. Post-pubertal learners, however, are not

28

Annotated suggestions for further reading 29

likely to perform in the native speaker range, and this holds true regardlessof the number of years they have resided in the L2 environment.

● Exceptions to the observed success and failure tendencies associated withage exist. Thus, some adult starters can achieve native-like levels in their L2,or at least extremely high levels that are near-native. Conversely, an earlystart does not guarantee complete and successful L2 acquisition in all cases,as some children who start learning the L2 at an age as early as four or eventwo may be found to differ from native speaker performance in subtle ways.In the former case, exceptions appear to be related to unusually highmotivation and high quality of instruction, whereas in the latter case theyappear to be associated with high L1-use levels (that is, with high L1activation or L1 dominance).

● Several explanations for the observed age effects have been proposed andare considered plausible by different SLA researchers. Those in favour of acritical period explanation posit that, after a certain age, it is biologicallyimpossible for the human brain to use the same processes that wereinvolved in learning the L1. Instead, other processes, such as reasoning andproblem solving, are summoned during post-pubertal L2 learning. Severalneurological and neurochemical causes have been considered (includinglateralization, plasticity, myelination and pubertal increases in oestrogen ortestosterone) but the empirical evidence is still unavailable for any of them.Of the researchers who favour non-biological explanations, some haveconsidered pre-existing knowledge of the L1 and others have emphasizedsocio-educational and affective-motivational forces.

● Recent research suggests that bilingualism effects (e.g. L1–L2 interactions)and language activation and dominance effects (i.e. relative amounts of L1versus L2 use) operate across all ages, beginning as early as age two. Thisevidence suggests that it may be misguided to compare bilingual attainmentto monolingual attainment. Thus, in the future, research programmes mayneed to shift away from the emphasis on a fundamental difference betweenmonolingual child L1 acquisition and monolingual-like adult L2 acquisitionand towards investigating changes in the brain and in cognitive processingthat are shaped by the experience that results from being exposed to morethan one language simultaneously or sequentially and across varying ages.

2.11 ANNOTATED SUGGESTIONS FOR FURTHER READING

Newcomers to the field of SLA are often overwhelmed by the many arguments anddata that appear to equally support and contradict the Critical Period Hypothesis inL2 learning. It is important to maintain an open mind and an attentive eye whenyou delve into this literature. With a topic as controversial as this one, it is good tofirst read the brief overviews by Scovel (2000) and Singleton (2001). You can

Age

profitably compare these two articles published in the same journal and incontiguous years by two seminal experts in the area, the former representing theposition in favour of a biological schedule explanation for the observed agedifferences and the latter espousing a more sceptical stance. At a more advancedlevel, it is good to read and compare the overviews by Hyltenstam andAbrahamsson (2003) and Birdsong (2006), which offer a more contemporary view,again the former definitely in favour of critical periods for L2 learning and the lattermore cautious about making final interpretations.

An accessible treatment of possible reasons for scepticism regarding the CriticalPeriod Hypothesis is Marinova-Todd et al. (2000), and I recommend you also readthe rebuttal by Hyltenstam and Abrahamsson (2001) and the authors’ response(Marinova-Todd et al., 2001). You could then deepen your understanding of theeducational ramifications of age effects by reading Nikolov and MihaljevicDjigunovic (2006).

If you are interested in gaining expertise in this topic, book-length readings arein order. The collection of papers edited by David Birdsong (1999a) is difficult toread but fascinating because of the balance between positions in favour and againstbiological explanations for age effects, and because of the range of theoretical andempirical arguments represented. Although specialized in treatment, the book byJulia Herschensohn (2007) offers a uniquely valuable window into the issue of agebecause it reviews in depth age effects in both L1 and L2 acquisition and it takes aninnatist approach but ends up arguing in favour of a critical period for first but notsecond language acquisition. Tom Scovel’s (1988) book is still a classic well worthreading if you are interested in foreign accents. The collections edited by GarcíaMayo and García Lecumberri (2003) and Muñoz (2006) will be exciting reading ifyour main interest is in L2 learning in foreign language contexts.

I hope your interest in definitions and limits of nativeness in SLA has beenspurred by our discussion in section 2.8. If this is the case, Grosjean (1989), Cook(1991, 2008) and Birdsong (2005) are good readings for you.

Finally, for readers who are or will be parents interested in knowing more aboutraising children with multiple languages, King and Mackey’s (2007) book isengaging and directed to a lay audience but well rooted in the best research aboutthe topic.

30

Interaction and Instructed Language Acquisition (Please answer in full sentences and use your own words)

1. What the relationship between current approaches to language teaching methodologies and the interactionist approach to instructed second language acquisition?

2. What ate the main constructs of interaction? How are they defined? How are they

operationalized?

3. What are some of the psycholinguistic benefits of peer interaction?

4. How is a task defined? Why is this important for the interactionist approach to language teaching?

Lang. Teach. (2018), 51.3, 285–329 c© Cambridge University Press 2018doi:10.1017/S0261444818000125

State-of-the-Art Article

Interaction and instructed second language acquisition

Shawn Loewen Michigan State University, Michigan, USAloewens@msu.edu

Masatoshi Sato Universidad Andres Bello, Santiago, Chilemasatoshi.sato@unab.cl

Interaction is an indispensable component in second language acquisition (SLA). This reviewsurveys the instructed SLA research, both classroom and laboratory-based, that has beenconducted primarily within the interactionist approach, beginning with the core constructs ofinteraction, namely input, negotiation for meaning, and output. The review continues with anoverview of specific areas of interaction research. The first investigates interlocutorcharacteristics, including (a) first language (L1) status, (b) peer interaction, (c) participationstructure, (d) second language (L2) proficiency, and (e) individual differences. The secondtopic is task characteristics, such as task conditions (e.g. information distribution, task goals),task complexity (i.e. simple or complex), and task participation structure (i.e. whole class,small groups or dyads). Next, the review considers various linguistic features that have beenresearched in relation to interaction and L2 learning. The review then continues withinteractional contexts, focusing especially on research into computer-mediated interaction.The review ends with a consideration of methodological issues in interaction research, such asthe merits of classroom and lab-based studies, and the various methods for measuring thenoticing of linguistic forms during interaction. In sum, research has found interaction to beeffective in promoting L2 development; however, there are numerous factors that impact itsefficacy.

1. Introduction

Interaction has played an important role in SLA theory for several decades. The interactionhypothesis, proposed by Long (1981, 1983) and revised in 1996 (Long 1996), was based ondiscourse analysis research during the 1970s (e.g. Wagner-Gough & Hatch 1975; Hatch 1978)and has developed and matured with burgeoning empirical research since then. Indeed, thehypothesis has developed into a theoretical approach (see Mackey & Gass 2015), containinga description of multiple processes associated with L2 learning (Jordan 2005; Mackey 2012;Pica 2013). Such processes include learners’ exposure to and production of language, andthe interplay of such input and output with learners’ cognitive resources and other individualdifferences (Long 1996; Gass 1997; Mackey 2012; Pica 2013; Gass & Mackey 2015; Long

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

2 8 6 S H A W N L O E W E N A N D M A S A T O S H I S A T O

2015; Loewen & Sato 2017). Investigation into the effects of interaction on L2 developmenthas been considered a component of instructed second language acquisition (ISLA) (e.g.Mackey 2006; Loewen 2015; Long 2017; VanPatten 2017b), which is concerned with themanipulation of the processes and mechanisms of L2 learning. Consequently, interactionresearch has occurred in both intact classrooms and in laboratory contexts, with results of thelatter studies generally considered in light of their implications for the L2 classroom. Thus,the current review considers any research conducted within the interaction approach.

Much of the early interactionist work explored how interaction occurred in variouscontexts. Seminal works investigated, among other topics, speech modifications andmiscommunication in native speaker/non-native speaker interaction as well as non-native/non-native speaker interaction (Gass & Varonis 1985; Varonis & Gass 1985; Doughty& Pica 1986; Porter 1986; Pica 1988; Gass & Varonis 1990; Loschky 1994). In particular,researchers were interested in the frequency with which negotiation of meaning, a series ofdiscourse moves to resolve a communication breakdown, occurred, how it occurred, andwhat factors influenced its occurrence (e.g. Long & Porter 1985; Pica et al. 1991; Pica1994; Lyster & Ranta 1997). As the characteristics of interaction became clearer, researchersbegan to investigate the effects of specific variables pertaining to interaction on (a) discoursemoves such as modified output (Swain 1985, 1995, 2005), (b) cognitive constructs suchas noticing (Schmidt 1990, 1995, 2001), and (c) most importantly, L2 development andacquisition (see Mackey 1999; Spada & Lightbown 2009; Mackey 2012). Variables underinvestigation, and which comprise a large part of the current review, can be categorized asfollows: those pertaining to (a) the interlocutors (e.g. L2 proficiency, L1 status, gender), (b) thetask characteristics (e.g. complexity, type of task), (c) linguistic targets, and (d) the interactionalcontext (e.g. setting, modality).

Indeed, the interest in interaction, its effects, and the variables impacting its effectivenesshave attracted considerable attention, resulting in numerous empirical research studies,reviews (Gass 2003; Plonsky & Gass 2011; Mackey, Abbuhl & Gass 2012; Goo & Mackey2013; Lyster & Ranta 2013; Lyster, Saito & Sato 2013; Plonsky & Brown 2015; Kim 2017),and meta-analyses (Keck, Iberri-Shea, Tracy-Ventura & Wa-Mbaleka 2006; Russell & Spada2006; Mackey & Goo 2007; Li, 2010; Lyster & Saito 2010; Brown 2016; Ziegler 2016a) ofinteraction in general or specific components of interaction. Overall, these works have foundthat interaction is beneficial for L2 acquisition. For example, Keck et al.’s (2006) meta-analysisof 14 quasi-experimental interaction studies revealed a large effect on immediate posttestsfor learners involved in interaction (and negotiation). Mackey & Goo’s (2007) meta-analysisof 28 interaction studies conducted both inside and outside the classroom found large effectsizes for learners who were engaged in interaction compared to those who were not. Theseresults were seen more clearly on delayed posttests. Indeed, in terms of interaction, there issome consensus that ‘there is a robust connection between interaction and learning’ (Gass &Mackey 2015: 181), with the result that researchers have shifted their focus from the generaleffectiveness of interaction to the details of which components of interaction might be moreor less effective in which contexts with which learners (Mackey et al. 2012).

Interaction research in ISLA, at its broadest, encompasses several areas of research thatfocus only on specific aspects of interaction; nevertheless, such research is supported by thetheoretical underpinnings of the interactionist approach. For example, corrective feedback

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 287

has a robust research tradition in its own right, as evidenced by multiple reviews and meta-analyses (e.g. Russell & Spada 2006; Li 2010; Lyster & Saito 2010; Lyster et al. 2013;Brown 2016); however, corrective feedback is also a central component of the interactionistapproach. In another example, task-based language teaching and learning is an area ofconsiderable current interest with its own research agendas (e.g. Baralt, Gilabert & Robinson2014; Skehan 2014; Long 2015; Gilabert, Manchon & Vasylets 2016; Plonsky & Kim 2016;Ziegler 2016a), but much of the rationale for tasks is based on the interaction approach(Bygate, Skehan & Swain 2001; Ellis 2003; Long 2016; Ellis 2017b; Kim 2017). In reviewingthe literature over 30 years since the initial proposal of the interaction hypothesis (Long 1981),Long (2015) emphasized that ‘it is this theory and related empirical findings that provide themain psycholinguistic underpinnings for TBLT [task-based language teaching]’ (p. 61). Thecurrent review will draw on studies from these various areas of research.

2. Components of interaction

Before going further in this review, it is important to consider the key constructs of interaction,namely input, negotiation, output, and noticing.

2.1 Input

Input is a necessary component of all theories of language acquisition, including theinteractionist approach (Gass & Mackey 2015). In particular, interactionist researchers havebeen interested in the input that learners receive, whether naturalistic, pre-modified (i.e.simplified and/or elaborated), or interactionally modified. Two issues here are the effects ofinput on comprehension and L2 development. In terms of comprehension, there is evidencethat interactionally modified input may be more beneficial (e.g. Pica, Young & Doughty1987; Loschky 1994). Furthermore, interactionally modified input, as exemplified in thesubsequent examples, has generally been found to be better for L2 acquisition than pre-modified or unmodified input (e.g. Ellis & He 1999; Mackey 1999). Because interactionallymodified input is promoted as an optimal type of input for acquisition, few studies withinthe interactionist approach have investigated simplified or elaborated input alone. However,Ellis & He (1999) compared pre-modified and interactionally modified input, finding nodifference between the two types of input in a task-based study of vocabulary learning. Otherresearchers have investigated input-based and output-based instruction, although much ofthis research is not directly situated with the interactionist approach and is consequentlybeyond the scope of this review (e.g. DeKeyser & Sokalski 1996; Erlam 2003; Shintani 2015).

2.2 Negotiation for meaning

Negotiation for meaning is at the heart of the interaction hypothesis, with a breakdown incommunication being posited as the driving force in improving learner comprehension andL2 development (Long 1996). When L2 learners and their interlocutors do not understand

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

2 8 8 S H A W N L O E W E N A N D M A S A T O S H I S A T O

each other, they may signal that a communication breakdown has occurred. Hence, keyelements of negotiation for meaning include clarification requests, confirmation checks, andcomprehension checks, all of which are responses to a communication breakdown. Theseare the primary discourse moves that have been investigated in research studies that examinethe occurrence of negotiation (e.g. Ellis, Basturkmen & Loewen 2001a; Loewen 2004; Gass,Mackey & Ross-Feldman 2005).

Confirmation checks aim to ensure that what has just been said has been heard correctlyand understood. Confirmation checks often take the form of a repetition of the troublesource with rising intonation, or a question such as Do you mean X? In Example 1, taken froma spot-the-differences task, two learners are discussing the objects in their respective pictures.Learner 2 checks to confirm her understanding of the information that Learner 1 has justprovided, to which Learner 1 answers in the affirmative.

Example 1: Confirmation check (indicated by SMALL CAPS) (Gass et al. 2005: 585)

Learner 1: En mi dibujo hay un pajaro. ‘In my drawing there is a bird.’

Learner 2: ¿SOLAMENTE UN? Tengo, uh, cinco pajaros con un hombre, en sus hombros.

‘ONLY ONE? I have, uh, five birds with a man, on his shoulders.’

Learner 1: Oh, oh, sı, sı. ‘Oh, oh, yes, yes.’

In contrast to confirmation checks, clarification requests attempt to elicit additionalinformation from the interlocutor regarding the meaning of their utterance, through questionssuch as What do you mean? In Example 2, which occurred in an information and opinion gaptask, Learner 2 simply requests more information from his interlocutor by asking What?

Example 2: Clarification request (indicated by SMALL CAPS) (Gass et al. 2005: 586)

Learner 1: ¿Que es importante a ella? ‘What is important to her?’

Learner 2: ¿COMO? ‘WHAT?’

Learner 1: ¿Que es importante a la amiga? ¿Es solamente el costo? ‘What is important to thefriend? Is it just the cost?’

The final major category of negotiation for meaning is comprehension checks, which areused to ensure that the addressee understands what is being said. Comprehension checks cantake the form of questions such as Do you understand what I’m saying? In Example 3, in whichlearners are engaging in a map description task, Learner 1 asks if Learner 2 would like herto repeat the information that she has just provided.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 289

Example 3: Comprehension check (indicated by SMALL CAPS) (Gass et al. 2005: 586–587)

Learner 1: La avenida siete va en una direccion hacia el norte desde la calle siete hasta la calle ocho.

¿QUIERES QUE REPITA? ‘Avenue Seven goes in one direction towards the northfrom Street Seven to Street Eight. DO YOU WANT ME TO REPEAT?’

Learner 2: Por favor. ‘Please.’

2.3 Negotiation of form

Although negotiation for meaning is sometimes present during communicative interaction,it has become clear that such negotiation does not always occur frequently in the classroom(Foster 1998; Eckerth 2009). Consequently, in addition to investigating negotiation formeaning, some interactionist researchers expanded their scope to include negotiation thatoccurs as a result of a desire for linguistic accuracy, rather than as a result of communicationbreakdown (e.g. Lyster & Ranta 1997; Ellis et al. 2001a; Lyster et al. 2013). This type ofnegotiation is particularly relevant to ISLA because it often involves teachers’ pedagogicalintervention. More precisely, negotiation of form can be attributed ‘a more didactic function’(Lyster 1998b: 190), often consisting of feedback with corrective intent. Corrective feedbackthat is didactic in nature (e.g. didactic recasts) occurs when someone, often the teacher,responds to a learner’s linguistically problematic utterance, even though the meaning of theutterance is clear. Thus, in Example 4, Will uses the wrong preposition in discussing his armyexperiences, to which the teacher responds with the correct preposition. Will repeats thecorrect form, and they continue with their interaction.

Example 4: Corrective feedback (indicated by SMALL CAPS) (Loewen 2005: 371)

Will: when I was soldier I used to wear the balaclava

Teacher: and why did you wear it Will, for protection from the cold or for another reason

Will: just wind uh protection to wind and cold

Teacher: PROTECTION FROM

Will: uh from wind and cold

Teacher: right, okay not for a disguise

Considerable research has investigated corrective feedback during interaction over thepast 25 years (e.g. Lyster & Ranta 1997; Long, Inagaki & Ortega 1998; Ammar & Spada2006; Ellis, Loewen & Erlam 2006; Mackey 2006; Yang & Lyster 2010; Li, Zhu &

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

2 9 0 S H A W N L O E W E N A N D M A S A T O S H I S A T O

Ellis 2016; Nakatsukasa 2016), resulting in numerous research syntheses (e.g. Long 2007;Lyster, Saito & Sato 2013; Nassaji 2013; Ellis 2017a) and meta-analyses (e.g. Russell &Spada 2006; Li 2010; Lyster & Saito 2010; Brown 2014). Several important distinctions havebeen made concerning corrective feedback, including (a) negative versus positive evidence(Leeman 2003), (b) input-providing versus output-prompting feedback (Lyster 2004; Goo& Mackey 2013; Lyster & Ranta 2013), and (c) explicit versus implicit feedback (Sheen &Ellis 2011; Lyster et al. 2013). Regarding the first distinction, negative evidence indicatesto learners what is not acceptable in the target language, while positive evidence providesexamples of well-formed utterances. Researchers have long argued that negative evidenceis an important component of corrective feedback (e.g. Schachter 1991); however, researchhas indicated that the positive evidence provided by recasts can also be important (Leeman2003).

Input-providing feedback provides the correct linguistic form for the learner. For example,a recast, which reformulates a learner’s incorrect utterance, provides the correct formimmediately after the learner’s erroneous utterance. In contrast, output-prompting correctivefeedback does not provide the correct form for the learner; rather, it attempts to elicit thecorrect form from the learner. The effectiveness of these two types of feedback has been widelydebated. Theoretical arguments and empirical investigation have been used to support thesuperiority of both input-providing feedback (e.g. Long 2007; Goo & Mackey 2013) andoutput-prompting feedback (e.g. Lyster 2004; Lyster & Ranta 2013), although some studieshave also found similar effects between the two (e.g. Loewen & Nabei 2007). This state ofaffairs has led some researchers to suggest that teachers should provide a variety of feedbacktypes in the classroom (Lyster & Ranta 2013; Ellis 2017a).

In terms of the explicit-implicit distinction (Lyster et al. 2013), several researchers (e.g. Long1996, 2007; Goo & Mackey 2013) have argued that feedback should be relatively implicit,primarily taking the form of a recast so as not to interrupt the flow of communication. Forexample, Long (2015) reiterates that recasts ‘have a solid track record in both L1A [firstlanguage acquisition] and SLA, and to the extent that implicit negative feedback does thejob, teachers and learners are freed up to devote their primary attention to tasks and subject-matter learning’ (p. 57). In contrast, other researchers (Lyster 2004; Ellis, Loewen & Erlam2006; Loewen & Philp 2006; Lyster & Ranta 2013) have suggested that more explicit types ofcorrective feedback, such as explicit correction or metalinguistic information, are more likelyto be noticed by learners and thus have the possibility to influence the learner’s interlanguagesystem.

Language-related episodes (LREs), which occur when one or more of the participants‘generate [linguistic] alternatives, assess [linguistic] alternatives, and apply the resultingknowledge to solve a linguistic problem’ (Swain & Lapkin 1998: 333), are also an exampleof negotiation arising out of a general concern for linguistic accuracy. In LREs, specificlinguistic items are topicalized by learners as they engage in meaning-focused communication.Numerous studies, both experimental and descriptive, with L1 or L2 interlocutors, haveemployed LREs to identify learning opportunities during interaction (e.g. Swain & Lapkin1998; Williams 2001b; Storch 2002; Loewen 2005; Kim & McDonough 2008; Yilmaz 2011;Garcıa Mayo & Azkarai 2016). In Example 5, a group of learners focus on a lexical issue(rascacielos/skyscrapers) in Spanish. Jenny does not recognize the word that Larry used. Larry

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 291

then explains the meaning of the word, which is subsequently accepted by the other membersof the group. As this example shows, an LRE does not necessarily contain corrective feedback;rather, an LRE consists of collaborative exchanges surrounding a linguistic issue.

Example 5: Language-related episode (Fernandez Dobao 2016: 40)

Larry: entre dos rascacielos, grandes ‘between two big skyscrapers’

Ruth: dos ‘two’

Jenny: que es? ‘what is it?’

Larry: skyscrapers

Jenny: rascacielos? ‘skyscrapers?’ oh!

Ruth: rascacielos rascacielos ‘skyscrapers skyscrapers’

Jenny: look at you

Larry: sı ‘yes’

Jenny: rascacielos ‘skyscrapers’

Ruth: okay

2.4 Output

The final construct to consider within the interactionist approach is output, namely, thelanguage that learners produce during meaning-focused interaction. Although some SLAtheories do not ascribe an important role for language production in L2 development(e.g. Krashen 1982, 2003), the interactionist approach draws strongly on Swain’s (1985,1995, 2005) Comprehensible Output Hypothesis, which claims that output is not merely arepresentation of L2 development but is a causal factor for L2 development in several ways.First, Swain argues that learners need to be pushed to produce language that is syntacticallymore complex and accurate. Learners tend to process language semantically, and if they cangrasp the meaning of an utterance without needing to process all of the linguistic features, theywill do so. But if learners have to produce language, then they have to consider which specificlinguistic forms encode which meanings (i.e. the noticing function). Second, Swain proposesthat output allows learners to test their linguistic hypotheses and possibly receive feedbackon these hypotheses (i.e. the hypothesis-testing function). For example, learners may decide

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

2 9 2 S H A W N L O E W E N A N D M A S A T O S H I S A T O

to try out a new structure during communication, but if such use results in communicationbreakdown or corrective feedback, learners may realize that they need to revise their originalhypothesis about the target language structure. Third, Swain asserts that output serves ametalinguistic function, ‘enabling [learners] to control and internalize linguistic knowledge’(Swain 1995: 126). Finally, as output necessarily involves language use, it facilitates productionpractice and, thus, the development of fluency and automaticity (see DeKeyser 2001; Lyster& Sato 2013; DeKeyser 2017a).

2.5 Attention

While the previously discussed constructs are discoursal, this final one is cognitive in nature.Long’s (1996) characterizations of the interaction hypothesis argues that interaction ‘connectsinput . . . ; internal learner capacities, particularly selective attention; and output . . . inproductive ways’ (451–452). Indeed, Schmidt’s noticing hypothesis (1990, 1995, 2001) hasplayed an integral role in the interaction approach, with the claim that L2 learning doesnot occur without awareness. Similarly, Robinson (1995, 1996, 2003) claims that attention,namely the ‘process that encodes language input, keeps it active in working and short-termmemory, and retrieves it from long-term memory’ (2003: 631), is essential for L2 learning.Although interaction researchers do not agree on the precise nature of these constructs, thereis general agreement that ‘the cognitive constructs of attention, awareness, and the relatedconstruct of noticing are part of the interaction–L2 learning process (Gass & Mackey 2015:191). Consequently, researchers have expended considerable effort investigating the effectsof interaction not only on L2 development but also on attention, awareness, and noticing(e.g. Mackey, Gass & McDonough 2000; Williams 2001b; Philp 2003; Gass & Alvarez Torres2005; Mackey 2006; Philp & Iwashita 2013; Sagarra & Abbuhl 2013).

Although it is important to identify the key components of interaction, the primary concernof interactionist research has been to examine how the various components of interaction,particularly negotiation for meaning, corrective feedback, and output, occur in differentcontexts and are influenced by different characteristics (e.g. Gass 2003; Keck et al. 2006;Russell & Spada 2006; Mackey & Goo 2007; Li, 2010; Lyster & Saito 2010; Plonsky & Gass2011; Mackey et al. 2012; Goo & Mackey 2013; Lyster & Ranta 2013; Lyster et al. 2013;Plonsky & Brown 2015; Ziegler 2016a; Brown 2016; Kim 2017). To that end, the review willnow focus on studies that have examined the influence of (a) interlocutor characteristics, (b)task characteristics, (c) linguistic targets, and (d) interactional contexts.

3. Interlocutor characteristics

From its inception, interaction researchers have been concerned with the effects that theinterlocutor has on interaction (e.g. Long 1983; Long & Porter 1985; Varonis & Gass1985; Pica 1988). Indeed, an important set of moderating factors related to L2 interactioninvolve the various characteristics of the individuals engaged in communication. This section

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 293

will investigate several of these features, including (a) L1 status, (b) peer interaction, (c)participation structure, (d) L2 proficiency, and (e) cognitive and psychological individualdifferences.

3.1 L1 status

Interactionist researchers have long been interested in the effect of L2 learners speakingwith L1 speakers (often operationalized as native speakers) and other L2 speakers (oftenoperationalized as non-native speakers) of the language (see Long & Porter 1985). Inparticular, researchers have explored whether L1 and L2 speakers are equally capable ofproviding the linguistic support that is arguably necessary for L2 development (e.g. inputmodifications and corrective feedback: see Pica 2013). Much of this L1–L2 interactionresearch has been conducted in the laboratory because not many L1 speakers, apart fromthe teacher, who may or may not be an L1 speaker, are present in instructed L2 contexts.Furthermore, not many studies have examined naturally occurring L2 learner interactionsoutside the classroom in L2 contexts, such as study abroad (see Perez-Vidal 2017). Of course,such interaction with L1 speakers is generally not possible in a foreign language contextwhere the L2 is not spoken in the wider society, although technology now allows L2 learnersaccess to L1 speakers, which was previously not possible (e.g. Sauro 2013b).

Research comparing L1–L2 and L2–L2 speaker interaction has investigated four keyinteractional features: input modifications, corrective feedback, modified output, and self-initiated modified output. First, studies that compared L1 and L2 speakers as input providersreported that L1 speakers tend to provide lexically richer and syntactically more complexinput than do L2 speakers. Pica et al. (1996), for instance, gave picture jigsaw tasks (picturedescription and story completion) to L1–L2 and L2–L2 dyads. The analysis of frequenciesof lexical and morphosyntactic modifications of the two types of interlocutors revealed thatL1 speakers provided more modifications after communication breakdowns but only for thepicture description task (see also Porter 1986). However, not all studies have found in favourof L1 speakers. For example, Garcıa Mayo & Pica (2000) replicated Pica et al. (1996) withEnglish-major university students whose paper-based TOEFL scores ranged from 580 to630. The researchers found that the advanced L2 interlocutors produced structurally richerinput to their partners than did the L1 speakers. In a more recent study, Sato (2015), bycomparing the density and complexity of L1 and L2 interlocutors’ speech production, foundthat learners can provide a comparable amount (i.e. mean length of utterance) and variety(i.e. the number of types of verbs) of input to that of L1 speakers primarily due to the linguisticsimplifications that the L1 speakers tended to produce (i.e. foreigner talk: see Ferguson 1975).However, learners sometimes provided their partners with grammatically incorrect input orconcluded their interaction with non-target-like solutions (see Bruton & Samuda 1980).

Second, concerning feedback, researchers have investigated learners’ signalling of non-understanding as well as learners’ provision of feedback. Regarding the first issue, researchindicates that learners are often more willing to indicate a lack of understanding when theyare interacting with another learner. In the aforementioned Pica et al. (1996) study, theanalysis of the amount of feedback showed that there were significantly more such instances

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

2 9 4 S H A W N L O E W E N A N D M A S A T O S H I S A T O

in L2–L2 interaction (for one of the two tasks implemented in the study). The researchersconcluded that L2–L2 interaction ‘did offer data of considerable quality, particularly in thearea of feedback’ (1996: 80). A similar conclusion was reached by Eckerth (2008) in hisstudy of L2–L2 interaction in two German L2 university classes. He found that learnersprovided each other with ‘feedback rich in acquisitional potential’ (2008: 133) on bothlinguistics structures targeted in the tasks as well as structures that arose incidentally during theinteraction.

Research has also found that learners tend to react to feedback by modifying theirinitial errors (modified output) more often during interaction with L2 peers rather thanwhen they interact with L1 speakers. Sato & Lyster (2007) compared the interaction ofuniversity-level Japanese learners of English with each other and with L1 speakers of Englishfrom Australia and Canada. Sato & Lyster found that the Japanese learners modified theirerroneous utterances after feedback from their peers more often than from native speakers.This tendency was consistent across feedback types. In Mackey, Oliver & Leeman (2003),lower-intermediate learners of English with a variety of L1 backgrounds interacted with L1English speakers in information gap tasks. The analysis of 24 adult dyads showed that whileL2 speakers provided more output-promoting feedback, the quality of the resulting modifiedoutput – modification of the original error – in the two types of dyads was comparable.However, Shehadeh (1999), who examined the interactions between adult English as asecond language (ESL) learners in the UK and between L2 learners and L1 speakers, showedthat learners were more likely to ‘make an initial utterance more accurate and/or morecomprehensible to their interlocutor(s)’ (1999: 644) when feedback was given by anotherL2 speaker than by an L1 speaker, especially when negotiation was extended between theinteractants (see also Fujii & Mackey 2009).

Finally, while much less investigated, research indicates that learners tend to self-correctmore while interacting with other learners than when interacting with L1 speakers. Self-corrections, which can be argued to be ‘overt manifestations of the monitoring process’(Kormos 2006: 123) and are included in Swain & Lapkin’s (1998) definition of LREs, arehypothesized to facilitate L2 processing in the same way as modified output triggered byfeedback (de Bot 1996). Shehadeh (2001), in a follow up examination of his earlier data(1999), showed that 93% of self-initiation led to modified output during L2–L2 interaction,which was proportionally higher than L2–L1 interaction, again after extended negotiations.In her investigation of L2 pair interaction among English as a foreign language (EFL) learnersin Thailand, McDonough (2004) found that learners rarely modified their output in responseto peer feedback, choosing instead to produce self-initiated modified output. Nevertheless,increased peer interaction led to improved production of English conditionals, the targetfeature (see also Buckwalter 2001; Toth 2008; Smith 2009).

One special type of peer group that has received recent attention is heritage language(HL) learners, who have had some degree of exposure to the target language at home. Forexample, Bowles, Toth & Adams (2014), comparing HL–L2 and L2–L2 dyads, found thatHL–L2 dyads were more likely to reach target-like outcomes during dyadic interaction andnegotiation. There was also a greater amount of talk and more LREs when an HL learnerwas involved. Bowles et al. suggest that there is an imbalance when L2 learners work with HLlearners, with the L2 learners receiving more benefit from the interaction than do the HL

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 295

learners. In addition, learners were more likely to stay in the target language when speakingwith an HL learner than with another L2 learner.

In sum, these studies indicate that L1 interlocutors can interact with L2 learners in positiveways, but it is not always the case that interaction with L1 speakers is better than interactionwith L2 peers. Indeed, L2 speakers can sometimes be better for interlocutors than L1 speakers,something that should perhaps be stressed in the classroom as support for interactive tasks(Long & Porter 1985). Furthermore, this is good news since opportunities for L1 speakerinteraction are considerably limited for most learners.

3.2 Peer interaction

In addition to the above mentioned L1–L2 distinction, another interlocutor characteristicis the difference between interacting with the teacher versus fellow students or peers.Confounded with this variable is the fact that teacher and student roles often overlapconsiderably with participation structure, with whole class interaction being primarilyteacher-led, while small group and dyadic interaction is generally with peers.

As early as 1985, Varonis & Gass (1985) concluded that peer interaction can be ‘a goodforum for obtaining input necessary for acquisition’ (p. 83). Indeed, peer interaction is oftenthought of as the primary participant structure for interaction to occur, and more researchhas recently investigated the effects of peer interaction in large part because it is probablythe most common type of interaction in many communicatively oriented classrooms. Inparticular, researchers have investigated learner interactions with other learners in order toexamine whether peers can also be a helpful learning resource (see, for reviews, Philp, Adams& Iwashita 2014; Sato & Ballinger 2016). Indeed, TBLT, which will be discussed in moredetail, utilizes peer interaction as a vehicle to enhance the effectiveness of tasks.

The psycholinguistic benefits of peer interaction may include relatively longer processingtime for input and output, which in turn leads to frequent feedback and opportunitiesfor testing out or practicing the language. In a recent comprehensive monograph of peerinteraction, for instance, Philp et al. (2014) argued that peer interaction gives learners ‘acontext for experimenting with the language’ (p. 17). In addition, peer interaction offers apsychological benefit. That is, an increased comfort level may positively affect learners’ L2processing by helping them notice errors in their partners’ speech, correct those errors, andmodify their own errors when given feedback. Moreover, a greater comfort level seems toincrease the amount of overall language production, which leads to more opportunities forlanguage practice. Philp et al. (2014) asserted that peer interaction is ‘generally felt to be lessstressful than teacher-led interaction, precisely because it will not be carefully monitored’(Philp et al. 2014: 198). In Sato’s (2013) study of learners’ perceptions of peer interaction,learners explained that they did not have to worry about making errors while talking to eachother, especially in comparison to a teacher-centered conversation.

However, Tomita & Spada (2013) found that social context was an important factor inclassroom interaction among Japanese high school learners of English. Students sometimeschose not to communicate in English during peer interaction because using English wassocially stigmatized as showing off. However, if learners could position themselves as

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

2 9 6 S H A W N L O E W E N A N D M A S A T O S H I S A T O

incompetent learners who struggled with the language, then using English was moresocially acceptable. A similar finding was reported by Yoshida (2013), who investigatedthe relationship between learner beliefs and actions. Learners of Japanese at an Australianuniversity expressed that although they believed that participation improves speaking skills, alack of confidence related to accuracy inhibited their actual participation. To this end, Storch(2002) suggests that learners’ level of comfort may depend on the relationship that a pair orgroup of learners construct together. Indeed, Sato (2016) examined the relationships among(a) learner psychology towards interlocutor and/or task, (b) peer interaction patterns, and (c)developmental outcomes, and discovered that the learners who approached peer interactionwith a positive mindset tended to engage with the tasks and benefited from interaction morethan those who exhibited a non-collaborative mindset.

Arguments suggesting that peer corrective feedback should be discounted because of itspoorer quality compared to teacher feedback have been investigated. Adams (2007) analyseddyadic interactions between adult ESL learners. Based on the feedback episodes, Adamsdeveloped a tailor-made posttest. Results showed that 59% of feedback led to learning acrossdifferent linguistic structures, showing a positive role of peer feedback. However, in anotherstudy, Adams, Nuevo & Egi’s (2011) correlation analyses between the frequency of feedbackreceived by learners and their gain scores showed that the provision of explicit correctionsand the development of past tense were significantly negatively correlated. The researchersconcluded that ‘feedback may not play as important a role in learner–learner interaction as itplays in native speaker–learner interactions’ (2011: 56). In a classroom setting, Sato & Lyster’s(2012) experiment showed positive effects of peer corrective feedback. In this study, learnerswere trained to provide feedback to each other in order to compensate for the weaknesses ofpeer feedback (e.g. infrequent, inaccurate, or unfocused feedback). This intervention resultedin an increase in grammatical accuracy. The researchers explained the effect by invokingLevelt’s (1983) perceptual loop theory and proposed that peer corrective feedback serves adual function and benefits both the provider and the receiver. That is, in order to providecorrective feedback to their classmates, learners need to notice errors in their classmates’speech first. This monitoring process facilitates L2 learning in the provider (see also Naughton2006; Fujii & Mackey 2009; Sato & Ballinger 2012; Fujii, Ziegler & Mackey 2016; Sato 2017).

These varied findings underscore the importance of a classroom environment wherelearners’ collaborative interaction is encouraged and, possibly, explicitly taught. In such anenvironment, learner psychology may be aligned to psycholinguistic processes conducive toL2 development. Indeed, the relationship between interaction and learner psychology hasbeen under-investigated, although as early as 2000, Dornyei & Kormos (2000) argued that‘all the cognitive and linguistic processes discussed in the L2 task literature depend, to someextent, on this initial condition [task attitudes]’ (p. 281). We will turn to this topic when wediscuss individual differences.

3.3 L2 proficiency

Interaction may be influenced by the proficiency levels of the interlocutors (e.g. Yule &Macdonald 1990; Leeser 2004). Two groups of studies have investigated the proficiency

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 297

effect on interaction patterns as well as on L2 learning. The first group compared learnerpairs of the same or different proficiency levels (Same vs Mixed). In this area of study,some research has reported that pairing learners with different proficiency levels promotesbeneficial interaction. For instance, Kim & McDonough (2008) examined peer interactionof adult Korean learners and found that mixed pairs resolved language-related issuesmore successfully (especially for lexical LREs). However, in one of the earliest studieson interaction and proficiency, Yule & Macdonald (1990) found that different proficiencypairings resulted in more frequent and more successful interaction only if the lowerproficiency had a dominant role in the interaction. In Watanabe & Swain’s (2007) design,the learners were paired with learners with both higher and lower proficiency. Theiranalysis on the degree of collaboration (the quantity and quality of LREs) indicatedthat learning outcomes were mediated more by how collaborative a pair was than byinterlocutor proficiency levels. Having found a similar effect for collaborative interaction,Storch & Aldosari (2013) concluded that to understand the relationship between proficiencylevels and L2 learning, ‘it is not only proficiency difference which needs to be takeninto consideration, but also the kind of relationship learners form when working inpairs’ (p. 46).

Meanwhile, other researchers have paired learners with the same proficiency level andcompared high and low proficiency pairs (High vs Low). The findings from these studieshave been mixed as well. Williams (2001b) analysed the amount and types of LREs thatoccurred between adult learners of English. Results indicated that higher proficiency pairsproduced more LREs and benefited more from the LREs, as evidenced by learners’ scoreson tailor-made posttests. Williams claimed that higher proficiency learners can pay moreattention to formal aspects of the target language (see also Williams 2001a). Similarly, Nassaji(2013) found that advanced learners benefited more from focus on form episodes (FFEs,which are similar to LREs) than did beginner or intermediate learners, even though therewere not substantial differences in FFE frequency, with 39% occurring in beginner, 24% inintermediate, and 37% in advanced level classes.

However, Iwashita’s (2001) study with adult learners of Japanese failed to show such aneffect for proficiency. She quantified the amount of corrective feedback and subsequentmodified output that learners produced. The comparisons between pairs with differentproficiency levels did not detect any statistical differences. Furthermore, Oliver’s (2002)study of younger learners found that the lower the proficiency of a pair, the morenegotiation for meaning (more clarification requests and confirmation checks) that occurred.Contrary to Williams (2001b), Oliver argued that when a dyad’s proficiency is low, thereis ‘a greater chance that communication breakdown will occur and, hence, a greaterneed for the use of negotiation strategies’ (2002: 107). Sato & Viveros (2016) comparedintact EFL classes in a high school in Chile that had been divided into high and lowproficiency classes. Their analysis of the recorded communicative group tasks suggestedthat the learners in the lower class engaged in more language-related collaboration.The researchers explained that the difference was caused by the degree of collaborationrather than the interlocutors’ proficiency levels, supporting the claims made by Watanabe& Swain (2007) and Storch & Aldosari (2013) regarding collaborative interaction andL2 learning.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

2 9 8 S H A W N L O E W E N A N D M A S A T O S H I S A T O

3.4 Individual differences

There are a number of interlocutor individual differences that have been posited to influenceinteraction, although the amount of research varies depending on the characteristic (Robinson2005; DeKeyser 2012; Mackey et al. 2012; Baddeley 2015; Li 2017). Individual differencesthat have received more attention include (a) anxiety, (b) cognitive abilities, including languageaptitude, and working memory, (c) willingness to communicate, (d) learner beliefs, and (e)age.

Anxiety, which has been described as ‘the subjective feeling of tension, apprehension,nervousness, and worry’ (Horwitz, Horwitz & Cope 1986: 125), may affect L2 interaction.Specifically, Horwitz et al. refer to communication apprehension, a type of state anxiety thatoccurs in specific contexts, which may hinder learners’ ability to process input and produceoutput during L2 interaction (Sheen 2008). Considerable research has investigated L2 anxiety(see Dewaele 2017 for an overview); however, only a handful of scholars have examined theeffects of anxiety within the interactionist approach. One of the first, Sheen (2008) examinedanxiety and corrective feedback with ESL learners in the US, finding that low-anxiety learnersbenefited more from feedback than did high-anxiety learners. Similarly, Rassaei (2015) foundthat low-anxiety EFL learners were more likely to perceive the corrective nature of feedbackthan were high-anxiety learners.

However, not all interaction studies have found an influence for anxiety. For example,Revesz (2011) used questionnaires to assess learners’ anxiety and investigate any moderatingeffects on learners’ production of accurate and complex language during simple and complextasks. Surprisingly, Revesz found no impact for high anxiety on task performance. Finally,in a study investigating the effects of interaction context on L2 anxiety, Baralt & Gurzynski-Weiss (2011) compared the levels of anxiety of L2 Spanish learners as they engaged inone-on-one task-based interaction with their instructor either face-to-face (FTF) or throughcomputer-mediated communication (CMC). They used an anxiety questionnaire, which theyadministered to learners half-way through the task, and then again at the end of the task.They found no differences in anxiety levels between modalities or at the two different timesof measurement. While these studies indicate that anxiety may influence the effectivenessof interaction for L2 acquisition, they also point to the need for more studies to investigatefactors affecting learner anxiety in order to ameliorate them during interaction.

Another area of individual differences that has begun to receive attention in interactionistresearch is cognitive differences in, primarily, language learning aptitude and workingmemory. More generally, language learning aptitude has a relatively long research historyin SLA (Carroll 1981, 1990) and has been defined as a set of cognitive abilities, such aslanguage analytic ability, phonetic coding ability, and rote memorization, that are ‘predictiveof how well, relative to other individuals, an individual can learn a foreign language’ (Carroll& Sapon 2002: 23). Similarly, working memory, which ‘involves the temporary storage andmanipulation of information that is assumed to be necessary for a wide range of complexcognitive activities’ (Baddeley 2003: 189), has a research tradition within SLA (see Juffs &Harrington 2012 for an overview), as well in the interactionist approach (Mackey et al. 2002;Sagarra 2007; Goo 2012; Mackey & Sachs 2012; Revesz 2012). There have been suggestionsthat working memory may be a component of language learning aptitude (Li 2017), but

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 299

regardless, the important consideration for the interaction approach is the relationshipbetween cognitive abilities and the effectiveness of L2 interaction.

The few studies of language aptitude and interaction seem to suggest that higher aptitudemay be beneficial for interaction because of the generally more implicit nature of attention tolanguage during interaction. One study by Li (2013) investigated the effects of cognitivedifferences on the effectiveness of implicit and explicit corrective feedback on implicitand explicit L2 knowledge of Chinese classifiers. For language analytic ability, which heoperationalized as sensitivity to grammatical structures, as measured by the Words inSentences subtest of the Modern Language Aptitude Test, Li found that language analyticability was a significant predictor of grammaticality judgement test gain scores for the implicitfeedback group but only on the delayed posttest. He concluded that ‘in the absence ofmetalinguistic information, learners with higher analytic ability achieved more’ (2013: 647).Thus, language learning aptitude may help compensate for the lack of explicit attention tolanguage, whereas such aptitude is less important in more traditional classroom activities inwhich direct and explicit information is provided about the target language (see also Long1996; Ranta 2002).

In a study of a suite of individual cognitive differences, Trofimovich, Ammar & Gatbonton(2007) examined learners’ working memory, phonological memory, analytical ability, andattention control in relation to learners’ ability to notice and benefit from recasts. Thirty-twoFrancophone learners of English received recasts on both grammatical and lexical items.Overall, learners were able to both notice and benefit from recasts, although there wasconsiderable individual variability. The individual differences found to account for the gainscores were attention control, phonological memory, and analytical ability, with attentioncontrol being the strongest predictor. Trofimovich et al. concluded that ‘in L2 interaction,attention control may characterize learners’ ability to efficiently switch attention amongdifferent aspects of language or among different cognitive tasks, thereby determining learners’success in using interactional feedback’ (2007: 194).

In terms of working memory, the hypothesis is that learners with greater working memorycapacity may benefit more from interaction than learners with smaller working memorybecause higher capacity learners will be able to hold and access more linguistic information intheir short-term cognitive system, which is particularly useful when, for example, comparinginterlocutor feedback with one’s own erroneous production (Mackey et al. 2002; Sagarra2007; Goo 2012; Mackey & Sachs 2012; Sagarra & Abbuhl 2013). In Revesz (2012), EFLlearners in Hungary were divided into three groups: recast, non-recast, and control. L2development was measured via a grammaticality judgement test, a written production task,and oral production tasks, while learners’ working memory was examined by digit span,non-word span, and reading span tests each of which tapped into different subcomponents ofworking memory. Results showed that the subcomponents of working memory mediated theeffects of recasts in different areas of L2 development. For example, significant correlationswere detected between the gain scores of the oral description task and the digit span andnon-word span tests, suggesting that, learners with high phonological short-term memorybenefited from recasts in the area of oral production. Li’s (2013) previously mentionedstudy also investigated working memory (operationalized using a listening span test) as amoderating variable in the effectiveness of implicit and explicit feedback on English L1

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 0 0 S H A W N L O E W E N A N D M A S A T O S H I S A T O

learners’ implicit and explicit knowledge. He found that working memory was a significantpredictor of grammaticality judgement test and elicited imitation test gain scores for theexplicit feedback group but only on the delayed posttests. In another study, Kim, Payant& Pearson (2015) examined the impact of task complexity and working memory on thenoticeability of recasts as well as the development of question formation. Regardless of thetask condition (simple or complex), the learners received recasts during task-based interactionwith a native speaker. They were instructed to repeat the last-heard utterance when thenative speaker knocked on the desk twice. Noticing was measured by type of response afterthe knocks (e.g. full repetition of the recast). The analysis of oral production tests given threetimes revealed that working memory was the only predictor of the level of noticing as wellas L2 development, suggesting ‘the benefits of high WM [working memory] on the learningof morphosyntactic features during interactional tasks while receiving feedback’ (Kim et al.2015: 574).

The relationship between interaction and psychological individual differences has been agrowing, yet under-investigated, area of research. For example, the link between motivationand interaction has largely been overlooked (Dornyei 2002). In particular, task motivation,which explains ‘why students behave as they do in a specific learning situation where theyare carrying out a specific task’ (Csizer 2017: 424–425), has been under-investigated (seeDornyei & Kormos 2000; Kormos & Prefontaine 2016; Poupore 2016). Also, willingness tocommunicate (WTC), which concerns ‘the probability that a learner will USE the language inauthentic interaction’ (emphasis in original: MacIntyre et al. 1998: 558), could considerablyimpact learners’ interaction behaviour (see MacIntyre, Burns & Jessome 2011). Cao and Philp(2006) compared learners’ self-reported WTC and their classroom behaviour, finding littlecorrelation between the two. In interviews, learners reported group size and self-confidenceas two important factors influencing WTC behaviour. More recently, Cao (2014) argued thatlearner WTC is better viewed as dependent on ‘dynamic situational variables’ (p. 789) ratherthan as a learner trait characteristic.

Another individual difference that has received some attention pertains to individuals’beliefs about interaction. The underlying assumption is that if teachers and learners valueinteraction and see it as a useful activity, then they are more likely to benefit from it. However,if learners think that an L2 classroom should consist of explicit grammar and vocabularyinstruction accompanied by different drills, then such learners may devalue interaction andbe less likely to benefit from it. In two early studies, Schulz (1996, 2001) examined the beliefsof students and teachers in the US and Colombia about grammar instruction and errorcorrection. She found that students in both contexts valued error correction; however, teacherswere much less positive about the provision of corrective feedback. Schulz commented thatsuch a mismatch could possibly hinder language learning if students’ expectations are notmet. In another study, Loewen et al. (2009) investigated the beliefs of over 700 learners ofvarious languages at a US university. They found that learners viewed corrective feedbackand grammar instruction as different constructs, which is positive given the different effectsof grammar instruction and corrective feedback on the development of L2 knowledge.Furthermore, Loewen et al. found that learners studying different languages had differingbeliefs about interaction and corrective feedback. For example, ESL learners were the mostpositive about interaction compared to foreign language learners. Such positive beliefs may

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 301

incline learners to be more receptive to interaction in the classroom, thereby increasingits benefits (see also Lasagabaster & Sierra 2005; Brown 2009). Importantly, unlike otherindividual differences pertaining to individuals’ capacities (e.g. WM), learner beliefs maybe altered by instruction. Sato’s (2013) ten-week intervention, designed to raise learners’awareness of the effectiveness of peer interaction and peer corrective feedback, showed thatlearner beliefs positively changed over time, which was reflected in the increased provisionof peer corrective feedback. Certainly, intervention research designed to enhance positivepsychology is called for (see MacIntyre & Gregersen 2016).

Learners’ age is another individual difference that is important to ISLA as instructorsmay adjust their teaching strategies according to the students’ age groups. While thereexists abundant research as to how the age of acquisition affects the ultimate attainment(see DeKeyser 2017b), investigation of the relationship between age and L2 interaction hasbeen limited partly because of the methodological difficulty in teasing apart L2 developmentfrom children’s general psychological and emotional development as well as L1 development(Oliver, Nguyen & Sato 2017). However, Oliver and colleagues’ studies (Oliver 2000; Oliver& Mackey 2003; Mackey, Kanganas, & Oliver 2007) have shown that children and adultlearners benefit from interaction in different ways. For instance, Oliver (1998) comparedher dataset of 196 children, aged from 8 to 13, with that of the adult learners in Long(1983). While both age groups negotiated for meaning, Oliver found that the child learnersused fewer clarification requests and confirmation checks. More recently, Oliver, Philp &Duchesne (2017) compared interactions between 11 pairs of younger (5–8 years) and 10 pairsof older (9–12 years) learners. Among several key differences, Oliver et al. identified learner’sengagement as a mediating factor of the effectiveness of interaction. For example, dependingon the topic of the task, older learners ‘simply wanted to get the task done’ (2017: 8), resultingin fewer instances of negotiation for meaning.

It is clear that learner individual differences may affect learner interaction, as well asthe effectiveness of that interaction. However, there is still much that is unknown aboutindividual differences and interaction. Only a few variables have been investigated, and thesewith only a handful of studies. In general, learner psychology requires further research asthe way in which L2 learners approach communicative tasks may influence their interactionbehaviour and determines the ultimate benefit of interaction (Sato 2016). Furthermore,individual differences such as personality and learning styles can reasonably be assumed toinfluence interaction (Dewaele 2017). How do more introverted or extroverted learners feelabout and engage in interactive tasks? What are the effects of using more social learningstrategies on interaction? One other new area of interaction research is individual creativity,with some studies suggesting that creativity can influence linguistic production during task-based interaction (McDonough, Crawford & Mackey 2015). These areas warrant furtherresearch.

4. Task characteristics

Apart from interlocutor characteristics, there are numerous task features that may impactinteraction (e.g. Doughty & Pica 1986; Robinson 2001; Pica 2005; Keck et al. 2006; Pica,

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 0 2 S H A W N L O E W E N A N D M A S A T O S H I S A T O

Kang & Sauro 2006; Kim 2008; Tavakoli & Foster 2008; Jenks 2009; Kormos & Trebits2012). In fact, in the task-based language learning and teaching literature, there has beenconsiderable interest in the characteristics of tasks and the impact of these characteristics oninteraction as well as L2 development (e.g. Bygate et al. 2001; Samuda & Bygate 2008; Ellis2017b). Again, while some studies, particularly those conducted in the earlier years of theinteractionist approach, were more descriptive in nature, there has been a growing concernwith the impact of different types of tasks on interaction patterns and L2 development (e.g.Gurzynski-Weiss & Revesz 2012; Baralt 2013; Nassaji 2013).

There have been several different conceptualizations of what constitutes a task (e.g. Skehan1998; Ellis 2003; Long 2015); however, there is some consensus that several key features mustbe included in a task. The main focus of a task should be on communication and meaningrather than linguistic forms. Learners should use primarily their own linguistic resources, andthe goal of the task should be non-linguistic in nature. This emphasis on communication isbroadly in line with an interactionist approach.

When one considers task design, there are several key features to take into account, such astask conditions (e.g. information configuration, interactant relationship, and goal orientation),task complexity (e.g. more versus less complex), and task participation structure (e.g. dyad,small group, or whole class).

4.1 Task conditions

The ways in which information is distributed among learners, and the ways in which theymust exchange information can impact learner interaction (Doughty & Pica 1986; Foster &Skehan 1996; Ellis 2003; Pica et al. 2006; Gilabert, Baron & Llanes 2009). For example,information configuration and interactant relationship refer to how the task information isheld and shared among the interlocutors. In one-way tasks, generally one person has the taskinformation and provides that information to other participants. However, in two-way tasks,the information is shared among the learners, and they must exchange that information witheach other to complete the task.

In terms of goal orientation, tasks may be convergent, in which learners must agree ona specific outcome, or divergent, in which learners may hold different opinions regardingthe task outcome. In general, information gap tasks are convergent, while opinion gaptasks are divergent. For example, Gilabert et al. (2009) investigated the effects of differenttasks types on interactional features such as negotiation of meaning, recasts, and LREs.They found that a convergent map task, with only one possible solution, resulted in morenegotiation of meaning than did more divergent tasks. Similarly, Gass et al. (2005) foundthat task type was the primary factor that significantly affected interaction in their study ofL2 Spanish learners. Their consensus task resulted in learners’ interaction containing fewerinteractional characteristics than did their picture differences and map tasks, a differencethat the researchers attributed to presence or absence of required information exchange.In the consensus tasks, both interlocutors had only to provide their opinion about a setof information that they shared in common. In the other two tasks, the interlocutors haddifferent pieces of information that they had to exchange in order to complete the task.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 303

The necessity of exchanging information, rather than expressing opinions, resulted in moreinteraction.

However, Newton (2013) found slightly different results in his investigation into vocabularyuse during convergent, information gap tasks and divergent opinion gap tasks. He found thatin the information gap task, one of which asked learners to exchange information about thelocation of animals in a zoo, learners were less likely to negotiate meaning than when theyhad to share their opinions. Newton suggests that the learners were able to complete theinformation gap task rather mechanically, while in the opinion gap tasks, the learners had tofocus more on the meaning of each other’s utterances.

In general, the amount of information each learner holds and the way in which theinformation is exchanged during interaction have an impact on interaction behaviour. Asthose features can be manipulated by task designs, this research warrants further investigationso as to provide classroom learners with more effective tasks.

4.2 Task complexity

Another task characteristic that has received considerable attention is task complexity (e.g.Robinson 2007, 2011; Revesz, Sachs & Hama 2014), which Gilabert & Baron (2013) defineas ‘the internal cognitive demands that language tasks impose on learners’ processing andlanguage use’ (p. 45). Thus, task complexity is affected by the attention, memory, andreasoning demands imposed by the task. Task complexity has been operationalized in severaldifferent ways. In one study, Kim (2012) used +/− reasoning demands and +/− fewelements, in tasks with three levels of complexity (simple, +complex, and ++complex).For instance, to complete a task called ‘university students’ part-time job,’ learners in thesimple group had to exchange information about university students provided on activitycards, while learners in the ++complex group were required to recommend appropriatejobs for those students based on their profiles. In another study, Revesz (2009) used +/−contextual support, in which learners had to narrate stories either with or without theaid of accompanying photos. Finally, Baralt (2013) used +/− intentional reasoning, which‘requires reflection on the intentional reasons and cognitive mental states that cause otherpeople to do certain actions’ (p. 698). For instance, in completing a story-retelling task,the simple-group learners were provided with a series of pictures with thought bubbleswhich contained information; however, this information was hidden for the complex-grouplearners.

The results of studies on task complexity and interaction have been mixed. For example,Baralt (2013) examined learner interaction on more or less complex tasks in both FTF andwritten synchronous CMC contexts, with learners receiving recasts on their erroneous useof the Spanish past subjunctive. She found that learners perceived difficulty based on themodality of interaction rather than on +/− intentional reasoning. Thus, learners thoughtthe task was more difficult if they had to do it FTF, even if they had to provide their ownintentional reasoning. Also, engaging in the more complex task resulted in more learning forthe FTF group but not the CMC one. Baralt argues that the combination of the recasts andthe CMC environment, especially the disjointed feedback that occurs in written chat, was

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 0 4 S H A W N L O E W E N A N D M A S A T O S H I S A T O

cognitively overwhelming for the learners, thereby blocking the expected benefits of morecomplex tasks.

In another study, Gilabert & Baron (2013) investigated task complexity and pragmaticmoves in dyadic interaction. They triangulated their own operationalization of complexity(i.e. tasks with more or less complex reasoning demands and time on task) with learners’impressions of difficulty, based on the amount of time learners thought they had spentperforming the task compared to the actual time on task. All three measures were found tocorrespond to each other. As for the effects of complexity on pragmatics, learners produceda greater number of pragmatic moves in the more complex tasks; however, there was atask effect for the types of pragmatic moves that were used. Conditionals were used morefrequently in the complex task than in the simple one, and want statements occurred in thesimple task but not the complex one.

Task complexity has received much attention from cognitive-interactionist researchers.Nonetheless, the research often lacks clear pedagogical implications, especially with taskcomplexity being operationalized in multiple ways. Future research may benefit fromexamining actual classroom materials or by implementing tasks with differing complexitiesin real learning settings.

4.3 Task participation structure

Task participation structure includes whole class, small group, and dyadic interaction.Generally, whole class interaction, and sometimes small group interaction, involves learnersinteracting with the teacher, while dyadic interaction, and small group interaction generallyinvolves learners interacting with their fellow L2 speaking peers. Another issue is whetherlearners are direct participants in or observers of the interaction.

Nassaji (2013), in his investigation into task participation structures, found that incidentalfocus on form occurred more frequently in whole class interaction. Also, preemptive FFEsoccurred more frequently than reactive ones. However, student-initiated FFEs occurred morefrequently in small group and dyadic interaction, while teacher-initiated FFEs were morecommon in whole class interaction. Nassaji investigated the effects of FFEs by using tailor-made post-hoc tests. Overall, students responded correctly 58% of the time; however, accuracywas greater if the FFEs occurred in dyads or small groups (69% and 66% respectively) thanwhole class interaction (48%). Furthermore, student-initiated FFEs had a 72% accuracy rate,while teacher-initiated had only a 46% rate. Nassaji concluded that ‘although the majorityof FFEs occurred in whole class interactions, learners were more likely to benefit from FonF[focus on form] when it occurred in small group and one on one interactions’ (2013: 861–862). Whole class contexts may limit learners’ opportunities for interaction and attention tolanguage, in part because more individuals are involved in the interaction.

There are other aspects that influence interactional participation structures. For example,McDonough & Hernandez Gonzalez (2013) found that pre-service teachers tended todominate interaction in conversation groups organized by the researchers specifically toprovide communication opportunities for the learners. The groups had between 2 and 16learners, with a median of 8; consequently, the groups may blur the line between whole class

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 305

and small group interaction. Nevertheless, learners produced less than 25% of the total words,suggesting that interaction with a teacher may not provide the best interactional opportunities.Alcon’s (2002) comparison between peer interaction and learner–teacher interaction alsoexhibited a similar phenomenon in the classroom. Her observation of EFL classes (oneteacher-led and the other peer interaction focused) indicated that learners employed morerequesting strategies with each other (i.e. a type of feedback) than when they interacted withthe teacher (see also Toth 2008).

Furthermore, in small group and whole class interaction, it is possible that not all learnersengage directly in interaction or negotiation for meaning (e.g. Foster 1998). For example,Loewen (2003) observed that not all students in his New Zealand ESL classes participatedequally in FFEs. In fact, roughly 13% of students were not involved in any direct exchanges,suggesting that any benefit for them would have to come through observing the interactionof others. Indeed, recent research has begun to investigate the effects of learners engagingdirectly in interaction or being an observer of interaction. For instance, Fernandez Dobao(2016) focused on a silent learner during group work. By examining participation patterns(LREs) as well as their subsequent impact on L2 learning, she discovered that the silentlearner benefited from interaction as much as those who actually produced language. Sheargued that being active listeners (observers of LREs) can help learners acquire new linguisticknowledge, possibly by engaging in sub-vocal private speech.

In another study, Yilmaz (2016) examined the effects of learners receiving feedback onerrors that they themselves made (direct condition) or errors that their fellow students hadmade (indirect) on two different Turkish linguistic structures. Yilmaz found that for theplural morpheme both receivers and non-receivers outperformed a comparison group onimmediate and delayed production tests, but only the receivers outperformed the comparisongroup on the recognition test. In addition, the receivers outperformed the non-receivergroup on all but one test. However, for the locative morpheme, there was no differencebetween the receivers and non-receivers. Also, neither group outperformed the comparisongroup. In another study, Philp & Iwashita (2013) examined learners in task-based dyadicinteraction to see what interactors and observers noticed, and to see if their role in theinteraction affected noticing. Seventeen students interacted, while nine observed a DVD ofinteraction. Noticing at the level of awareness was measured through stimulated recall and exitquestionnaires. Interactors commented more on the target (French gender agreement/passecompose) and other linguistic structures than did the observers. The observers commentedmore on comprehension, participants’ feelings, and task strategies. Half of the participants,regardless of group, reported noticing the target forms after or during participation.

Due to the somewhat differing results, more research is needed to explore the differencesbetween engaging directly in or observing interaction, especially in contexts where interactionoccurs more often in whole class contexts. In this sense, Batstone & Philp’s (2013) discussionof private spaces may be relevant. The researchers distinguished classroom interaction inpublic and private spaces. Public spaces are open and include teacher-led interaction as wellas subgroup work in which discourse is meant for all members of the group. Private spacesincluded only specific members and are not open to the larger group. Batstone & Philpargued that such private talk has a compensatory function in that it allows learners to relatethe larger interaction to their own learning needs.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 0 6 S H A W N L O E W E N A N D M A S A T O S H I S A T O

5. Linguistic targets

The role of linguistic structures during interaction varies, with some tasks allowinginterlocutors to target linguistic structures incidentally, while other tasks attempt to drawlearners’ attention to specific forms (Ellis 2003). Additionally, not all aspects of language havebeen found to benefit from interaction equally (e.g. Mackey & Goo 2007).

One meditating factor impacting learners’ success in L2 development during interactionmay be the saliency of certain linguistic features. Saliency of a particular linguistic structureduring interaction can be controlled by instructional interventions, for instance, givinglearners tasks designed to elicit specific linguistic features (i.e. focused tasks), exposing them tocertain features to facilitate production of the same structures later (i.e. priming), or drawingtheir attention to certain features by providing corrective feedback. In this sense, DeKeyser(2012) discussed the relationship between instruction and linguistic structures and argued that‘different structures have a differential need for instruction, in particular instruction designedto enhance their salience (by drawing learners’ attention to the existence, patterning, andimportance of the non-salient structures)’ (p. 196). Additionally, Long (2015) suggests that theperceptual saliency of linguistic structures combined with learners’ sensitivity to input canaccount for ‘success and failure’ (p. 60) in acquiring individual structures (see Gass, Spinner& Behney 2018 for a recent volume on saliency).

5.1 Incidental linguistic features

Research suggests that L2 learners, especially adult learners, struggle in acquiring featuresthat are non-salient in input (e.g. VanPatten 1990, 2017a). This difficulty is evident from abody of research revealing that learners’ attention is drawn to certain features more thanothers when there is no external manipulation, regardless of the target language (e.g. English:Mackey 2006; French: Ayoun 2001; Japanese: Loschky 1994). In general, learners tendto focus on lexical items over other aspects of language. For example, Williams (2001a)observed adult intensive ESL classes and found that more than 80% of LREs were lexicalrather than morphosyntactic in nature. Similarly, Jeon’s (2007) quasi-experimental study ofEnglish-speaking learners of Korean showed that conversational interaction was morebeneficial for lexical features (nouns and verbs) compared to morphosyntactic ones (objectrelative clauses and subject–verb agreement). The influence of selective attention can beobserved within different grammatical structures as well. For instance, Shintani (2014a)compared acquisition of English plural –s and copula be by providing young Japanese learnersof English with focus on form tasks. The analysis of the productive knowledge tests showedthat after nine lessons of 30 minutes, the learners gained knowledge of the plural –s but notcopula be.

The differential impact of interaction on L2 learning depending on linguistic features hasseveral explanations. First, the amount of exposure during interaction might be a factoraffecting which features are more readily acquired. Collins et al.’s (2009) corpus studyexamined different degrees of saliency depending on linguistic structures when L2 learnerswere aurally exposed to them. The researchers compared different grammatical structures

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 307

(the simple past tense, the progressive, and the possessive determiners) and discoveredthat more difficult-to-acquire structures were not necessarily salient in the classroom input,implying that infrequent input may partially explain why learners struggle with those features.Second, the communicative value of linguistic features affects the developmental patterns ofdifferent features. That learners tend to attend to lexical items more than grammatical itemscan be explained by the fact that nouns carry more important value when negotiating formeaning (Spada & Lightbown 2008). In the same vein, Shintani (2014a) argued that learnersacquired plural –s but not copula be because –s was necessary to complete the task. Third,the acquisition of particular features can be promoted by their saliency because learners’attention is drawn to more salient features. Perceptual saliency, however, is susceptible to therelationship between a learner’s L1 and L2 as some morphosyntactic features may or may notexist in the learner’s L1 system (see Tokowicz & MacWhinney 2005). Also, learner readinessmay affect which features are more salient for a particular learner (Lightbown 2013).

5.2 Targeted linguistic features

While incidental focus on form does promote L2 learning (Loewen 2005), learning ofdifferent linguistic features can be facilitated by providing learners with focused tasks. Focusedtasks elicit specific linguistic features, either by design or by the use of methodologicalprocedures that focus attention to form during the implementation of the task. Ellis (2003)separates focused tasks into three types: structure-based production, comprehension tasks, andconsciousness-raising tasks, among which structure-based production pertains to interaction.Structure-based production tasks are designed to incorporate specific linguistic features forlearners to reproduce (e.g. dictogloss, picture differences, picture description, and storycompletion tasks). For example, Fortune (2005) examined four dictogloss tasks. Eight advancedEFL learners from various L1 backgrounds worked in pairs to reconstruct texts. Althoughlearners were not required to use the same type of syntactic structure or lexical items fromthe text, the analysis of the LREs revealed that the learners’ attention was successfully drawnto the linguistic targets as shown by the metalanguage they produced (see also Leeser 2004).

There exist pedagogical challenges in eliciting specific linguistic structures through focusedtasks. First, tasks often do not work as the teacher planned. Breen (1987) observed differencesbetween outcomes derived from task design (task-as-workplan) and actual outcomes derivedfrom learners (task-in-process). He argued that this divergence resulted from an unpredictableinteraction between the learner, task, and situation. In other words, learners may arrive ata task with different assumptions about their roles and the demands of the task. Second,tasks may not work in the way they were designed (but see Ellis 2017b). Collins & White(2014) analysed classroom interaction of Grade 6 ESL learners in Quebec and discoveredthat certain features (e.g. possessive determiners) were more difficult to elicit even when thetask was designed to do so (e.g. talking about family members). Consequently, relying on tasksto elicit specific linguistic features may be less than ideal.

Priming may be another effective way of manipulating learners’ production of specificlinguistic features. Among different types of priming (see Trofimovich & McDonough 2011),syntactic priming concerns elicitation of specific grammatical features, and is described by

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 0 8 S H A W N L O E W E N A N D M A S A T O S H I S A T O

McDonough & Mackey (2008: 32) as ‘a speaker’s tendency to produce a syntactic structureencountered in the recent discourse, as opposed to an alternate structure’. In order to testwhether priming is effective in eliciting certain grammatical structures, researchers select astructure with several alternative forms. Then, learners are provided with primes of onlyone alternative to see if their production patterns change and they start to use the primedstructure. For instance, Thai EFL learners in McDonough & Chaikitmongkol’s (2010) studyoften alternated between two wh-question structures, one of which is grammatical (e.g. how

do people damage their health?) and the other ungrammatical (e.g. how people damage their health?).The learners were provided with grammatically correct primes, which led to significantlymore production of the grammatically correct alternative. More recently, McDonough,Trofimovich & Neumann (2015) tested three targets in the classroom context and revealedthat priming was successful for relative clauses and adverbial clauses but not for passives. Theysuggested that passives were easier to elicit in English for academic purposes classes and thusdid not yield a significant difference between the control and experimental groups. In otherwords, the pedagogical strength of priming activities lies in the fact that they ‘help studentsproduce a difficult or infrequent structure’ (McDonough et al. 2015: 78) in the classroomcontext.

It is perhaps unsurprising that vocabulary should figure so prominently in meaning-focusedinteraction, especially when learners are left to their own devices. Consequently, there may beless of a need for manipulation of lexical items before, during, and after interactive tasks, anda greater need to draw learners’ attention to grammar. For example, Newton (2013) arguesthat ‘communication tasks [are] a potentially valuable source of opportunities for incidentalvocabulary learning’ (p. 165) because learners are likely to encounter unfamiliar words.Newton examined learners’ negotiation of lexical items in opinion gap and information gaptasks. He found gain scores of roughly 4 words per 30 minutes of task. However, there was alsoconsiderable improvement on vocabulary items that were not negotiated, leaving Newton tosuggest that negotiation of meaning is only one way that learners can learn vocabulary duringcommunicative tasks. He suggests that encountering words in context, as well as observingnegotiation can have an influence. Zhao & Bitchener’s (2007) observation of a university-levelESL class also revealed that vocabulary was the most attended linguistic aspect during bothlearner-learner and learner-teacher interactions, a finding that was confirmed by Simard &Jean (2011) who observed L2 classes of young learners (Grade 9 to Grade 12) of English andFrench.

Among a variety of focus on form techniques that teachers can utilize, corrective feedbackhas generated a considerable amount of research, as can be seen in the number of recent meta-analyses of corrective feedback (e.g. Russell & Spada 2006; Li 2010; Lyster & Saito 2010;Brown 2016). Two issues are relevant as far as different linguistic features and correctivefeedback are concerned: (a) which linguistic features do teachers tend to react to more and (b)which linguistic features benefit most from feedback? First, a body of classroom observationstudies examining different types of linguistic targets in different languages (see Loewen 2012)suggests that teachers tend to provide more corrective feedback on morphosyntactic errorsthan on other types of errors. Brown’s (2016) meta-analysis of 21 studies showed that 43%of corrective feedback was directed to morphosyntactic errors, followed by lexical (28%),and phonological errors (22%). The reason for this tendency is as yet unknown. However,

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 309

Brown’s (2016) moderator analysis revealed that the teaching context (second vs foreignlanguage contexts) affected the linguistic foci: teachers in foreign language contexts focusedon grammatical errors more than those in L2 teaching contexts. Brown’s study also foundthat teachers with more experience tended to focus on lexical rather than pronunciationerrors, suggesting a change in teachers’ beliefs about which linguistic features to correct(Basturkmen, Loewen & Ellis 2004).

Contrary to teachers’ tendency to provide feedback more on morphosyntax errors thanon lexical errors, feedback effectiveness is more observable in the development of lexisthan morphosyntax. Noticeability may explain this gap. Mackey, Gass & McDonough’s(2000) stimulated recall data showed that feedback that was intended to correct lexical,semantic, and phonological errors was perceived more accurately than feedback targetingmorphosyntactic errors (see also Egi 2007). In this sense, the nature of different linguisticfeatures may affect the degree of noticeability. Among several studies that compared differentlinguistic targets, Yang & Lyster (2010) examined corrective feedback on English regularand irregular past tense. Having found differential impacts on the two forms, the researchersinvoked Skehan’s (1998) dual-mode system which distinguishes an analytic rule-based system(e.g. regular past tense) and a memory-driven exemplar-based system (e.g. irregular pasttense). Yang & Lyster argued that the noticeability of irregular forms (e.g. bought) was higherthan regular (e.g. shopped) and thus feedback affected the development of lexical items moreeasily than rule-based items regardless of type of feedback (recasts and prompts). In anotherstudy, Yilmaz (2016) found different effects for corrective feedback on Turkish plural andlocative morphemes, which he attributed, in part, to their morphophonological differences.The plural morpheme has two allomorphs, while the locative has four. Yilmaz also suggeststhat the relationship between the participants’ L1 (English) and Turkish might also play apart. English and Turkish plurality is similar, but locative expression is not. Finally, Li (2014)investigated relationships between feedback types, linguistic targets (Chinese noun classifiersvs perfective aspect particle le), and learners’ proficiency levels. The results revealed thatfeedback effectiveness was mediated by linguistic targets. Li argued that proficiency levelsmay also have played a role. Because learners with higher proficiency had more cognitiveresources, they were able to take advantage of feedback given on the non-salient aspect makers.

Several studies of focus on form have found that pronunciation is not targeted frequentlyduring classroom interaction (e.g. Ellis et al. 2001a), which may be somewhat unexpecteddue to the influence of pronunciation in learner comprehensibility (Kennedy & Trofimovich2017). Similarly, there has been a lack of research studies focusing on interaction andpronunciation; however, this lacuna is being addressed (see special issue of Studies in Second

Language Acquisition Gurzynski-Weiss, Long & Solon 2017). For example, Parlak & Ziegler(2017), in an investigation of the effects of recasts on English lexical stress errors in FTFinteraction and oral synchronous computer-mediated communication (SCMC), found nostatistical differences in comparisons with control groups. However, the researchers foundthat the FTF recast group did alter vowel duration following feedback. Additionally, resultsrevealed that learners perceived the SCMC context as being less stressful and more helpfulfor language learning. In the same issue, Loewen & Isbell (2017) investigated pronunciation-focused LREs in learner–learner FTF interaction and oral SCMC, finding that only 14%of LREs addressed pronunciation with 90% targeting segmental rather than suprasegmental

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 1 0 S H A W N L O E W E N A N D M A S A T O S H I S A T O

features. Prior to these studies, much pronunciation-focused interaction research has beenconducted by Saito, who in a series of studies (Saito & Lyster 2012; Saito 2013; Saito& Wu 2014; Saito 2015), found mixed results for the effects of interaction and correctivefeedback on pronunciation. For example, Saito (2013) found added benefits for form-focusedinstruction combined with recasts for L1 Japanese learners of English and their productionof /ɹ/. However, Saito & Wu (2014) found only marginal effects for the addition of correctivefeedback to form-focused instruction on Cantonese L1 speakers’ perception of Mandarintones. Finally, in another study, Lee & Lyster (2015) targeted perception of the /i-I/ distinctionin English, comparing instruction alone with instruction plus corrective feedback. Participantswere given five hours of instruction over five days, and the researchers found that the correctivefeedback group improved more than the instruction alone group. These results suggest thatpronunciation during interaction warrants more investigation.

Pragmatics is an area of linguistics that has been generally overlooked by interactionistresearchers, especially in comparison to grammar and vocabulary. Consequently, Mackeyet al. (2012) point to the need for further research in this area due to ‘the fact that fewstudies investigating the effects of interaction have addressed pragmatic and phonologicaltargets’ (p. 10). For instance, Mackey & Goo (2007) were not able to include any pragmaticsstudies in their meta-analysis of interaction research; the studies included investigated onlylexis and grammar. However, there have been several studies that have examined theeffects of interaction on pragmatic development. For example, Alcon (2002) investigated thedevelopment of requests in teacher–student versus student–student interaction in SpanishEFL classes. Learners were presented with role-play contexts that necessitated the useof requests. In addition, they were provided with explicit instruction and recordings ofinteraction focused on requests. Learners in both groups improved in their use of requests;however, the student–student groups engaged in more interaction, while the teacher providedcorrective feedback more often in the teacher-led group. In another study, Sykes (2005)compared the effects of interaction in three different modalities (written chat, oral chat, andFTF) on L2 Spanish learners’ development of invitation refusals. Sykes found that participantsin the written chat group outperformed the other two groups in producing more native-likerefusals, a result she attributed to the additional processing time available in written chatinteraction as well as the need for more linguistic responses due to the lack of paralinguisticfeatures such as intonation and laughter. More recently, Gilabert & Baron (2013) examinedthe effects of task complexity on the production of request and suggestions by English L2learners in Spain. They found that learners produced more requests in more complex tasks,and the linguistic forms used to produce the requests varied according to tasks. There is clearlya need for more investigation into the effects of interaction on the use and development ofpragmatics (for a review, see Bardovi-Harlig 2017). Indeed, it is conceivable that task-basedinteraction could provide interactional opportunities for L2 learners that would allow them toexperience a range of social contexts that are not typically found in the classroom (Eisenchlas2011). In many cases, these social contexts might need to be artificially created.

Finally, very few studies within the cognitive-interactionist paradigm have investigateddiscourse level phenomenon or features that might be considered markers of interactionalcompetence, such as turn-taking. One exception is Ziegler et al. (2013) who looked atthe development of conversational features, such as overlapping speech, turn-taking and

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 311

floor-holding, during naturally occurring L2 German conversation groups. The study foundthat learners who adapted to the more German conversational style became more activeinteractors, while those who failed to do so became more passive, but they were still able toattend to the input and benefit from the interaction. All learners viewed the conversationgroups as a positive experience. Due to the paucity of research in this area, there is a needfor research that investigates the discourse level components of interactional competence todetermine if the benefits of meaning-focused interaction that have been found to generallyapply to grammar and lexis also apply to larger discourse level aspects.

6. Interactional context

6.1 Instructional setting

In the late 1990s and early 2000s, a series of studies investigated the occurrence of interactioncharacteristics in different instructional contexts. One of the earliest was Lyster & Ranta’s(1997) description of the frequency and types of corrective feedback that occurred duringcommunicative activities in a French immersion high school context in Canada. They foundthat the teachers provided a variety of feedback, which was subsequently categorized intosix types, with recasts being the most frequent. The researchers also identified differentlearner responses in reaction to feedback (i.e. uptake). In following analyses of the samedata set, Lyster (1998a, 1998b) examined more closely the types of corrective feedback thatoccurred, as well as the types of output that learners provided in response to different typesof feedback. Results indicated that learner uptake occurred more frequently after output-prompting feedback, suggesting that recasts were less useful than other types of feedbackduring interaction in promoting L2 production.

A series of studies in New Zealand ESL classes for young adults (Ellis et al. 2001a, 2001b;Loewen 2004) also examined the types of corrective feedback and output responses thatoccurred during interaction. Results indicated that recasts were the most common typeof corrective feedback, and successful uptake of the feedback occurred roughly 75% ofthe time. In a synthetic analysis of the data from Lyster’s and Ellis et al.’s studies, Sheen(2004) compared the interaction that happened in those contexts with her own data fromcommunication classes in Korea. She found that uptake rates in the Canadian context(Lyster & Ranta, French immersion: 54.8%; Panova & Lyster, Canadian ESL: 46.6%) weresignificantly lower than in New Zealand (80.4%) and Korea (82.3%). A similar tendency wasfound for the relationship between uptake and recasts: students in Korean and New Zealandcontexts responded significantly more to recasts (82.5% and 72.9%, respectively) than didESL (39.8%) and immersion (30.7%) students in Canada. The results suggest that contextualvariables such as type of classroom and sociolinguistic status of the target language may affecthow learners respond to corrective feedback.

These initial descriptive studies were important for understanding the occurrence ofinteraction in classroom contexts, and indeed there are still some recent examples of purelydescriptive studies. For example, Bowles et al. (2014) examined the types of interaction that

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 1 2 S H A W N L O E W E N A N D M A S A T O S H I S A T O

occurred when L2 learners of Spanish performed tasks with heritage learners of the language.They found that interaction was largely similar regardless of whether the interlocutor was anHL learner or not. In another recent descriptive study, Basterrechea & Garcıa Mayo (2013)looked at the effects of instructional context on the occurrence of LREs, comparing contentand language integrated learning (CLIL) and EFL settings in Spain. CLIL, a primarilyEuropean term, refers to ‘the learning of a non-language subject through a foreign languagewhere the subject and language have a joint role’ (p. 28); it is similar to immersion programmesin North America (Lyster 2007). Basterrechea & Garcıa Mayo found that LREs were presentin student interaction in both CLIL and EFL contexts, but there were more LREs in theCLIL context, suggesting that there were important differences between the two contexts(see also Llinares & Lyster 2014).

6.2 Modality

Although interactionist research has traditionally been interested in oral, as opposed towritten, interaction, the use of technology in communication has expanded the purviewof interactionist research. In particular, researchers have investigated the features andeffectiveness of synchronous CMC (SCMC), both oral and written (e.g. Smith 2003; Sauro2013b; Adams, Alwi & Newton 2015). Written SCMC occurs primarily as text chat, in whichtwo or more interlocutors are present in an online forum where they can exchange typedmessages. Several researchers have proposed advantages for this context (e.g. Sauro & Smith2010), and in fact text chat has sometimes been referred to as communication in slow motionbecause the modality allows for learners to take more time as they communicate. This slowerpace of interaction is proposed to benefit learners by giving them more time to draw on theirexplicit linguistic knowledge as they formulate their text. In addition, researchers argue thattext chat can help with noticing because there is a written record for learners to refer to, thusallowing them to scroll back through the conversation to check on linguistic forms (Beauvois1992; Smith & Sauro 2009).

However, some studies have found impoverished linguistic production in written SCMC,compared with FTF or oral SCMC. For example, Loewen & Wolff (2016) examinednegotiation for meaning, LREs, and recasts in task interaction in the three abovementionedmodalities. They found that the oral SCMC and FTF interaction had similar numbersof interactional features; however, the numbers were significantly reduced in the writtenSCMC condition. Similarly, Kim (2014) found that more communication strategies, such ascircumlocution and direct appeal for assistance, were present in FTF communication, whilemore avoidance strategies occurred in the written SCMC context. Rouhshad & Storch’s(2016) data also suggested that learners engaged in the task and with each other during FTFinteractions than during text chat. However, Sauro (2013) examined written SCMC as partof an authentic classroom partnership in which US college students interacted with Swedishhigh school students. The US students were given specific linguistic forms to target duringinteraction, and were told to provide corrective feedback. Sauro found that attention to formdid occur during the interaction; however, the provision of corrective feedback was not alwaysconsistent.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 313

In a recent study, Saito & Akiyama (2017) investigated the longitudinal development ofL2 English learners engaged in video-based negotiations for nine weeks with L1 Englishspeakers, who were trained to provide feedback in response to comprehension difficulties.Results indicated that L1 interlocutors generally provided feedback as trained. Furthermore,the interaction ‘seemed to have a significant impact on comprehensibility, fluency, vocabulary,and grammar but not necessarily accentedness and pronunciation’ (2017: 23). Additionalstudies of such long-term computer-based interaction are needed.

In an effort to understand how SCMC affects noticing and/or L2 development, severalstudies have used self-report or eye-tracking data (see an overview in Michel & Smith 2017).For instance, Gurzynski-Weiss & Baralt (2014) found no difference in learners’ accurateperception of feedback in either FTF (68%) or SCMC (71%). Learners were most accuratein their perception of lexical and semantic feedback regardless of modality. In addition,there were considerable differences in the number of times learners had the opportunity tomodify their output in the two modalities – 91% in FTF but only 53% in SCMC. In aneye-tracking study, Smith & Renaud (2013) measured the amount of time learners gazed atcorrective feedback during SCMC interaction. Learners fixated on corrective recasts 72%of the time, and there was a significant relationship between fixation and correct posttestscores for lexical and grammatical forms. Finally, Smith (2012) used both eye-tracking andstimulated recall to measure learners’ attention to recasts in written SCMC. He found astrong, positive correlation between the two measures, with learners noticing semantic andsyntactic features more than morphological ones.

Research has also compared levels of learner anxiety in FTF interaction and writtenSCMC. While Baralt & Gurzynski-Weiss (2011) found that there was no difference inlearners’ anxiety during either FTF or CMC interaction, Baralt, Gurzynski-Weiss & Kim(2016) reported mode differences related to learners’ affective aspects. The researchersanalysed the interactions in terms of cognitive, affective, and social engagement, andshowed that, first, learners’ level of attention to form (language awareness) was higher inthe FTF mode. Second, learners’ attitude towards the task was more positive in the FTFcontext. In addition, learners engaged with each other more in the FTF than in the CMCinteractions.

In a meta-analysis of 14 studies, Ziegler (2016a) compared the effects of modality, whetherSCMC or FTF, on interaction. Although there was a slight advantage for SCMC interms of raw scores of L2 development tests, the meta-analysis did not reveal statisticaldifferences between modalities, leading Ziegler to conclude that ‘mode of communicationhas no statistically significant impact on the positive developmental benefits associated withinteraction’ (p. 2). Further studies investigating the ever-changing modalities of interactioncan provide additional evidence in this regard.

Indeed, further research will surely continue to investigate these and other modalitiesof interaction, and as new modes of interaction appear, interactionist research will needto continue to extend its boundaries. For example, there are some studies investigatingvoice-based (Bueno-Alastuey 2013) and video-based CMC interaction (Yanguas 2010).Additionally, texting via cell phones is an increasingly popular means of communication;however, no studies, to our knowledge, have investigated this context in terms of features andeffectiveness of L2 interaction.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 1 4 S H A W N L O E W E N A N D M A S A T O S H I S A T O

7. Methodological issues

In order to ensure the validity and reliability of interactionist research, it is important toconsider the methodological choices that are made (Plonsky & Gass 2011). This sectionreviews some of the primary methodological issues that face interactionist researchers as theyinvestigate the impact of interaction on L2 development.

7.1 Laboratory and classroom research

L2 interaction can happen naturalistically outside of the classroom (e.g. Schegloff 2000), butit is interaction that occurs in pedagogical contexts and/or is manipulated in some way topromote L2 development that is of primary concern to ISLA researchers as they endeavourto determine the characteristics of input, interaction, and output that are most beneficialfor L2 acquisition (e.g. Mackey 2012; Loewen 2015; Loewen & Sato 2017). Much researchhas been conducted inside the classroom, which provides high ecological validity and insightinto the daily realities that face teachers and students as they interact. However, because it isdifficult to control moderating and confounding variables in the classroom (Shadish, Cook &Campbell 2002), researchers have also used laboratory studies to investigate the nature andeffectiveness of interaction. For example, Mackey & Goo (2007) found that 64% of the 28studies in their meta-analysis were laboratory-based, while only 36% were conducted in aclassroom.

The generalizability of laboratory research to the classroom has been questioned (e.g.Foster 1998; Eckerth 2009, although see Gass et al. 2005). Laboratory research mayinvolve learners in heightened awareness of language forms, even if they are engaged incommunicative interaction, because of the decontextualized nature and potential novelty ofthe tasks. Consequently, learners may engage in more noticing of linguistic forms, which mayresult in disproportionately higher levels of L2 development. Evidence suggesting a greaterimpact was found in Mackey & Goo’s (2007) meta-analysis of 28 studies in which the effectsof interaction on immediate posttests were greater in laboratory-based research. However,in contrast, Russell & Spada’s (2006) meta-analysis of 15 studies did not find differencesbetween the two settings. Subsequently, Mackey et al. (2013) commented that more authenticclassroom research is needed – particularly quasi-experimental research, testing the effectsof interaction on L2 development.

7.2 Descriptive and quasi-experimental research

There is a trend within interaction research for studies to become more interventionistover time. That is to say, early studies of interaction were primarily observational innature (e.g. Gass & Varonis 1986). Some studies were primarily concerned with describinginteraction, while others were correlational in nature, with researchers comparing theoccurrence of interactional features to noticing or the use of target structures. However,

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 315

as the characteristics of interaction, and the variables that influence them, become betterunderstood, the tendency has been for more quasi-experimental studies that manipulatespecific variables to determine their effects on L2 development. Still, descriptive studies havenot entirely disappeared as researchers continue to consider features affecting interactionthat have not yet been investigated. For example, Bowles et al. (2014) is one of the firstinteractional studies to investigate HL learners in interactional dyads. Other descriptivestudies have investigated the influence of technology on interaction (e.g. Loewen & Wolff2016).

In terms of investigating noticing, there have been several different methods used, butthey can be grouped into concurrent and retrospective measures. Concurrent measures arethose that attempt to measure noticing as it happens during interaction. One of the leastinvasive ways of doing this is to look at interactional discourse for evidence of noticing.This is what studies of uptake have done (e.g. Ellis et al. 2001a; Loewen 2005) whenthey have looked at how learners responded to corrective feedback that was providedduring meaning-focused interaction. These studies have argued on theoretical grounds thatlearners’ production may provide some evidence that linguistic forms have been noticed(Lightbown 1998). However, these studies have acknowledged that discourse features arenot equivalent to cognitive processes or to L2 learning. Consequently, it may be possible forlearners to repeat corrective feedback without noticing the form, for example; conversely,just because learners do not produce the correct form does not mean that noticing hasnot occurred. Therefore, other methods of concurrent measurement of noticing have beenemployed.

One such measure involves learners responding to a research-provided stimulus, such asa knocking sound, at critical junctures in the interaction. Several studies (e.g. Philp 2003;Bigelow et al. 2006) asked learners to repeat what they had just heard after a recast, whileother studies (e.g. Egi 2007) asked learners to report what they were thinking immediatelyupon hearing the stimulus. The former option is similar to the discourse measures mentionedin the previous paragraph, and the approach is predicated on the idea that learner productioncan provide evidence of noticing. However, that latter approach, in which learners are askedto state what they were thinking at that moment, provides additional information about thecognitive activities that learners are engaged in. Of course, here, noticing is measured at thelevel of awareness, with the assumption that learners will report, at minimum, something thatthey are aware of noticing. Or they may go one step further and provide evidence of noticingat the level of understanding.

Finally, in regard to concurrent measures of noticing, we should mention the role thattechnology is playing. For example, eye-tracking has recently been used to measure learners’noticing of linguistic features during written SCMC (Sauro & Smith 2010; Smith & Renaud2013) as well as during online FTF communication (McDonough et al. 2015) under theassumption of an eye-mind link in which the focus and duration of eye gaze is said tocorrespond with cognitive attention (see, for a review, McDonough 2017). Other technologythat may provide information about noticing during interaction is brain imaging; however, weare not aware of any studies that have examined this issue yet, probably because brain-imagingtechnique (e.g. event-related potentials (ERPs), functional magnetic resonance imaging(fMRI)) does not permit learners to engage in spontaneous interaction invoking different

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 1 6 S H A W N L O E W E N A N D M A S A T O S H I S A T O

cognitive functions (see Steinhauer 2014; Morgan-Short, Faretta-Stutenberg & Bartlett-Hsu2015).

Although they can provide some indication about learners’ noticing, concurrent measuresare somewhat problematic, in large part because they often interrupt the flow of conversation.As a result, retrospective measures have been more popular in measuring noticing duringinteraction. One popular method is stimulated recall (Gass & Mackey 2000) in which learnersare provided with audio or video recordings of their previous interaction, and they are askedwhat they were thinking at specific moments during the interaction. Studies of stimulatedrecall have found that different aspects of language, such as lexis and phonology, may bemore likely to be noticed during interaction than morphosyntax (Mackey et al. 2000).In addition to stimulated recall, other retrospective measures include learner journals orresponses to focused questions about the interaction. For example, Mackey (2006) usedlearning journals during class time, stimulated recall, and post-interaction written responsesto investigating learner noticing. She found that learners’ reports of noticing related tolinguistic development in some areas (e.g. questions) but not in others (e.g. past tense).Mackey also found that while there was some correspondence in the noticing identifiedby the different measures, there were also discrepancies, leading her to suggest thatnoticing might be best viewed as a continuum. In the case of all retrospective measures,the challenge is to maintain the veridicality of the data, ensuring that the learners’subsequent data reports their thoughts at the time of interaction rather than at the timeof reporting.

Some studies have triangulated data using both concurrent and retrospective measures ofnoticing. For example, Gurzynski-Weiss & Baralt (2015) examined the relationship betweenmodified output and noticing by looking at the type of modified output that occurred andthe reports of noticing that accompanied stimulated recalls based on corrective feedback.They found that partial modified output most accurately predicted correct noticing in bothFTF and SCMC contexts. Full modified output also predicted accurate noticing, but onlyin FTF. Gurzynski-Weiss & Baralt argue that full modified output is not conversationallyappropriate in SCMC contexts and often, if something is repeated in full, then it ismore likely to be copied and pasted, which abnegates any effects for deeper cognitiveprocessing.

8. Conclusion

The interaction approach, with its emphasis on input, interaction, and output, has beenwidely investigated within the broader ISLA framework, and theorists (Mackey et al. 2012;Pica 2013; Long 2015; Mackey & Gass 2015; Long 2017) have noted that there has been ashift from investigating IF interaction is helpful for L2 development to how and under whatcircumstances it is effective. Consequently, there is a need for ongoing research. First andforemost, Mackey et al. (2012) have pointed to the need for replication studies to supportthe findings of previous studies, particularly given the methodological shortcomings of manyprevious studies (Plonsky & Gass 2011). As Porte (2015) summarizes, replication studies canshow ‘how far we can separate knowledge from the particular circumstances of time, place,

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 317

procedure, or subjects which were part of the original experiment or study’ (p. 140). Giventhe complexity of SLA and dynamics of ISLA in the classroom setting, replication studiesof various aspects of interaction are warranted. A variable that has been found to influencethe effect of interaction on L2 development is linguistic target (e.g. Brown 2016). Whilethe benefits of interaction have been found for grammar and vocabulary, more recentlystudies investigating its effects on pronunciation have appeared (Derwing & Munro 2015). Inaddition, studies that explore the potentially beneficial effects of interaction on pragmaticsare needed (Bardovi-Harlig 2017).

Individual differences affecting interaction is another important area for expansion. Whileit may not be feasible to deliver instruction attuned to individual learners in the classroom,research focusing on aptitude-treatment interaction (DeKeyser 2017a) will further ourunderstanding of the effect of interaction on L2 development. Moreover, the interactionapproach thus far has focused primarily on cognitive factors, but scant attention has beenpaid to social and sociocognitive issues (Toth & Davin 2016). Much research has investigatedyoung adults in university contexts; more research on other populations, such as child learners(e.g. Garcıa Mayo & de los Angeles Hidalgo 2017), HL learners (e.g. Bowles et al. 2014),refugees with limited L1 literacy (e.g. Bigelow et al. 2006; Bigelow & King 2016), and other‘non-traditional’ students, would help extend the parameters of the interaction approach.Additionally, there are psychological differences (whether they be static or situational), forexample related to task motivation and WTC, that can impact learner interaction (Yashima,MacIntyre & Ikeda 2016). The extent to which learners are engaged with tasks (engagement)is a growing research area as well (Cao & Philp 2006; Philp & Duchesne 2016; Sato 2016).Finally, the role of gestures during interaction has begun to receive attention. For example,Nakatsukasa (2016) found that the combination of gestures and recasts on English locativeprepositions lead to sustained improvement on a delayed posttest compared to recasts withoutgestures.

Interactional contexts continue to develop as technology changes the ways in whichindividuals communicate, and interaction research needs to keep pace. Researchers haveconducted a fair amount of research on, especially, written SCMC; however, moreinvestigation into audio and audio-visual SCMC is warranted (e.g. Akiyama & Saito 2016).Even less attention has been paid to other types of technology-enhanced interaction, such astexting or game-based interaction, although these communication contexts are being usedfor L2 developmental purposes, and only time will tell how communication and interactioncontinues to evolve. Interactionist researchers will need to continue to explore the implicationsof such new modalities for L2 development.

In terms of research methods, there has been an important call for longitudinal researchand delayed testing to investigate the sustainability of the effects of interaction. Othertypes of research methods, such as those employed in neuroscience research, can provideadditional information about the cognitive effects of interaction (e.g. Morgan-Short et al.2012).

Interaction plays a crucial role in SLA theory and pedagogy, and there appears to beno slowdown in this regard. Consequently, interactionist research will need to continue toexplore how best to implement interaction so that L2 learners receive maximal benefit forL2 development and communication skills.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 1 8 S H A W N L O E W E N A N D M A S A T O S H I S A T O

Questions arising

• What are some ways to enhance the effects of different components of interaction(input, feedback, negotiation for meaning, negotiation for form, output)?

• How does learner psychology moderate the effect of interaction? How should thistopic best be investigated?

• What are some ways of transferring the findings from task-based research intopedagogical interventions?

• What are some methodological weaknesses of cognitive-interactionist research?• What challenges do new methods of communication bring to interactionist

research?

References

Adams, R. (2007). Do second language learners benefit from interacting with each other? In A. Mackey(ed.), Conversational interaction in second language acquisition: A collection of empirical studies. Oxford: OxfordUniversity Press, 29–51.

Adams, R., A. M. Nuevo & T. Egi (2011). Explicit and implicit feedback, modified output, and SLA:Does explicit and implicit feedback promote learning and learner–learner interactions? ModernLanguage Journal 95.SI, 42–63.

Adams, R., N. Alwi & J. Newton (2015). Task complexity effects on the complexity and accuracy ofwriting via text chat. Journal of Second Language Writing 29, 64–81.

Akiyama, Y. & K. Saito (2016). Development of comprehensibility and its linguistic correlates:A longitudinal study of video-mediated telecollaboration. Modern Language Journal 100.3,585–609.

Alcon Soler, E. (2002). Relationship between teacher-led versus learners’ interaction and thedevelopment of pragmatics in the EFL classroom. International Journal of Educational Research 37.3/4,359–77.

Ammar, A. & N. Spada (2006). One size fits all? Recasts, prompts, and L2 learning. Studies in SecondLanguage Acquisition 28.4, 543–574.

Ayoun, D. (2001). The role of negative and positive feedback in the second language acquisition of thepasse compose and the imparfait. Modern Language Journal 85.2, 226–243.

Baddeley, A. (2003). Working memory and language: An overview. Journal of Communication Disorders36.3, 189–208.

Baddeley, A. (2015). Working memory in second language learning. In Z. Wen, M. Mota & A. McNeil(eds.), Working memory in second language acquisition and processing. Bristol: Multilingual Matters, 17–28.

Baralt, M. (2013). The impact of cognitive complexity on feedback efficacy during online versusface-to-face interactive tasks. Studies in Second Language Acquisition 35.4, 689–725.

Baralt, M. & L. Gurzynski-Weiss (2011). Comparing learners’ state anxiety during task-basedinteraction in computer-meditation and face-to-face communication. Language Teaching Research 15.2,201–229.

Baralt, M., R. Gilabert & P. Robinson (2014). Task sequencing and instructed second language learning. London:Bloomsbury.

Baralt, M., L. Gurzynski-Weiss & Y. Kim (2016). The effects of task complexity and classroomenvironment on learners’ engagement with the language. In M. Sato & S. Ballinger (eds.), Peerinteraction and second language learning: Pedagogical potential and research agenda. Amsterdam: John Benjamins,209–239.

Bardovi-Harlig, K. (2017). Acquisition of L2 pragmatics. In S. Loewen & M. Sato (eds.), The Routledgehandbook of instructed second language acquisition. New York: Routledge, 224–245.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 319

Basterrechea, M. & P. Garcıa Mayo (2013). Language-related episodes during collaborative tasks: Acomparison of CLIL and EFL learners. In K. McDonough & A. Mackey (eds.), Second languageinteraction in diverse educational contexts. Philadelphia: John Benjamins, 25–43.

Basturkmen, H., S. Loewen & R. Ellis (2004). Teachers’ stated beliefs about incidental focus on formand their classroom practices. Applied Linguistics 25.2, 243–272.

Batstone, R. & J. Philp (2013). Classroom interaction and learning opportunities across time and space.In K. McDonough & A. Mackey (eds.), 109–125.

Beauvois, M. H. (1992). Computer-assisted classroom discussion in the foreign language classroom:Conversation in slow motion. Foreign Language Annals 25.5, 455–464.

Bigelow, M. & K. King (2016). Peer interaction while learning to read in a new language. In M. Sato& S. Ballinger (eds.), 349–376.

Bigelow, M., R. Delmas, K. Hansen & E. Tarone (2006). Literacy and the processing of oral recasts inSLA. TESOL Quarterly 40.4, 665–689.

Bowles, M., P. Toth & R. Adams (2014). A comparison of L2-L2 and L2-heritage learner interactionsin Spanish language classrooms. Modern Language Journal 98.2, 497–517.

Breen, M. (1987). Learner contributions to task design. In C. Candlin & D. Murphy (eds.), Languagelearning tasks. London: Prentice Hall, 23–46.

Brown, A. (2009). Students’ and teachers’ perceptions of effective foreign language teaching: Acomparison of ideals. Modern Language Journal 93.1, 46–60.

Brown, D. (2014). The type and linguistic foci of oral corrective feedback in the L2 classroom: Ameta-analysis. Language Teaching Research 20.4, 436–458.

Brown, D. (2016). The type and linguistic foci of oral corrective feedback in the L2 classroom: Ameta-analysis. Language Teaching Research 20, 436–458.

Bruton, A. & V. Samuda (1980). Learner and teacher roles in the treatment of oral error in group work.RELC Journal 11.2, 49–63.

Bueno-Alastuey, M. C. (2013). Interactional feedback in synchronous voice-based computer mediatedcommunication: Effect of dyad. System 41.5, 543–559.

Buckwalter, P. (2001). Repair sequences in Spanish L2 dyadic discourse: A descriptive study. ModernLanguage Journal 85.3, 380–397.

Bygate, M., P. Skehan & M. Swain (2001). Task-based learning: Language teaching, learning and assessment.London: Longman.

Cao, Y. (2014). A sociocognitive perspective on second language willingness to communicate. TESOLQuarterly 48.4, 789–814.

Cao, Y. & J. Philp (2006). Interactional context and willingness to communicate: A comparison ofbehavior in whole class, group and dyadic interaction. System 34.4, 480–493.

Carroll, J. B. (1981). Twenty-five years of research on foreign language aptitude. In K. Diller(ed.), Individual differences and universals in language learning aptitude. Rowley, MA: Newbury House,83–118.

Carroll, J. B. (1990). Cognitive abilities in foreign language aptitude: Then and now. In T. S. Parry &C. W. Stansfield (eds.), Language aptitude reconsidered. Washington, DC: Center for Applied Linguistics,11–29.

Carroll, J. B. & S. Sapon (2002). Modern language aptitude test (MLAT). Manual 2002 edition. N. Bethesda,MD: Second Language Testing, Inc.

Collins, L. & J. White (2014). The quantity and quality of language practice in typical interactivepair/group tasks. TESL Canada Journal 31.8, 47–67.

Collins, L., P. Trofimovich, J. White, W. Cardoso & M. Horst (2009). Some input on the easy/difficultgrammar question: An empirical study. Modern Language Journal 93.3, 336–353.

Csizer, K. (2017). Motivation in the L2 classroom. In S. Loewen & M. Sato (eds.), 418–432.de Bot, K. (1996). The psycholinguistics of the output hypothesis. Language Learning 46.3, 529–555.DeKeyser, R. M. (2001). Automaticity and automatization. In P. Robinson (ed.), Cognition and second

language instruction. New York: Cambridge University Press, 125–151.DeKeyser, R. M. (2012). Interactions between individual differences, treatments, and structures in

SLA. Language Learning 62.2, 189–200.DeKeyser, R. M. (2017a). Knowledge and skill in ISLA. In S. Loewen & M. Sato (eds.), 15–32.DeKeyser, R. M. (2017b). Age in learning and teaching grammar. In J. Liontas (ed.), The TESOL

encyclopedia of English language teaching. New York: Wiley. doi: 10.1002/9781118784235.eelt0106.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 2 0 S H A W N L O E W E N A N D M A S A T O S H I S A T O

DeKeyser, R. M. & K. J. Sokalski (1996). The differential role of comprehension and productionpractice. Language Learning 46.4, 613–642.

Derwing, T. M. & J. M. Munro (2015). Pronunciation fundamentals: Evidence-based perspectives for L2 teachingand research. Amsterdam: John Benjamins.

Dewaele, J.-M. (2017). Personality and anxiety. In S. Loewen & M. Sato (eds.), 433–450.Doughty, C. & T. Pica (1986). ‘Information gap’ tasks: An aid to second language acquisition? TESOL

Quarterly 20.2, 305–325.Dornyei, Z. (2002). The motivational basis of language learning tasks. In P. Robinson (ed.), Individual

differences and instructed language learning. Amsterdam: John Benjamins, 137–158.Dornyei, Z. & J. Kormos (2000). The role of individual and social variables in oral task performance.

Language Teaching Research 4.3, 275–300.Eckerth, J. (2008). Investigating consciousness-raising tasks: Pedagogically targeted and non-targeted

learning gains. International Journal of Applied Linguistics 18.2, 119–145.Eckerth, J. (2009). Negotiated interaction in the L2 classroom. Language Teaching 42.1, 109–130.Egi, T. (2007). Interpreting recasts as linguistic evidence: The role of linguistic target, length, and

degree of change. Studies in Second Language Acquisition 29.4, 511–537.Eisenchlas, S. A. (2011). On-line interactions as a resource to raise pragmatic awareness. Journal of

Pragmatics 43, 51–61.Ellis, R. (2003). Task-based language learning and teaching. Oxford: Oxford University Press.Ellis, R. (2017a). Oral corrective feedback in L2 classrooms: What we know so far. In H. Nassaji &

E. Kartchava (eds.), Corrective feedback in second language teaching and learning: Research, theory, applications,implications. New York: Routledge, 3–18.

Ellis, R. (2017b). Task-based language teaching. In S. Loewen & M. Sato (eds.), 108–125.Ellis, R. & X. He (1999). The roles of modified input and output in the incidental acquisition of word

meanings. Studies in Second Language Acquisition 21.2, 285–301.Ellis, R., H. Basturkmen & S. Loewen (2001a). Learner uptake in communicative ESL lessons. Language

Learning 51.2, 281–318.Ellis, R., H. Basturkmen & S. Loewen (2001b). Preemptive focus on form in the ESL classroom. TESOL

Quarterly 35.3, 407–432.Ellis, R., S. Loewen & R. Erlam (2006). Implicit and explicit corrective feedback and the acquisition

of L2 grammar. Studies in Second Language Acquisition 28.3, 339–368.Erlam, R. (2003). Evaluating the relative effectiveness of structured-input and output-based instruction

in foreign language learning: Results from an experimental study. Studies in Second Language Acquisition25.4, 559–582.

Ferguson, C. (1975). Toward a characterization of English foreigner talk. Anthropological Linguistics 17.1,1–14.

Fernandez Dobao, A. (2016). Peer interaction and learning: A focus on the silent learner. In M. Sato& S. Ballinger (eds.), Peer interaction and second language learning: Pedagogical potential and research agenda.Amsterdam: John Benjamins, 33–61.

Foster, P. (1998). A classroom perspective on the negotiation of meaning. Applied Linguistics 19.1, 1–23.Foster, P. & P. Skehan (1996). The influence of planning and task type on second language performance.

Studies in Second Language Acquisition 18.3, 293–323.Fortune, A. (2005). Learners’ use of metalanguage in collaborative form-focused L2 output tasks.

Language Awareness 14.1, 21–38.Fujii, A. & A. Mackey (2009). Interactional feedback in learner-learner interactions in a task-based

EFL classroom. International Review of Applied Linguistics in Language Teaching 47.3, 267–301.Fujii, A., N. Ziegler & A. Mackey (2016). Peer interaction and metacognitive instruction in the EFL

classroom. In M. Sato & S. Ballinger (eds.), 63–89.Garcıa Mayo, M. P. & T. Pica (2000). L2 learner interaction in a foreign language setting: Are learning

needs addressed? International Review of Applied Linguistics 38.1, 35–58.Garcıa Mayo, M. P. & A. Azkarai (2016). EFL task-based interaction: Does task modality impact on

language-related episodes? In M. Sato & S. Ballinger (eds.), 241–266.Garcıa Mayo, M. P. & M. de los Angeles Hidalgo (2017). L1 use among young EFL mainstream and

CLIL learners in task-supported interaction. System 67, 132–145.Gass, S. M. (1997). Input, interaction and the second language learner. Mahwah, NJ: Lawrence Erlbaum.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 321

Gass, S. M. (2003). Input and interaction. In C. J. Doughty & M. H. Long (eds.), The handbook of secondlanguage acquisition. Malden, MA: Blackwell Publishing, 224–255.

Gass, S. M. & M. J. Alvarez Torres (2005). Attention when? An investigation of the ordering effect ofinput and interaction. Studies in Second Language Acquisition 27.1, 1–31.

Gass, S. M. & E. Varonis (1985). Variation in native speaker speech modification to non-native speakers.Studies in Second Language Acquisition 7.1, 37–57.

Gass, S. M. & E. Varonis (1986). Sex differences in NNS/NNS interactions. In R. Day (ed.), Talking tolearn: Conversation in second language interaction. Rowley, MA: Newbury House, 327–351.

Gass, S. M. & E. Varonis (1990). Miscommunication in nonnative speaker discourse. In N. Coupland,H. Giles & J. Wiemann (eds.), ‘Miscommunication’ and problematic talk. London, New Delhi: SagePublications, 121–145.

Gass, S. M. & A. Mackey (2000). Stimulated recall methodology in second language research. Mahwah, NJ:Lawrence Erlbaum.

Gass, S. M. & A. Mackey (2015). Input, interaction, and output in second language acquisition. InB. VanPatten & J. Williams (eds.), Theories in second language acquisition (2nd edn.). New York: Routledge,180–206.

Gass, S. M., A. Mackey & L. Ross-Feldman (2005). Task-based interactions in classroom and laboratorysettings. Language Learning 55.4, 575–611.

Gass, S. M., P. Spinner & J. Behney (2018) (eds.). Salience in second language acquisition. New York:Routledge.

Gilabert, R. & J. Baron (2013). The impact of increasing task complexity on L2 pragmatic moves. InK. McDonough & A. Mackey (eds.), 45–69.

Gilabert, R., J. Baron & A. Llanes (2009). Manipulating cognitive complexity across task types andits impact on learners’ interaction during oral performance. International Review of Applied Linguistics47.3–4, 367–395.

Gilabert, R., R. Manchon & O. Vasylets (2016). Mode in theoretical and empirical TBLT research:Advancing research agendas. Annual Review of Applied Linguistics 36, 117–135.

Goo, J. (2012). Corrective feedback and working memory capacity in interaction-driven L2 learning.Studies in Second Language Acquisition 34.3, 445–474.

Goo, J. & A. Mackey (2013). The case against the case against recasts. Studies in Second Language Acquisition35.1, 1–39.

Gurzynski-Weiss, L. & A. Revesz (2012). Tasks, teacher feedback, and learner modified output innaturally occurring classroom interaction. Language Learning 62.3, 851–879.

Gurzynski-Weiss, L. & M. Baralt (2014). Exploring learner perception and use of task-basedinteractional feedback in FTF and CMC modes. Studies in Second Language Acquisition 36.1,1–37.

Gurzynski-Weiss, L. & M. Baralt (2015). Does type of modified output correspond to learner noticingof feedback? A closer look in face-to-face and computer-mediated task-based interaction. AppliedPsycholinguistics 36.6, 1393–1420.

Gurzynski-Weiss, L., A. Y. Long & M. Solon (2017). TBLT and L2 pronunciation: Do thebenefits of tasks extend beyond grammar and lexis? Studies in Second Language Acquisition 39.3,213–224.

Hatch, E. (1978). Discourse analysis and second language acquisition. In E. Hatch (ed.), Second languageacquisition: A book of readings. Rowley, MA: Newbury House, 401–435.

Horwitz, E. K., M. B. Horwitz & J. Cope (1986). Foreign language classroom anxiety. Modern LanguageJournal 70.2, 125–132.

Iwashita, N. (2001). The effect of learner proficiency on interactional moves and modified output innonnative-nonnative interaction in Japanese as a foreign language. System 29.2, 267–287.

Jenks, C, J. (2009). Exchanging missing information in tasks: old and new interpretations. ModernLanguage Journal 93.2, 185–194.

Jeon, K. S. (2007). Interaction-driven L2 learning: Characterizing linguistic development. In A. Mackey(ed.), 379–403.

Jordan, G. (2005). Theory construction in second language acquisition. Amsterdam: John Benjamins.Juffs, A. & M. Harrington (2012). Aspects of working memory in L2 learning. Language Teaching 44.2,

137–166.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 2 2 S H A W N L O E W E N A N D M A S A T O S H I S A T O

Keck, C., G. Iberri-Shea, N. Tracy-Ventura & S. Wa-Mbaleka (2006). Investigating the empirical linkbetween task-based interaction and acquisition: A meta-analysis. In J. Norris & L. Ortega (eds.),Synthesizing research on language learning and teaching. Philadelphia: John Benjamins, 91–132.

Kennedy, S. & P. Trofimovich (2017). Pronunciation acquisition. In S. Loewen & M. Sato (eds.),260–279.

Kim, H.-Y. (2014). Learning opportunities in synchronous computer-mediated communication andface-to-face interaction. Computer Assisted Language Learning 27.1, 26–43.

Kim, Y. (2008). The contribution of collaborative and individual tasks to the acquisition of L2vocabulary. Modern Language Journal 92.1, 114–130.

Kim, Y. (2012). Task complexity, learning opportunities and Korean EFL learner question development.Studies in Second Language Acquisition 34.4, 627–658.

Kim, Y. (2017). Cognitive-interactionist approaches to L2 instruction. In S. Loewen & M. Sato (eds.),126–135.

Kim, Y. & K. McDonough (2008). The effect of interlocutor proficiency on the collaborative dialoguebetween Korean as a second language learners. Language Teaching Research 12.2, 211–234.

Kim, Y., C. Payant & P. Pearson (2015). The intersection of task-based interaction, task complexity,and working memory. Studies in Second Language Acquisition 37.3, 549–581.

Kormos, J. (2006). Speech production and second language acquisition. Mahwah, NJ: Lawrence Erlbaum.Kormos, J. & Y. Prefontaine (2016). Affective factors influencing fluent performance: French learners’

appraisals of second language speech tasks. Language Teaching Research. doi: 1362168816683562.Kormos, J. & A. Trebits (2012). The role of task complexity, modality, and aptitude in narrative task

performance. Language Learning 62.2, 439–472.Krashen, S. (1982). Principles and practice in second language acquisition. Oxford: Pergamon.Krashen, S. (2003). Explorations in language acquisition and use. Portsmouth, NH: Heinemann.Lasagabaster, D. & J. M. Sierra (2005). Error correction: Students’ versus teachers’ perceptions. Language

Awareness 14, 112–127.Lee, A. H. & R. Lyster (2015). The effects of corrective feedback on instructed L2 speech perception.

Studies in Second Language Acquisition 38.1, 35–64.Leeman, J. (2003). Recasts and second language development: Beyond negative evidence. Studies in

Second Language Acquisition 25.1, 37–63.Leeser, M. (2004). Learner proficiency and focus on form during collaborative dialogue. Language

Teaching Research 8.1, 55–81.Levelt, W. (1983). Monitoring and self-repair in speech. Cognition 14.1, 41–104.Li, S. (2010). The effectiveness of corrective feedback in SLA: A meta-analysis. Language Learning 60.2,

309–365.Li, S. (2013). The interactions between the effects of implicit and explicit feedback and individual

differences in language analytic ability and working memory. Modern Language Journal 97.3, 634–654.Li, S. (2014). The interface between feedback type, L2 proficiency, and the nature of the linguistic

target. Language Teaching Research 18.3, 373–396.Li, S. (2017). Cognitive differences and ISLA. In S. Loewen & M. Sato (eds.), 396–417.Li, S., Y. Zhu & R. Ellis (2016). The effects of the timing of corrective feedback on the acquisition of a

new linguistic structures. Modern Language Journal 100.1, 276–295.Lightbown, P. M. (1998). The importance of timing in focus on form. In C. Doughty & J. Williams

(eds.), Focus on form in classroom second language acquisition. Cambridge: Cambridge University Press,177–196.

Lightbown, P. M. (2013). Learner readiness. The encyclopedia of applied linguistics. Hoboken, NJ: Wiley-Blackwell.

Llinares, A. & R. Lyster (2014). The influence of context on patterns of corrective feedback andlearner uptake: A comparison of CLIL and immersion classrooms. The Language Learning Journal42.2, 181–194.

Loewen, S. (2003). Variation in frequency and characteristics of incidental focus on form. LanguageTeaching Research 7.3, 315–345.

Loewen, S. (2004). Uptake in incidental focus on form in meaning-focused ESL lessons. LanguageLearning 54.1, 153–187.

Loewen, S. (2005). Incidental focus on form and second language learning. Studies in Second LanguageAcquisition 27.3, 361–386.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 323

Loewen, S. (2012). The role of feedback. In S. M. Gass & A. Mackey (eds.), The Routledge handbook ofsecond language acquisition. New York: Routledge, 24–40.

Loewen, S. (2015). Introduction to instructed second language acquisition. New York: Routledge.Loewen, S. & T. Nabei (2007). Measuring the effects of oral corrective feedback on L2 knowledge. In

A. Mackey (ed.), 361–377.Loewen, S. & J. Philp (2006). Recasts in the adult English L2 classroom: Characteristics, explicitness,

and effectiveness. Modern Language Journal 90.4, 536–556.Loewen, S. & D. Wolff (2016). Peer interaction in F2F and CMC contexts. In M. Sato & S. Ballinger

(eds.), 163–184.Loewen, S. & D. R. Isbell (2017). Pronunciation in face-to-face and audio-only

synchronous computer-mediated learner interaction. Studies in Second Language Acquisition 39.2,225–256.

Loewen, S. & M. Sato (2017). Instructed second language acquisition (ISLA): An overview. In S.Loewen & M. Sato (eds.), 1–12.

Loewen, S., S. Li, F. Fei, A. Thompson, K. Nakatsukasa, S. Ahn & X. Chen (2009). Second languagelearners’ beliefs about grammar instruction and error correction. Modern Language Journal 93.1,91–104.

Long, M. H. (1981). Input, interaction, and second language acquisition. Annals of the New York Academyof Sciences 379, 259–278.

Long, M. H. (1983). Native speaker/non-native speaker conversation and the negotiation ofcomprehensible input. Applied Linguistics 4.2, 126–141.

Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W.C. Ritchie & T. K. Bhatia (eds.), Handbook of research on language acquisition (vol. 2). New York: AcademicPress, 413–468.

Long, M. H. (2007). Problems in SLA. New York: Routledge.Long, M. H. (2015). Second language acquisition and task-based language teaching. New York: Wiley.Long, M. H. (2016). In defense of tasks and TBLT: Nonissues and real issues. Annual Review of Applied

Linguistics 36, 5–33.Long, M. H. (2017). Instructed second language acquisition (ISLA): Geopolitics, methodological

issues, and some major research questions. Instructed Second Language Acquisition 1,7–44.

Long, M. H. & P. A. Porter (1985). Group work, interlanguage talk, and second language acquisition.TESOL Quarterly 19.2, 207–228.

Long, M. H., S. Inagaki & L. Ortega (1998). The role of implicit negative feedback in SLA: Modelsand recasts in Japanese and Spanish. Modern Language Journal 82.3, 357–371.

Loschky, L. (1994). Comprehensible input and second language acquisition: What is the relationship?Studies in Second Language Acquisition 16.3, 303–323.

Lyster, R. (1998a). The ambiguity of recasts and repetition in L2 classroom discourse. Studies in SecondLanguage Acquisition 20.1, 51–81.

Lyster, R. (1998b). Negotiation of form, recasts, and explicit correction in relation to error types andlearner repair in immersion classrooms. Language Learning 48.2, 183–218.

Lyster, R. (2004). Differential effects of prompts and recasts in form-focused instruction. Studies in SecondLanguage Acquisition 26.3, 399–432.

Lyster, R. (2007). Learning and teaching languages through content: A counterbalanced approach. Amsterdam: JohnBenjamins.

Lyster, R. & L. Ranta (1997). Corrective feedback and learner uptake: Negotiation of form incommunicative classrooms. Studies in Second Language Acquisition 19.1, 37–66.

Lyster, R. & K. Saito (2010). Oral feedback in classroom SLA: A meta-analysis. Studies in Second LanguageAcquisition 32.2, 265–302.

Lyster, R. & L. Ranta (2013). Counterpoint piece: The case for variety in corrective feedback research.Studies in Second Language Acquisition 35.1, 1–18.

Lyster, R. & M. Sato (2013). Skill acquisition theory and the role of practice in L2 development. InM. Garcıa Mayo, J. Gutierrez-Mangado & M. Martınez Adrian (eds.), Contemporary approaches to secondlanguage acquisition. Amsterdam: John Benjamins, 71–92.

Lyster, R., K. Saito & M. Sato (2013). Oral corrective feedback in second language classrooms. LanguageTeaching 46.1, 1–40.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 2 4 S H A W N L O E W E N A N D M A S A T O S H I S A T O

MacIntyre, P. D., C. Burns & A. Jessome (2011). Ambivalence about communicating in a secondlanguage: A qualitative study of French immersion students’ willingness to communicate. ModernLanguage Journal 95.1, 81–96.

MacIntyre, P. D., T. Gregersen & S. Mercer (2016) (eds.). Positive psychology in SLA. Clevedon: MultilingualMatters.

MacIntyre, P. D., Z. Dornyei, R. Clement & K. A. Noels (1998). Conceptualizing willingness tocommunicate in a L2: A situational model of L2 confidence and affiliation. Modern Language Journal82.4, 545–562.

Mackey, A. (1999). Input, interaction, and second language development: An empirical study ofquestion formation in ESL. Studies in Second Language Acquisition 21.4, 557–587.

Mackey, A. (2006). Feedback, noticing and instructed second language learning. Applied Linguistics 27.3,405–430.

Mackey, A. (2012). Input, interaction, and corrective feedback in L2 learning. Oxford: Oxford University Press.Mackey, A. & J. Goo (2007). Interaction research in SLA: A meta-analysis and research synthesis. In

A. Mackey (ed.), 407–449.Mackey, A. & R. Sachs (2012). Older learners in SLA research: A first look at working memory,

feedback, and L2 development. Language Learning 62.3, 704–740.Mackey, A. & S. M. Gass (2015). Second language research: Methodology and design. New York: Routledge.Mackey, A., R. Abbuhl & S. M. Gass (2012). Interactionist approach. In S. M. Gass & A. Mackey (eds.)

The Routledge handbook of second language acquisition. New York: Routledge, 7–24.Mackey, A., S. M. Gass & K. McDonough (2000). How do learners perceive interactional feedback?

Studies in Second Language Acquisition 22.4, 471–497.Mackey, A., R. Oliver & J. Leeman (2003). Interactional input and the incorporation of feedback: An

exploration of NS-NNS and NNS-NNS adult and child dyads. Language Learning 53.1, 35–66.Mackey, A., A. P. Kanganas & R. Oliver (2007). Task familiarity and interactional feedback in child

ESL classrooms. TESOL Quarterly 41.2, 285–312.Mackey, A., J. Philp, T. Egi, A. Fujii & T. Tatsumi (2002). Individual differences in working memory,

noticing of interactional feedback and L2 development. In P. Robinson (ed.), 181–209.Mackey, A., A. Fujii, S. Biesenbach-Lucas, H. Weger, N. Jacobsen, L. Fogle, J. Lake, K. Sondermann,

K. Tagarelli, M. Tsujita, A. Watanabe, R. Abbuhl & K. Kim (2013). Tasks and traditional practiceactivities in a foreign language context. In K. McDonough & A. Mackey (eds.), 71–87.

McDonough, K. (2004). Learner-learner interaction during pair and small group activities in a ThaiEFL context. System 32.2, 207–224.

McDonough, K. (2017). Experimental research methods. In S. Loewen & M. Sato (eds.), 562–576.McDonough, K. & A. Mackey (2008). Syntactic priming and ESL question development. Studies in

Second Language Acquisition 30.1, 31–47.McDonough, K. & W. Chaikitmongkol (2010). Collaborative syntactic priming activities and EFL

learners’ production of wh- questions. The Canadian Modern Language Review 66.6, 817–841.McDonough, K. & T. Hernandez Gonzalez (2013). Language production opportunities during whole-

group interaction in conversation group settings. In K. McDonough & A. Mackey (eds.), 293–314.McDonough, K., W. J. Crawford & A. Mackey (2015). Creativity and EFL students’ language use

during a group problem-solving task. TESOL Quarterly 49.1, 188–199.McDonough, K., P. Trofimovich & H. Neumann (2015). Eliciting production of L2 target structures

through priming activities. The Canadian Modern Language Review 71.1, 75–95.McDonough, K., D. Crowther, P. Kielstra & P. Trofimovich (2015). Exploring the potential role of eye

gaze in eliciting English L2 speakers’ responses to recasts. Second Language Research 31.4, 563–575.Michel, M. C. & B. Smith (2017). Eye-tracking research in computer-mediated language learning. In

S. Thorne & S. May (eds.), Language, education, and technology. New York: Springer, 453–464.Morgan-Short, K., M. Faretta-Stutenberg & L. Bartlett-Hsu (2015). Contributions of event-related

potential research to issues in explicit and implicit second language acquisition. In P. Rebuschat(ed.), Explicit and implicit learning of languages. Amsterdam: John Benjamins, 349–384.

Morgan-Short, K., K. Steinhauer, C. Sanz & M. T. Ullman (2012). Explicit and implicit secondlanguage training differentially affect the achievement of native-like brain activation patterns. Journalof Cognitive Neuroscience 24.4, 933–947.

Nakatsukasa, K. (2016). Efficacy of recasts and gestures on the acquisition of locative prepositions.Studies in Second Language Acquisition 38.4, 771–799.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 325

Nassaji, H. (2013). Participation structure and incidental focus on form in adult ESL classrooms.Language Learning 63.4, 835–869.

Naughton, D. (2006). Cooperative strategy training and oral interaction: Enhancing small groupcommunication in the language classroom. Modern Language Journal 90.2, 169–184.

Newton, J. (2013). Incidental vocabulary learning in classroom communication tasks. Language TeachingResearch 17.2, 164–187.

Oliver, R. (1998). Negotiation of meaning in child interactions. The relationship between conversationalinteraction and second language acquisition. Modern Language Journal 82.3, 372–386.

Oliver, R. (2000). Age difference in negotiation and feedback in classroom and pair work. LanguageLearning 50.1, 119–151.

Oliver, R. (2002). The patterns of negotiation for meaning in child interactions. Modern Language Journal86.1, 97–111.

Oliver, R. & A. Mackey (2003). Interactional context and feedback in child ESL classrooms. ModernLanguage Journal 87.4, 519–533.

Oliver, R., B. Nguyen & M. Sato (2017). Child instructed SLA. In S. Loewen & M. Sato (eds.), 468–487.

Oliver, R., J. Philp & S. Duchesne (2017). Children working it out together: A comparison of youngerand older learners collaborating in task based interaction. System 69, 1–14.

Panova, I. & R. Lyster (2002). Patterns of corrective feedback and uptake in an adult ESL classroom.TESOL Quarterly 36.4, 573–595.

Parlak, O. & N. Ziegler (2017). The impact of recasts on the development of primary stress in asynchronous computer-mediated environment. Studies in Second Language Acquisition 39.2, 257–285.

Perez-Vidal, C. (2017). Study abroad and ISLA. In S. Loewen & M. Sato (eds.), 339–360.Philp, J. (2003). Constraints on ‘noticing the gap’: Nonnative speakers’ noticing of recasts in NS-NNS

interaction. Studies in Second Language Acquisition 25.1, 99–126.Philp, J. & S. Duchesne (2016). Exploring engagement in tasks in the language classroom. Annual Review

of Applied Linguistics 36, 50–72.Philp, J. & N. Iwashita (2013). Talking, tuning in and noticing: Exploring the benefits of output in

task-based peer interaction. Language Awareness 22.4, 353–370.Philp, J., R. Adams & N. Iwashita (2014). Peer interaction and second language learning. New York: Routledge.Pica, T. (1988). Interlanguage adjustments as an outcome of NS-NNS negotiated interaction. Language

Learning 38.1, 45–73.Pica, T. (1994). Research on negotiation: What does it reveal about second language learning conditions,

processes, and outcomes? Language Learning 44.3, 493–527.Pica, T. (2005). Classroom learning, teaching, and research: A task-based perspective. Modern Language

Journal 89.3, 339–352.Pica, T. (2013). From input, output and comprehension to negotiation, evidence, and attention: An

overview of theory and research on learner interaction and SLA. In M. Garcıa Mayo, J. Gutierrez-Mangado & M. M. Adrian (eds.), Contemporary approaches to second language acquisition. Amsterdam: JohnBenjamins, 49–69.

Pica, T., R. Young & C. Doughty (1987). The impact of interaction in comprehension. TESOL Quarterly21.4, 737–758.

Pica, T., H. Kang & S. Sauro (2006). Information gap tasks, their multiple roles and contributions tointeraction research methodology. Studies in Second Language Acquisition 28.2, 301–338.

Pica, T., F. Lincoln-Porter, D. Paninos & J. Linnell (1996). Language learners’ interaction: Howdoes it address the input, output, and feedback needs of L2 learners? TESOL Quarterly 30.1,59–84.

Pica, T., L. Holliday, N. Lewis, D. Berducci & J. Newman (1991). Language learning through interaction.Studies in Second Language Acquisition 13.2, 343–376.

Plonsky, L. & S. M. Gass (2011). Quantitative research methods, study quality and outcomes: The caseof interaction research. Language Learning 61.2, 325–366.

Plonsky, L. & D. Brown (2015). Domain definition and search techniques in meta-analyses of L2research (or why 18 meta-analyses of feedback have different results). Second Language Research 31.2,267–278.

Plonsky, L. & Y. Kim (2016). Task-based learner production: A substantive and methodological review.Annual Review of Applied Linguistics 36, 73–97.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 2 6 S H A W N L O E W E N A N D M A S A T O S H I S A T O

Porte, G. (2015). Replication research in quantitative research. In J. D. Brown & C. Coombe (eds.),The Cambridge guide to research in language teaching and learning. Cambridge: Cambridge University Press,140–145.

Porter, P. (1986). How learners talk to each other: Input and interaction in task-centered discussions.In R. Day (ed.), Talking to learn: Conversation in second language acquisition. Rowley, MA: Newbury House,200–222.

Poupore, G. (2016). Measuring group work dynamics and its relation with L2 learners’ task motivationand language production. Language Teaching Research 20.6, 719–740.

Ranta, L. (2002). The role of learners’ language analytic ability in the communicative classroom. In P.Robinson (ed.), 159–179.

Rassaei, E. (2015). The effects of foreign language anxiety on EFL learners’ perceptions of oralcorrective feedback. Innovation in Language Learning and Teaching 9.2, 87–101.

Revesz, A. (2009). Task complexity, focus on form, and second language development. Studies in SecondLanguage Acquisition 31.3, 437–470.

Revesz, A. (2011). Task complexity, focus on L2 constructions, and individual differences: A classroom-based study. Modern Language Journal 95.2, 162–181.

Revesz, A. (2012). Working memory and the observed effectiveness of recasts on different outcomemeasures. Language Learning 62.1, 93–132.

Revesz, A., R. Sachs & M. Hama (2014). The effects of task complexity and input frequency on theacquisition of the past counterfactual construction through recasts. Language Learning 64.3, 615–650.

Robinson, P. (1995). Attention, memory and the ‘noticing’ hypothesis. Language Learning 45.2, 99–140.Robinson, P. (1996). Consciousness, rules and instructed second language acquisition. New York: Lang.Robinson, P. (2001). Task complexity, task difficulty, and task production: Exploring interactions in a

componential framework. Applied Linguistics 22.1, 27–57.Robinson, P. (2003). Attention and memory during SLA. In C. J. Doughty & M. H. Long (eds.), 631

678.Robinson, P. (2005). Aptitude and second language acquisition. Annual Review of Applied Linguistics 25,

46–73.Robinson, P. (2007). Task complexity, theory of mind, and intentional reasoning: Effects on L2

speech production, interaction, uptake and perceptions of task difficulty. International Review of AppliedLinguistics 45.3, 193–213.

Robinson, P. (2011). Second language task complexity: Researching the Cognition Hypothesis of language learning andperformance. Amsterdam: John Benjamins.

Rouhshad, A. & N. Storch (2016). A focus on mode: Patterns of interaction in face-to-face andcomputer-mediated contexts. In M. Sato & S. Ballinger (eds.), 267–289.

Russell, J. & N. Spada (2006). The effectiveness of corrective feedback for the acquisition of L2 grammar:A meta-analysis of the research. In J. Norris & L. Ortega (eds.), Synthesizing research on language learningand teaching. Philadelphia: John Benjamins, 133–164.

Sagarra, N. (2007). From CALL to face-to-face interaction: The effect of computer-delivered recastsand working memory on L2 development. In A. Mackey (ed.), 229–248.

Sagarra, N. & R. Abbuhl (2013). Optimizing the noticing of recasts via computer-delivered feedback:Evidence that oral input enhancement and working memory help second language learning. ModernLanguage Journal 97.1, 196–216.

Saito, K. (2013). The acquisition value of recasts in instructed language speech learning: Teachingthe perception and production of English /ɹ/ to adult Japanese learners. Language Learning 63.3,499–529.

Saito, K. (2015). Variables affecting the effects of recasts on L2 pronunciation development. LanguageTeaching Research 19.3, 276–300.

Saito, K. & R. Lyster (2012). Effects of form-focused instruction and corrective feedback on L2pronunciation development of /ɹ/ by Japanese learners of English. Language Learning 62.2, 595–633.

Saito, K. & X. Wu (2014). Communicative focus on form and second language suprasegmental learning:Teaching Cantonese learners to perceive Mandarin tones. Studies in Second Language Acquisition 36.4,647–680.

Saito, K. & Y. Akiyama (2017). Video-based interaction, negotiation for comprehensibility, and secondlanguage speech learning: A longitudinal study. Language Learning 67.1, 43–74.

Samuda, V. & M. Bygate (2008). Tasks in second language learning. Basingstoke: Palgrave MacMillan.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 327

Sato, M. (2013). Beliefs about peer interaction and peer corrective feedback: Efficacy of classroomintervention. Modern Language Journal 97.3, 611–633.

Sato, M. (2015). Density and complexity of oral production in interaction: The interactionist approachand an alternative. International Review of Applied Linguistics in Language Teaching 53.3, 307–329.

Sato, M. (2016). Interaction mindsets, interactional behaviors, and L2 development: An affective-social-cognitive model. Language Learning 67.2. doi.org/10.1111/lang.12214.

Sato, M. (2017). Oral peer corrective feedback: Multiple theoretical perspectives. In H. Nassaji & E.Kartchava (eds.), 19–34.

Sato, M. & S. Ballinger (2012). Raising language awareness in peer interaction: A cross-context,cross-method examination. Language Awareness 21, 157–179.

Sato, M. & S. Ballinger (2016). Understanding peer interaction: An overview of the research. In M.Sato & S. Ballinger (eds.), 1–30.

Sato, M. & R. Lyster (2007). Modified output of Japanese EFL learners: Variable effects of interlocutorvs. feedback types. In A. Mackey (ed.), 123–142.

Sato, M. & R. Lyster (2012). Peer interaction and corrective feedback for accuracy and fluencydevelopment: Monitoring, practice, and proceduralization. Studies in Second Language Acquisition 34.4,591–262.

Sato, M. & P. Viveros (2016). Interaction or collaboration? Group dynamics in the foreign languageclassroom. In M. Sato & S. Ballinger (eds.), 91–112.

Sauro, S. (2013a). SCMC for SLA. A research synthesis. CALICO Journal 28.2, 1–23.Sauro, S. (2013b). The cyber language exchange. In K. McDonough & A. Mackey (eds.), 129–145.Sauro, S. & B. Smith (2010). Investigating L2 performance in text chat. Applied Linguistics 31.4,

554–577.Schachter, J. (1991). Corrective feedback in historical perspective. Second Language Research 7.2, 89–102.Schegloff, E. (2000). When ‘others’ initiate repair. Applied Linguistics 21.2, 205–243.Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics 11,

129–158.Schmidt, R. (1995). Attention and awareness in foreign language learning. Honolulu: University of Hawai’i

Press.Schmidt, R. (2001). Attention. In P. Robinson (ed.), Cognition and second language instruction. Cambridge:

Cambridge University Press, 3–32.Schulz, R. (1996). Focus on form in the foreign language classroom: Students’ and teachers’ views on

error correction and the role of grammar. Foreign Language Annals 29.3, 343–364.Schulz, R. (2001). Cultural differences in student and teacher perceptions concerning the role

of grammar instruction and corrective feedback: USA: Colombia. Modern Language Journal 85,244–258.

Shadish, W. R., T. D. Cook & D. T. Campbell (2002). Experimental and quasi-experimental designs forgeneralized causal inference. Boston, MA: Houghton Mifflin Company.

Sheen, Y. (2004). Corrective feedback and learner uptake in communicative classrooms acrossinstructional settings. Language Teaching Research 8.3, 263–300.

Sheen, Y. (2008). Recasts, language anxiety, modified output, and L2 learning. Language Learning 58.4,835–874.

Sheen, Y. & R. Ellis (2011). Corrective feedback in language teaching. In E. Hinkel (ed.), Handbook ofresearch in second language teaching and learning, vol. 2. New York: Routledge, 593–610.

Shehadeh, A. (1999). Non-native speakers’ production of modified comprehensible output and secondlanguage learning. Language Learning 49.4, 627–675.

Shehadeh, A. (2001). Self- and other-initiated modified output during task-based interaction. TESOLQuarterly 35.3, 433–457.

Shintani, N. (2014). The incidental grammar acquisition in focus on form and focus on forms instructionfor young beginner learners. TESOL Quarterly 49.1, 115–140.

Shintani, N. (2015). The effectiveness of processing instruction and production-based instruction onL2 grammar acquisition: A meta-analysis. Applied Linguistics 36.3, 306–325.

Simard, D. & G. Jean (2011). An exploration of L2 teachers’ use of pedagogical interventions devisedto draw L2 learners’ attention to form. Language Learning 61.3, 759–785.

Skehan, P. (1998). A cognitive approach to language learning. Oxford: Oxford University Press.Skehan, P. (2014). Processing perspectives on task performance. Amsterdam: John Benjamins.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

3 2 8 S H A W N L O E W E N A N D M A S A T O S H I S A T O

Smith, B. (2003). Computer-mediated negotiated interaction: An expanded model. Modern LanguageJournal 87.1, 38–57.

Smith, B. (2009). The relationship between scrolling, negotiation, and self-initiated self-repair in anSCMC environment. CALICO Journal 26.2, 231–245.

Smith, B. (2012). Eye tracking as a measure of noticing: A study of explicit recasts in SCMC. LanguageLearning & Technology 16.3, 53–81.

Smith, B. & C. Renaud (2013). Using eye tracking as a measure of foreign language learners’ noticingof recasts during computer-mediated writing conferences. In K. McDonough & A. Mackey (eds.),147–165.

Smith, B. & S. Sauro (2009). Interruptions in chat. Computer Assisted Language Learning 22.3, 229–247.Spada, N. & P. Lightbown (2008). Form-focused instruction: Isolated or integrated? TESOL Quarterly

42.2, 181–207.Spada, N. & P. Lightbown (2009). Interaction research in second/foreign language classrooms.

In A. Mackey & C. Polio (eds.), Multiple perspectives on interaction. New York: Routledge,157–175.

Spada, N. & Y. Tomita (2010). Interactions between type of instruction and type of language feature:A meta-analysis. Language Learning 60.2, 263–308.

Steinhauer, K. (2014). Event-related potentials (ERPs) in second language research: A brief introductionto the technique, a selected review, and an invitation to reconsider critical periods in L2. AppliedLinguistics 35, 393–417.

Storch, N. (2002). Patterns of interaction in ESL pair work. Language Learning 52.1, 119–158.Storch, N. & A. Aldosari (2013). Pairing learners in pair work activity. Language Teaching Research 17.1,

31–48.Swain, M. (1985). Communicative competence: Some roles of comprehensible input and

comprehensible output in its development. In S. Gass & C. Madden (eds.), Input in second languageacquisition. Rowley, MA: Newbury, 235–253.

Swain, M. (1995). Three functions of output in second language learning. In G. Cook & B. Seidlhofer(eds.), Principle and practice in applied linguistics. Oxford: Oxford University Press, 125–144.

Swain, M. (2005). The output hypothesis: Theory and research. In E. Hinkel (ed.), 471–483.Swain, M. & S. Lapkin (1998). Interaction and second language learning: two adolescent French

immersion students working together. Modern Language Journal 82.3, 320–337.Swain, M. & S. Lapkin (2001). Focus on form through collaborative dialogue: Exploring task effects.

In M. Bygate, P. Skehan & M. Swain (eds.), Researching pedagogic tasks: Second language learning, teachingand testing. Harlow: Longman, 99–118.

Sykes, J. (2005). Synchronous CMC and pragmatic development: Effects of oral and written chat.CALICO Journal 22, 399–431.

Tavakoli, P. & P. Foster (2008). Task design and second language performance: The effect of narrativetype on learner output. Language Learning 58.2, 439–473.

Tokowicz, N. & B. MacWhinney (2005). Implicit and explicit measures of sensitivity to violationsin second language grammar: An event-related potential investigation. Studies in Second LanguageAcquisition 27.1, 173–204.

Tomita, Y. & N. Spada (2013). Form-focused instruction and learner investment in L2 communication.Modern Language Journal 97.3, 591–610.

Toth, P. D. (2008). Teacher- and learner-led discourse in task-based grammar instruction:Providing procedural assistance for L2 morphosyntactic development. Language Learning 58.2,237–283.

Toth, P. D. & K. J. Davin (2016). The sociocognitive imperative of L2 pedagogy. Modern Language Journal100.S1, 148–168.

Trofimovich, P. & K. McDonough (2011). Using priming methods to study L2 learning and teaching.In P. Trofimovich & K. McDonough (eds.), Applying priming methods to L2 learning, teaching and research.Amsterdam: John Benjamins, 3–17.

Trofimovich, P., A. Ammar & E. Gatbonton (2007). How effective are recasts: The role of attention,memory, and analytical ability. In A. Mackey (ed.), 171–195.

VanPatten, B. (1990). Attending to form and content in the input. Studies in Second Language Acquisition12.3, 287–301.

VanPatten, B. (2017a). Processing instruction. In S. Loewen & M. Sato (eds.), 166–180.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

I N T E R A C T I O N A N D I N S T R U C T E D S L A 329

VanPatten, B. (2017b). Situation instructed language acquisition: Facts about second languageacquisition. Instructed Second Language Acquisition 1.1, 45–60.

Varonis, E. & S. M. Gass (1985). Miscommunication in native/non-native conversation. Language inSociety 14.3, 327–343.

Wagner-Gough, J. & E. Hatch (1975). The importance of input data in second language acquisitionstudies. Language Learning 25.2, 297–307.

Watanabe, Y. & M. Swain (2007). Effects of proficiency differences and patterns of pair interactionon second language learning: Collaborative dialogue between adult ESL learners. Language TeachingResearch 11.2, 121–142.

Williams, J. (2001a). Learner-generated attention to form. Language Learning 51.S1, 303–346.Williams, J. (2001b). The effectiveness of spontaneous attention to form. System 29.3, 325–340.Yang, Y. & R. Lyster (2010). Effects of form-focused practice and feedback on Chinese EFL learners’

acquisition of regular and irregular past tense forms. Studies in Second Language Acquisition 32.2, 235–263.

Yanguas, I. (2010). Oral computer-mediated interaction between L2 learners: It’s about time. LanguageLearning and Technology 14.1, 72–93.

Yashima, T., P. D. MacIntyre & M. Ikeda (2016). Situated willingness to communicate inan L2: Interplay of individual characteristics and context. Language Teaching Research. doi:1362168816657851.

Yilmaz, Y. (2011). Task effects on focus on form in synchronous computer-mediated communication.Modern Language Journal 95.1, 115–132.

Yilmaz, Y. (2016). The role of exposure condition in the effectiveness of explicit correction. Studies inSecond Language Acquisition 38.1, 65–96.

Yoshida, R. (2013). Conflict between learners’ beliefs and actions: speaking in the classroom. LanguageAwareness 22.4, 371–388.

Yule, G. & D. Macdonald (1990). Resolving referential conflicts in L2 interaction: The effect ofproficiency and interactive role. Language Learning 40.4, 539–556.

Zhao, S.-Y. & J. Bitchener (2007). Incidental focus on form in teacher–learner and learner–learnerinteractions. System 35.4, 431–447.

Ziegler, N. (2016a). Synchronous computer-mediated communication and interaction: A meta-analysis.Studies in Second Language Acquisition 38.3, 553–586.

Ziegler, N. (2016b). Taking technology to task: Technology-mediated TBLT, performance, andproduction, Annual Review of Applied Linguistics 36, 136–163.

Ziegler, N., C. Seals, S. Ammons, J. Lake, P. Hamrick & P. Rebuschat (2013). Interaction in conversationgroups: The development of L2 conversational styles. In K. McDonough & A. Mackey (eds.),269–292.

SHAWN LOEWEN (Ph.D. University of Auckland) is Professor in the Second Language Studies andMATESOL programmes in the Department of Linguistics, Germanic, Slavic, Asian and AfricanLanguages at Michigan State University. His research interests include instructed SLA, L2 interaction,and quantitative research methodology. He has conducted descriptive and quasi-experimental researchin both Spanish and ESL classrooms. He has co-authored two books, Key concepts in second languageacquisition and An A-Z of applied linguistics research methods, and his sole-authored book, Introduction toinstructed second language acquisition, appeared in 2015. He also co-edited The Routledge handbook of instructedsecond language acquisition (2017) with Masatoshi Sato.

MASATOSHI SATO (Ph.D. McGill University) is Associate Professor in the Department of English atUniversidad Andres Bello, Chile. His research interests include peer interaction, corrective feedback,teacher cognition, and the development of communicative skills especially of EFL learners. He hasexplored these topics in relation to CMC, teacher education, and language testing. He has publishedin various international journals and recently co-edited books from John Benjamins (Peer interaction andsecond language learning with Susan Ballinger: 2016) and Routledge (The Routledge handbook of instructed secondlanguage acquisition with Shawn Loewen: 2017). He is the recipient of the 2014 ACTFL/MLJ PaulPimsleur Award.

terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0261444818000125Downloaded from https://www.cambridge.org/core. Washington University School of Medicine - St Louis, on 10 Jun 2018 at 01:34:11, subject to the Cambridge Core

Saying what we mean: Making a case for ‘language acquisition’ to become ‘language development’

(Please answer in full sentences and use your own words)

1. What is a complex dynamic system? Why does Larsen-Freemen believe that language and language learning happens as part of a complex dynamic system? Can you give a language based example?

2. How are the words acquisition and development different when talking about

language learning? Why does Larsen-Freeman believe that the term development is better for describing the language learning process?

3. Based on this reconceptionalization of the language learning being a developmental rather than a process based on acquisition, what doe this mean for language teachers? ?

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

Lang. Teach.: page 1 of 15 c© Cambridge University Press 2014doi:10.1017/S0261444814000019

Plenary Speech

Saying what we mean: Making a case for ‘language acquisition’to become ‘language development’

Diane Larsen-Freeman University of Michigandianelf@umich.edu

As applied linguists know very well, how we use language both constructs and reflects ourunderstanding. It is therefore important that we use terms that do justice to our concerns. Inthis presentation, I suggest that a more apt designation than MULTILINGUAL or SECOND

LANGUAGE ACQUISITION (SLA) is MULTILINGUAL or SECOND LANGUAGE DEVELOPMENT (SLD). I give anumber of reasons for why I think SLD is more appropriate. Some of the reasons that I pointto are well known. Others are more current, resting on a view of language from a complexsystems perspective. Such a perspective rejects the commodification of language implied bythe term ‘acquisition’, instead imbuing language with a more dynamic quality, implied by theterm ‘development’, because it sees language as an ever-developing resource. It alsoacknowledges the mutable and interdependent norms of bilinguals and multilinguals. Inaddition, this perspective respects the fact that from a target-language vantage point, regressin learner performance is as characteristic of development as progress. Finally, and mostappropriately for AILA 2011, the term SECOND LANGUAGE DEVELOPMENT fits well with the themeof the congress – harmony in diversity – because it recognizes that there is no commonendpoint at which all learners arrive. For, after all, learners actively transform their linguisticworld; they do not merely conform to it.

1. Introduction

Modern-day interest in second language acquisition (SLA) began over forty years ago. A lothas been learned in these four decades. However, it may be time to re-examine its foundingpremise, specifically, that what is of scholarly concern is the ACQUISITION of second languages.As applied linguists know very well, how we use language both constructs and reflects ourunderstanding, also potentially transforming it. For this reason, it is important that we useterms that do justice to our concerns as applied linguists. In this plenary, I suggest that amore apt designation for SLA is MULTILINGUAL or SECOND LANGUAGE DEVELOPMENT (SLD).I adopt the lens of Complexity Theory (CT) to give twelve reasons why I think SLD is a

Revised version of a plenary address given at the AILA Congress, Beijing, 28 August 2011.

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

2 P L E N A R Y S P E E C H

better way to conceptualize this academic area. Although, admittedly, some of the reasonsare overlapping, I maintain that it makes a big difference to our understanding when we use‘development’ rather than ‘acquisition’. I will conclude by saying what the implications arefor this understanding in research and teaching.

So, to begin: we applied linguists, of all people, know that how we use language makesa difference. It makes a difference in how we conceive objects of interest. I coined theterm GRAMMARING, for instance, to challenge the notion of grammar as a static systemwith a finite number of rules. The term LEXICO-GRAMMAR (which I first encountered inthe work of Michael Halliday) is a portmanteau that suggests that it is unproductive toseparate lexicon from grammar. Although Halliday used the term before the advent ofcomputer-assisted corpus analysis, corpus linguistics has certainly demonstrated the factthat grammar and the lexicon are intertwined in the production of phrasal units. WhenEnglish speakers use separate lexical items in speaking of language and culture, insteadof LANGACULTURE as Michael Agar (1995) proposes, we imply that each is different,overlooking the links between the two that cognitive linguists argue are there to bemade. Further, Tecumseh Fitch (2011: 141) speculates that because English lacks a wordfor A DRIVE TO SHARE OUR THOUGHT WITH OTHERS, ‘perhaps this propensity to talkhas escaped detailed consideration in the literature on language evolution’ (written byEnglish speakers). Fortunately, German has the perfect term (Mitteilungsbedurfnis), whichrefers specifically to verbal communication and the basic human drive to talk and sharethoughts and feelings with others (cf. Schumann’s ‘interactional instinct’). I could goon, but I hope I have made my point. It may seem like a simple substitution to useDEVELOPMENT in place of ACQUISITION, but there are important consequences. Languagedoes not just reflect thought, it construes it, and with different construals, new awarenessesarise.

Before I elaborate on the theme of my central message, though, I would like to relate a briefanecdote. I do so in recognition of the fact that I may be one of the few people here who knowswhy the term SLA was not used right at the beginning of its short life, even though we hada model in ‘first language acquisition’. I conducted an informal genealogy of the term SLAto see if I could determine when it was first used. I wasn’t entirely successful in pinpointingits first use, but I did learn that the early articles that most people consider seminal, ‘Thesignificance of learners’ errors’ (Corder 1967) and ‘Interlanguage’ (Selinker 1972), did notcontain the term. Corder and Selinker used ‘second language learning’, as did Cook in 1969,although in the same article Cook used FIRST language acquisition. Later, of course, Krashen(1982) contrasted ‘second language acquisition’ with ‘second language learning’, but that wasfor a different purpose. So although we had an analogue in first language acquisition, whenthe term ‘acquisition’ was initially used to refer to second languages, it was not SLA. Theterm used was ‘L2 acquisition’, and of course this term retains a currency today.

It was likely that SLA was not used originally, in North America at least, because aroundthe same time that studies of SLA were taking off, there was an urban militant group whichwent by the name Simbionese Liberation Army (SLA) and which was active between 1973and 1975 in the United States. Notorious for their bank robberies and their kidnapping ofmedia heiress Patty Hearst, ‘SLA’ appeared in the newspaper headlines for months. Clearly,‘SLA’ was not a good acronym with which to anoint an emerging field of study.

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

D I A N E L A R S E N - F R E E M A N : S A Y I N G W H A T W E M E A N 3

In any case, after the Simbionese Liberation Army was no longer making headlines, theterm ‘SLA’ for L2 acquisition became acceptable. However, the purpose of my talk today isto go one step further and to recommend that we replace SLA with SLD, a move that de Bot,Lowie & Verspoor (2005) also support. I do not really expect to succeed in this campaign,but as I have just mentioned, the field has already changed its name once, so why not again?In the time I have with you, I will suggest that there is a great deal to be gained by renamingthe field and thereby encouraging us to think differently. I will give you twelve reasons forreplacing SLA with SLD, using the lens that CT has given me.

Complexity theorists study complex dynamic systems, in which complexity is emergent,not built in. Complex systems comprise many components. As the components interact, theygive rise to patterns at another level of complexity. In this way, it can be said that complexsystems are self-organizing, showing the emergence of order without central direction or anysort of preformationism. The patterns are dynamic and inseparable from the environment.An example from the natural world, which might help to make these qualities clear, is thatof an eddy in a stream. The whirling eddy is a pattern that emerges from water moleculesas they pass through a channel of a particular dimension and a streambed of a particularcontour. The water molecules that pass through the eddy are always different, yet the patternendures in the flux, until the environment or the rate of water flow changes.

I believe that language is a complex system, one which grows and organizes itself as peopleuse it, just as the water molecules in an eddy self-organize through the flow. As speakersinteract, their language resources change. Importantly, while rules can be used to describesuch resources, they are not the product of rules (Larsen-Freeman 1997).

Now with this very brief introduction to a CT-inspired view of language, let me put forthtwelve reasons for replacing SLA with SLD.

2. Twelve reasons for preferring SLD to SLA

2.1 Avoiding the commodification of language

First, the term ‘acquisition’ implies that language is a commodity to be acquired; somethingthat moves from the external to the internal, something that gets taken in. The word comesfrom the Latin acquistionem, the act of obtaining or getting. According to Douglas Harper’sOnline Etymology Dictionary (www.etymonline.com), its meaning of ‘thing obtained’ came aboutin the fifteenth century. Seen in this light, the end result of language learning could beconceived of as ‘having something’ (Larsen-Freeman 2010).

Such conceptions of language as ‘a thing one has’ are commonplace in applied linguistics.Linguistic knowledge is characterized as a collection of context-independent symbols, such aswords, accompanied by morphosyntactic rules that specify the relationships between them.In SLA terms, it is assumed that once learners have learned an L2 word or rule, they HAVE it.

Indeed, teachers often speak of whether or not their students ‘have’ the past tense, forinstance, and SLA researchers report acquisition orders and staged sequences of development,with the claim that one form is acquired or ‘had’ before another. The (at least) implicit goal

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

4 P L E N A R Y S P E E C H

of teaching and research from this perspective is to explain language acquisition as a processof taking in of linguistic forms as a mental act, albeit one that takes place through interactionwith others. After learners acquire forms, they can then do something with them.

Of course, it is well known in SLA research that establishing when something is acquired(i.e. defined as available for use at another time) is complicated, due to the commonly observedfluctuations in performance. In recognition of the nonlinearity of the learning process, SLAresearchers have gone to considerable lengths to define the point at which it could be saidWITH ANY ASSURANCE that a form has been acquired. For example, soon after the foundingof SLA as a separate area of inquiry, researcher Hakuta (1974) adapted Cazden et al.’s (1975)definition to his own longitudinal study of L2 morphological acquisition. In doing so, hedefined his research subject’s point of acquisition as the ‘first of three consecutive two-weeksamples in which the morpheme is supplied in over 90% of obligatory contexts’. Notice notonly that this definition recognizes the nonlinearity of learning (i.e. the acquisition criterion isdefined in terms of three successive samples), but it also centers on consistency of native-likeuse over time, a point to which I will return later.

Such definitions of acquisition may be necessary, but they also lead to the ‘longstandingtendency towards the objectification of language’ (Dewey 2009: 67). Now, linguists’descriptions of language as an object can be helpful to educators.1 However, as I suggestedearlier, it also possible to conceive of language differently, not as an object, but as a complexdynamic system. Since language is continually changing, even the term ‘target language’ inlanguage teaching is misleading because the target is always moving (Larsen-Freeman 1997).Instead, perhaps we should see language as AN EVER-DEVELOPING RESOURCE, rather thansomething to be acquired once and for all.

Influenced by CT and its close relative, Dynamic Systems Theory (DST), researchers havefocused on an examination of the flexible, transient, dynamic aspects of learner languagewhich emerge from use (Ellis & Larsen-Freeman 2006). Language development is no longerseen as a process of acquiring abstract rules, but as the EMERGENCE of language abilitiesthrough use in real time. Constructions emerge in learner production in a bottom-up fashionfrom frequently occurring patterns of language use2 rather than as a priori components offixed, autonomous, closed, and synchronic systems.

You might be asking how the patterns arise. An image of the Forbidden City, created bythe Swiss/French artist Corinne Vionnet, offers a clue (see www.corinnevionnet.com/site/1-photo-opportunities.html).3 You see that the image appears blurry. The artist created theimage by superimposing many photographs of this landmark. In other words, photos, takenby different tourists, are layered one on top of another. Each tourist who has taken a photo ofthe Forbidden City chooses an ideal vantage point from which to capture the facade. Whatis remarkable is the consistency of the viewpoint, despite the fact that there is no marking onthe pavement that indicates where a tourist should stand to capture the optimal image. Theartist muses that tourists may be seeking perfect symmetry in their composition, or perhaps

1 As I am the co-author of an 800-page book on English grammar for teachers, it would be hypocritical of me to suggestotherwise.2 This is not to suggest that frequency is the sole explanation for the emergence of new constructions in learner language.3 Thanks to Nick Ellis for alerting me to the website.

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

D I A N E L A R S E N - F R E E M A N : S A Y I N G W H A T W E M E A N 5

they frame the photo as they do because they are socially conditioned, having seen earlierimages of the Forbidden City on postcards or in travel brochures, for instance.

This perspective on the Forbidden City, a product of overlapping images, seems to meto be a good way of understanding language emergence. Each exemplar that a languageuser/learner encounters is similar to earlier ones, but also a bit different. On the basis of thisexperience, a user/learner constructs his or her own version, with similarity, but not absolutefidelity, to previous exemplars. Indeed, phoneticians have known for years that even withinthe same speaker, the pronunciation of the same word varies with each use (Milroy & Milroy1999). A dynamic system is built up through iteration or recursion, just as the image of theForbidden City is constructed from many photographs. However, as we have also seen withthe image of the Forbidden City, iterations of a pattern result in an imprecise or blurry image.By analogy, in the case of language use, we can come to understand why no two speakersspeak exactly the same way. Iteration is not repetition or exact copying.

Because language is a dynamic system, continuously changing, its potential is boundless,never fully realized. In CT terms, the state space of a language-using system represents thelandscape of possibilities of a system (Larsen-Freeman & Cameron 2008).

2.2 Acknowledging regress as well as progress (from the perspective of target-like use)

Acquisition is irreversible, but development, at least as the term has been used with secondlanguages, allows for the possibility of regress from ‘target-like use’ as well as progress. ‘Byusing “development” rather than “acquisition”, we want to make it clear that linguistic skillscan grow and decline, and that following from this, language acquisition and language attritionare equally relevant outcomes of developmental processes’ (de Bot & Larsen-Freeman 2011:6). In other words, what is used by a language learner on one occasion may not be used bythe same learner on a subsequent occasion, despite its relevance. One construction may seemsecure, but when a new, competing (potentially overlapping) construction is introduced, thesemantic space they both occupy needs to be renegotiated and reallocated. In the meantime,the gains that appear to have been made with regard to target-like use of the first constructionseem to disappear. For example, this is commonly observed in learners of English who appearto understand and to be able to use the simple past tense; however, when the present perfectis introduced, it disrupts, at least temporarily, the learners’ correct use of the past tense.

Here is a trilingual example from a paper written in 2010 by Maria Mueller, one of myAustrian students at the University of Innsbruck:

A German speaker from South Tyrol (Austria), studying Italian as an L2, introducedherself:

Mi chiamo Anna e ho passato tre anni alla scuola media.‘My name is Anna and I spent three years in middle school.’

Even though at this point in time the learner’s Italian utterance is grammatically correct, thisdoesn’t mean that her language use has become fixed. After subsequently studying English,she later introduced herself this way:

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

6 P L E N A R Y S P E E C H

Me chiamo Anna e ho spento tre anni alla scola media.

Here, the learner has combined her L3 with the L2. The English verb ‘to spend’ can beused with expressions of time (e.g., ‘to spend a week’); however, in Italian, the verb spendere

only collocates with money, while time is expressed with the Italian verb passare. In otherwords, in this second introduction, the learner used the English verb for spending time andadded to it the Italian past tense morpheme. Her use of Italian thus ‘regressed’, in that herfirst exemplar was more target-like while the second was a mix of Italian and English. Froma traditional way of looking at learner language production, Anna’s second utterance couldbe described as an error. From a CT perspective, however, such linguistic innovations arecharacteristic of both development and use. As Long (2009: 380) states, ‘development is notalways target-oriented’. Herdina & Jessner (2002) assert that growth and decline are normalphenomena in developing systems, both are developmental, but the direction of changedepends on the impact of internal and external resources.

2.3 ‘Acquiring’ a language implies there is an endpoint

‘Acquisition’ suggests completion, but ‘development’ is never complete. For one thing, ourlanguage resources change. For another, we use our resources differently with differentindividuals and also when we encounter and engage with different genres. While our languageresources are not bounded, they are nevertheless shaped by the in-person, remote, and virtualinteractions in which we engage. As Thelen & Bates (2003) note, the patterns are created anddissolved as tasks and environments change. Preferred patterns become stabilized throughfrequency of use and the strengthening of connection weights in neural networks (Hebb1949), but they are always subject to change. As I have subtitled one of my publications,‘There is no end and there is no state’ (Larsen-Freeman 2006a), contrasting with the idea ofthere being ‘end-state’ grammars.

2.4 Creating new patterns

A fourth reason for talking about development rather than acquisition is that each learner has‘the capacity to create his or her own patterns with meanings and uses (morphogenesis) andto expand the meaning potential of a given language, not just to internalize a ready-madesystem’ (Larsen-Freeman & Cameron 2008: 116).

Whether the gatekeepers of a given language allow learners to do this, i.e. allow a neologismto endure, is of course another question – a political question of who possesses the politicalcapital to innovate – but that is another matter, which I take up later. In any case, it can besaid that

‘language is not a fixed code’ (Harris 1996) that exists independently of its users, and that is ready-madefor users before they start using it, but rather it is created, or at the very least assembled from conventionalunits, each time it is used (Larsen-Freeman & Cameron 2008: 111).

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

D I A N E L A R S E N - F R E E M A N : S A Y I N G W H A T W E M E A N 7

Typically, the innovations come in the form of analogizing from other exemplars. Speakersof English have no trouble when nouns such as ‘e-mail’ come into existence. They easilyconvert them into verbs: ‘I e-mailed to see when the meeting was taking place’. Languageusers (and I include learners) actively transform their linguistic world; they do not just conformto it (Larsen-Freeman 2006b: 285).

That means that there is no homogeneity. We create new patterns from old by analogywhen we want to make new meanings, going beyond what is present in the input (Larsen-Freeman 1997). Furthermore, we use our language resources differently with different others.There is no single homogenized language competence.

2.5 Making meaning, not merely acquiring forms

A closely related fallacy has to do with our definition of what is being acquired. Linguists’descriptions of language can undoubtedly be helpful. However, despite the early morphemestudies of SLA, in which I myself took part, I have come to understand that it is not necessarilya morpheme affixation rule that is being acquired from the learners’ point of view. One reasonis that learners, particularly untutored learners, do not see language as composed of linguists’units. What do such learners know about morphemes? A more important reason is thatlearners are intent on making meaning, and to do this, they will use whatever languageresources are available to them. Not only do their resources not necessarily coincide withlinguists’ units, but they can also, as we have seen from corpus linguistics, be diverse formsof various lengths. For instance, it seems that the verbs to which morphemes are attachedare often perceived by learners as monomorphemes or as one unit. I say this because, as Ihave seen from my own research in the mid-1970s, there is often a preference to use certainmorphemes with certain verbs; that is, with some verbs, a morpheme is much more likelyto be used than with others. Learners’ performance cannot stem from acquiring form-basedrules, therefore, since learners’ development is uneven: much more gradient in nature.

2.6 Participating as much as acquiring

It has been argued that learners participate in language as much as they acquire it (Sfard1998). In other words:

Language is socially constructed [ . . . ]. Language use, social roles, language learning, and consciousexperience are all socially situated, negotiated, scaffolded, and guided. (Ellis & Larsen-Freeman 2006:572)

In CT/DST terms, we may speak of the [social] process of co-adaptation, in which thelanguage resources of two or more people are modified through their interaction. As speakersinteract, they modify their language, in part to align, or to show solidarity, with the otherspeaker, and in part to accommodate to the other’s comprehension. For instance, as a child

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

8 P L E N A R Y S P E E C H

and its caretaker interact, their language resources are dynamically altered, as each adaptsto the other. The social context thus affords the possibility for co-adaptation.

Shanker & King (2002) note that an apt metaphor for an information processing viewof communication is a fax machine, where messages are transmitted from one machine toanother. A CT/DST approach, on the other hand, finds the metaphor of a delicate dancemore appropriate. A dance involves many different steps, each unique, and each negotiatedas the dance progresses. According to Shanker & King, perfectly matched dancers display‘interactional synchrony’ in which the partners are mutually attuned in an interactionalprocess. Even such dancers, however, do not always perform seamlessly. They exhibit wavesof asynchrony and synchrony, striving to adapt to their partners’ motion (de Bot, Lowie &Verspoor 2007: 9).

In CT/DST terms, we would say that the person is coupled with his or her environment.Van Geert & Fischer (2009: 327) write that development applies to person–context assembliesacross time. Contexts are no longer seen as independent variables, as background to themain developmental drama. Of course, it is analytically possible to separate the person fromthe context; however, that separation requires the untenable assumption that the two areindependent (van Geert & Steenbeek 2005).

Quoting Keats, Kramsch (2002) makes a similar point. She asks ‘How can we know thedancer from the dance?’ From a CT perspective, we would not draw the line between aperson and context because development is never a function of a person or context alone,but results as a function of their dynamic interaction (Thelen & Smith 1998: 575). Indeed,we can never have complete information about a system in that ‘[a] system is never optimallyadapted to an environment since the process of evolution of the system will itself change theenvironment so that a new adaptation is needed, and so on’ (Heylighen 1989: 2).

2.7 Discouraging comparisons with monolinguals

An SLA perspective may not discourage comparisons with monolinguals. As I noted earlier,it is often the case that for research purposes L2 acquisition is measured by distance from amonolingual native speaker target. Bley-Vroman (1983) cautioned against this practice thirtyyears ago when he argued that it was important to look at the learner’s developing secondlanguage system in its own right. Nonetheless, the ‘comparative fallacy’, whereby learners’language production is compared to that of monolingual speakers of the language, endures tothis day. Others have also objected, using such concepts as ‘multicompetence’ (Cook 1991),which aims to counter a deficit view of language development. In other words, rather thanseeing L2 learners as somehow defective compared with native speakers, Cook feels thatthey should be perceived as multicompetent individuals with knowledge of more than onelanguage in the same mind.

When we speak about the multicompetence of a plurilingual person, we are not just talkingabout someone who can use more than one language; we are also adopting a different image,one in which several languages and several cultures interact to create a global, complexcompetence – a competence, which includes the L1, or L1s, and other known languages.

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

D I A N E L A R S E N - F R E E M A N : S A Y I N G W H A T W E M E A N 9

This competence changes and evolves as the person goes through new linguistic and culturalexperiences. It is fluid, not static.

In her plenary address at the American Association for Applied Linguistics Conferencelast year, Ortega proposed that monolingualism should not be the starting point or endpoint or the normal point of comparison for SLA, although it usually is. Elaborating on thisideological position, Ortega (2010) stated that when SLA researchers eliminate ‘ . . . bilinguals’other languages from analysis, the bilinguals’ repertoires are made to look like “less” insteadof “more”. In contrast, insights into bilingualism ‘counter the myth of immutable, fixed,inalienable privileges of the language(s) owned by birth . . . ’. I think that conceiving of theprocess as developmental rather than acquisitional is more likely to discourage comparisonswith monolingual performance.

2.8 Recognizing the sensitivity of a complex dynamic system

Complex systems are dependent on their initial conditions, a phenomenon known in CT asthe BUTTERFLY EFFECT. What this means is that complex systems are highly sensitive: evena small difference at one point in a developmental trajectory can make a big difference as itproceeds. This is because as a learner’s system develops, it ‘functions as a resource for its ownfurther development’ (Larsen-Freeman & Cameron 2008: 158). In other words, language isautopoietic; it contains the seeds for its own development. Learners learn a new language byreferring to what they already know, and what they already know constantly changes withmeaningful language use.

Speaking in biological terms,

Taking a [complex] systems perspective on developmental processes means, among other things, attendingto the ways in which the developing organism functions as a resource for its own further development.The organism helps determine which other resources will contribute to that development, as well asthe impact they will have. The roles played by the vast and heterogeneous assembly of interactants thatcontribute to a life-course are system-dependent and change over time. (Oyama, Griffiths & Gray 2001:5).

2.9 Acknowledging variation

There is no one target language. Even if one claimed that the target was a standard form oflanguage, it would be inadequate. For one thing, standard languages differ depending on theirgeographic location. Standard British English is different from Standard North AmericanEnglish, for instance. Furthermore, there is a great deal of variation in language use, thatis, even within a dialect there is little homogeneity. This is certainly the case in the currenthistorical period, where people ‘shuttle rapidly between communities and communicativecontexts, in both virtual and physical space’ (Canagarajah 2006: 25). However, variation inlanguage use is not only symptomatic of postmodern globalization.

Walker (2010) asserts that variation is an entirely natural phenomenon and a basic fact oflanguage life. Indeed, without variation, languages would be unable to serve speakers’ needs:

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

1 0 P L E N A R Y S P E E C H

‘Heterogeneity is . . . necessary to satisfy the linguistic demands of everyday life’ (Labov 1982:17). Walker goes on to point out the ways in which variation originates, not only through theevolution prompted by geographical distance and accompanying isolation, but also thoughsocial contact, creativity, and deliberate self-renewal as languages adapt in order to meet theneeds of generations of speakers. He offers the growing use of English as a Lingua Franca(ELF) as a case in point.

Thus, as we have seen already, a language is not a single homogeneous construct to beacquired. Rather, a complex systems view sees any stability in language arising from use. Sucha view ‘foregrounds the centrality of variation among different speakers and their developingawareness of the choice they have in how they use patterns within a social context’ (Larsen-Freeman & Cameron 2008: 116). This awareness fits much better with a developmentalperspective, I believe.

2.10 Acquiring a language is not a homogeneous activity either

What does it mean to say that one has acquired another language? Caspi (2010), for example,discusses the ‘receptive-productive’ vocabulary gap and finds evidence for a continuum ofdevelopment from a learner’s perception/recognition of new words to their (spontaneous)production. In other words, in the learning of vocabulary words alone, we cannot speakof acquisition as a completed process because different facets of development proceed atdifferent paces. Caspi explains that this unevenness is due to the interactions between levelsof vocabulary knowledge and to the limited resources of the developing system.

A highly successful language learner, Elka Todeva (2009: 68), adds to this her subjectiveexperience:

One of the questions that I always find difficult to answer is ‘How many languages do you speak?’ Oftenasked this question, I find myself prefacing my answer with ‘Well, I have formally studied quite a fewlanguages and have a high level of comprehension in still more’. Then, as a rule, I explain that Englishand Russian are perhaps the only languages I speak fluently, in addition to my L1, of course. I often wishpeople would ask instead ‘How many languages can you read in?’ or ‘In how many languages do youhave decent listening comprehension?’

The point is, when we speak of acquisition, we must remain mindful that, just as withlanguage itself, development is not homogeneous.

2.11 Recognizing the bi-directionality of transfer

Transfer of the patterns of one language to another by learners is not unidirectional. Indeed,McWhorter (2002) shows that English was adopted by Scandinavian invaders of the BritishIsles, but that it was also altered by them. Acquisition in SLA has always implied directionalityfrom an L1 to an L2, but there is research to support the fact that interactions in the L2 caninfluence the L1 as well (Cenoz, Hufeisen & Jessner 2001; Cook 2003).

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

D I A N E L A R S E N - F R E E M A N : S A Y I N G W H A T W E M E A N 11

Pavlenko & Jarvis (2002), for instance, found evidence of what they called ‘bidirectionaltransfer’. The English spoken by the Russian L2 learners in their study was influenced by theirnative Russian, and the same L2 learners’ English in turn influenced their Russian. Grosjean’sclaim that the nature of multicompetence is dynamic and flexible, and that bilinguals arenot two monolinguals in one (Grosjean 1992), further supports this observation. As Pavlenko& Jarvis write ‘the evidence of bidirectional transfer underscores the unstable nature of“native-speakerness”’ (2002: 210).

Kroll, Gerfen & Dussias (2008: 109) observe that most of the evidence suggests that lexicalaccess is nonselective in that alternatives in both languages appear to be activated in parallelin bilinguals when words are processed. Others, too, have found evidence of the influenceof the L2 on the L1 in the areas of phonology (Flege 1987), rhetorical preferences (Kubota1998), and expression of motion (Brown & Gullberg 2008); for further discussion, see Ortega(2010). In fact, this phenomenon is not limited to incipient bilinguals. Reverse interactionshappen even at intermediate proficiency levels (Jarvis & Pavlenko 2008).

2.12 Remembering the learner

Acquisition is applied to language; development is, too, but development can also apply tothe learner.

Van Geert & Fischer (2009: 313–314) tell us that:

Etymologically, the term ‘development’ means unwrapping, that is taking something out of its originalwraps, thus uncovering something that is present at the beginning but concealed. As a verb develop appearsin a transitive as well as an intransitive, more precisely reflexive, form. For instance, you can develop aphoto (people used to do these things before the digital camera) or develop a real estate project. However,it is a bit odd to say that a parent develops a child. One can say that the child develops grasping, or thatthe child develops. In the latter case, develop is used as a reflexive verb, that is, a verb where the agent andthe patient are the same . . . Thus, development is an internally driven process of growing that is in someway potentially present from the beginning. This is not, in any way, a form of nativism, claiming that theend state (or at least some essential core of it) is present from the beginning. Potential is derived from theLatin word for power, thus meaning that something at the beginning has the power of developing intosome particular final state through transformations that unwrap the potential by making it actual.

Although I would take exception to their use of ‘final state’, I think that van Geert &Fischer’s way of looking at development, a way that involves the learner in the developmentalprocess, is a much more accurate portrayal of the learner’s involvement than the use of theterm ‘acquisition’. The learner develops his/her language resources, and in so doing, thelearner develops. This way of putting it seems to me to be much more respectful of thelearner’s agency.

3. Conclusion

As I said at the outset of this talk, I do not really expect professional discourse to change. SLAis a well-established acronym for the field, and is likely to remain so. However, there is value

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

1 2 P L E N A R Y S P E E C H

to entertaining a different term and therefore a different way of thinking, one that centers onlearning and learners.

One implication, for instance, is that researchers must always remember to look at what alearner is doing from the learner’s point of view. The learner may not be thinking in termsof linguists’ units or linguists’ descriptions. It would also mean that, seeing development as acontinuous process of change over time, we could ask questions such as ‘How do the moment-to-moment aspects of language learning map onto the longer timescales of development?’(Elman 2003); ‘Where do novel language behaviors come from?’; ‘What are the motors ofchange?’ (Thelen & Corbetta 2002).

Furthermore, the process of development cannot truly be understood if we confineourselves to looking at the SLA process in the usual way, which is through investigatingassociations between variables across populations. Nor can we solely calculate group meansand believe that we are getting the full picture. Looking only at averages obscures theindividual differences between learners and makes us overlook statistical ‘outliers’, whosebehavior may be interesting and relevant.

Another implication is that assessment would take on an entirely different character. If weare not comparing the learners’ performance to that of native speakers of the language, then,according to Ortega (2010),

As much as possible, we would analyze complete repertoires of bilinguals across languages, so as to alwayssearch for ‘more’ rather than ‘less’ in bilinguality and so as to keep in sight that neither L2 nor L1 areimmutable or fixed.

Further,

We would conduct non-normative comparisons with multilingual natives and other diverse multiple-language learning profiles over the life span.

To these calls, I add we would maintain an empirical focus on experience over time, lookingfor periods of increased variability in a learner’s production as a sign that a phase transitionis about to take place. We would also look for expressions of creativity, likely identifiable morereadily from a retrodictive examination (Larsen-Freeman & Cameron 2008).

From a teaching perspective, we would come to understand that in multilingual languagedevelopment, it is most important to know what the learner is entertaining about the languagethat is developing. After all, we don’t only teach language, we teach learners. And teachinga language does not involve the transmission of a closed system of knowledge. Learners arenot engaged in simply learning fixed forms or sentences, but rather in learning to adapt theirbehavior to an increasingly complex environment.

Furthermore, learners learn to use language, like the steps in a dance, through repeatedactivity in slightly different situations. Learning is not a linear, additive process, but an iterativeone. Linguistic structures do not only symbolize reality; they are used to actively constructreality in interaction with others (Kramsch & Whiteside 2008).

We would also need to teach language as the meaning-making resource it is, teachingreasons as much as rules (Larsen-Freeman 2003). If students understand why something

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

D I A N E L A R S E N - F R E E M A N : S A Y I N G W H A T W E M E A N 13

functions the way that it does in a language – the reason behind the form – they are muchmore empowered to go beyond the given to create something anew.

Finally, and most appropriately for AILA 2011, the term SECOND LANGUAGE DEVELOPMENT

fits well with the theme of our congress – harmony in diversity – because it recognizes thatthere is no common endpoint at which all learners arrive. It is in the variation that strengthresides. This variation is a sign of our uniqueness, but is also a stimulus for growth. For afterall, as I said earlier, learners/users of a language actively transform their linguistic world;they do not merely conform to it. And this, we do together.

Acknowledgements

I wish to thank the hosts of the Congress for their hospitality and attentiveness: China EnglishLanguage Education Association, Beijing Foreign Studies University, National ResearchCentre for Foreign Language Education, and the Foreign Language Teaching and ResearchPress.

References

Agar, M. (1995). Language shock: Understanding the culture of conversation. New York: William Morrow.Bley-Vroman, R. (1983). The comparative fallacy in interlanguage studies: The case of systematicity.

Language Learning 33, 1–17.Brown, A. & M. Gullberg (2008). Bidirectional crosslinguistic influence in L1-L2 encoding of manner

in speech and gesture: A study of Japanese speakers of English. Studies in Second Language Acquisition30, 225–251.

Canagarajah, A. S. (2006). Globalization of English and changing pedagogical priorities: Thepostmodern turn. In B. Beaven (ed.), IATEFL 2005 Cardiff conference selections. Canterbury, UK:IATEFL, 15–25.

Caspi, T. (2010). A dynamic perspective on second language development. Unpublished doctoraldissertation, University of Groningen.

Cazden, C., H. Cancino, E. Rosansky & J. Schumann (1975). Second language acquisition sequences inchildren, adolescents, and adults. Cambridge, MA: Harvard University Graduate School of Education.

Cenoz, J., B. Hufeisen & U. Jessner (2001). Cross-linguistic influence in third language acquisition: Psycholinguisticperspectives. Clevedon, UK: Multilingual Matters.

Cook, V. (1969). The analogy between first and second language learning. IRAL VII(3), 207–216.Cook, V. (1991). The poverty-of-the-stimulus argument and multicompetence. Second Language Research

7, 103–117.Cook, V. (ed.) (2003). Effects of the second language on the first. Clevedon, UK: Multilingual Matters.Corder, S. P. (1967). The significance of learners’ errors. International Review of Applied Linguistics 5,

161–170.de Bot, K. & D. Larsen-Freeman (2011). Researching second language development from a dynamic

systems theory perspective. In M. Verspoor, K. de Bot & W. Lowie (eds.), A dynamic approach to secondlanguage development: Methods and techniques. Amsterdam: Benjamins, 5–24.

de Bot, K., W. Lowie & M. Verspoor (2005). Second language acquisition: An advanced resource book. London:Routledge.

de Bot, K., W. Lowie & M. Verspoor (2007). A dynamic systems theory approach to second languageacquisition. Bilingualism, Language and Cognition 10.1, 7–21.

Dewey, M. (2009). English as a lingua franca: Heightened variability and theoretical implications.English as a lingua franca: Studies and findings. Newcastle upon Tyne: Cambridge Scholars Publishing,60–83.

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

1 4 P L E N A R Y S P E E C H

Ellis, N. C. & D. Larsen-Freeman (2006). Language emergence: Implications for applied linguistics.Introduction to the special issue. Applied Linguistics 27.4, 558–589.

Elman, J. (2003). Development: It’s about time. Developmental Science 6, 430–433.Fitch, W. T. (2011). ‘Deep homology’ in the biology and evolution of language. In A. M. Di Sciullo &

C. Boeckx (eds.), The biolinguistic enterprise: New perspectives on the evolution and nature of the human languagefaculty. Oxford: Oxford University Press, 135–166.

Flege, J. (1987). The production of ‘new’ and ‘similar’ phones in a foreign language: Evidence for theeffect of equivalence classification. Journal of Phonetics 15, 47–65.

Grosjean, F. (1992). Another view of bilingualism. In R. Harris (ed.), Cognitive processing in bilinguals.Amsterdam: North-Holland, 91–103.

Hakuta, K. (1974). A preliminary report on the development of grammatical morphemes in a Japanesegirl learning English as a second language. Working Papers on Bilingualism 3, 18–43.

Harris, R. (1996). Signs, language, and communication. London: Routledge.Hebb, D. (1949). The organization of behavior. New York: Wiley.Herdina, P. & U. Jessner (2002). A dynamic model of multilingualism. Clevedon, UK: Multilingual Matters.Heylighen, F. (1989). Self-organization, emergence and the architecture of complexity. Proceedings of the

1st European Conference on System Science, AFCET, Paris, 23–32.Jarvis, S. & A. Pavlenko (2008). Crosslinguistic influence in language and cognition. Mahwah, NJ: Erlbaum.Kramsch, C. (ed.) (2002). Language acquisition and language socialization: Ecological perspectives. London:

Continuum.Kramsch, C. & A. Whiteside (2008). Language ecology in multilingual settings: Towards a theory of

symbolic competence. Applied Linguistics 29.4, 645–671.Krashen, S. D. (1982). Principles and practice in second language acquisition. Oxford: Pergamon.Kroll, J. F., C. Gerfen & P. E. Dussias (2008). Laboratory designs and paradigms: Words, sounds,

and sentences. In L. Wei & M. Moyer (eds.), The Blackwell guide to research methods in bilingualism andmultilingualism. Cambridge, MA: Blackwell, 108–131.

Kubota, R. (1998). An investigation of L1-L2 transfer in writing among Japanese university students:Implications for contrastive rhetoric. Journal of Second Language Writing 7, 69–100.

Labov, W. (1982). Building on empirical foundations. In W. P. Lehman & Y. Malkiel (eds.), Perspectiveson historical linguistics. Amsterdam: Benjamins, 17–92.

Larsen-Freeman, D. (1997). Chaos/complexity science and second language acquisition. AppliedLinguistics 18.2, 141–165.

Larsen-Freeman, D. (2003). Teaching language: From grammar to grammaring. Boston, MA: Heinle/Thomson.Larsen-Freeman, D. (2006a). Second language acquisition and the issue of fossilization: There is no

end, and there is no state. In Z.-H. Han & T. Odlin (eds.), Studies of fossilization in second languageacquisition. Clevedon, UK: Multilingual Matters, 189–200.

Larsen-Freeman, D. (2006b). On the need for a new understanding of language and its development.Journal of Applied Linguistics 3l.3, 281–304.

Larsen-Freeman, D. (2010). Having and doing: Learning from a complexity theory perspective. InP. Seedhouse, S. Walsh & C. Jenks (eds.), Conceptualising learning in applied linguistics. Basingstoke, UK:Palgrave Macmillan, 52–68.

Larsen-Freeman, D. & L. Cameron (2008). Complex systems in applied linguistics. Oxford: Oxford UniversityPress.

Long, M. H. (2009). Methodological principles for language teaching. In M. H. Long & C. J. Doughty(eds.), Handbook of language teaching. Oxford: Blackwell, 373–394.

McWhorter, J. (2002). The power of Babel: A natural history of language. London: Heinemann.Milroy, J. & L. Milroy (1999). Authority in language (3rd edn). New York: Routledge.Ortega, L. (2010). The bilingual turn in SLA. Plenary address delivered at the Annual Conference of the

American Association for Applied Linguistics. Atlanta, Georgia, March 6–9.Oyama, S., P. Griffiths & R. Gray (eds.) (2001). Cycles of contingency, development systems and evolution.

Cambridge, MA: MIT Press.Pavlenko, A. & S. Jarvis (2002). Bidirectional transfer. Applied Linguistics 23, 190–214.Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics 10.3, 209–231.Sfard, A. (1998). On two metaphors for learning and the dangers of choosing just one. Educational

Researcher 27, 4–13.Shanker, S. & B. King (2002). The emergence of a new paradigm in ape language research: Target

article. Behavioral & Brain Sciences 25, 605–626.

http://journals.cambridge.org Downloaded: 25 Jun 2014 IP address: 134.153.184.170

D I A N E L A R S E N - F R E E M A N : S A Y I N G W H A T W E M E A N 15

Thelen, E. & E. Bates (2003). Connectionism and dynamic systems: Are they really different?Developmental Science 6, 378–391.

Thelen, E. & D. Corbetta (2002). Microdevelopment and dynamic systems: Applications to infantmotor development. In N. Granott & J. Parziale (eds.), Microdevelopment. Cambridge: CambridgeUniversity Press, 59–79.

Thelen, E. & L. Smith (1998). Dynamic systems theories. In W. Damon & R. Lerner (eds.), Handbookof child psychology Volume 1: Theoretical models of human development (5th edn). New York: Wiley, 563–634.

Todeva, E. (2009). Multilingualism as a kaleidoscopic experience: The mini-universes within. InE. Todeva & J. Cenoz (eds.), The multiple realities of multingualism. Berlin: Mouton De Gruyter, 53–74.

van Geert, P. & F. W. Fischer (2009). Dynamic systems and the quest for individual-based models ofchange and development. In J. P. Spencer, M. S. C. Thomas & J. McClelland (eds.), Toward a newgrand theory of development? Connectionism and dynamic systems theory reconsidered. Oxford: Oxford UniversityPress, 313–336.

van Geert, P. & H. Steenbeek (2005). Explaining after by before: Basic aspects of a dynamic systemsapproach to the study of development. Developmental Review 25, 408–442.

Walker, R. (2010). Teaching the pronunciation of English as a lingua franca. Oxford: Oxford University Press.

DIANE LARSEN-FREEMAN is Professor Emerita of Education, Professor Emerita of Linguistics, andResearch Scientist Emerita at the University of Michigan. She is also a Visiting Professor of EducationalLinguistics at the University of Pennsylvania. She has published widely in the areas of English linguistics,second language development, language teaching, language teacher education, and ComplexityTheory.

Second Language Learning Explained? SLA across 10 Contemporary Theories

(Please answer in full sentences and use your own words)

1. What are the 10 observations regarding language learning that need to be explained by any theory of second language acquisition or development?

2. Which theories of take the position that formal instruction can help or is necessary in

the language learning process? ?

3. When assessing second language proficiency and attainment should we compare them to monolingual users? Why or why not? What research supports this point of view?

As a field, Second Language Acquisition (SLA) is vibrant. It began in the late 1960s, with key initial developments during the decade of the 1970s. Most schol-ars agree that the coming of age as an autonomous discipline happened some time at the end of the 20th century, after less than 40 years of exponential growth. Since then, a prodigious expansion in research and theorizing has occurred that continues unabated at the time of this writing. The field of SLA is also decidedly interdisciplinary, both in its origins and its development, a quality that is felt in the epistemological diversity of its theories as well. SLA interconnects with four neighboring fields, some of them also relative newcomers in academia: language teaching, linguistics, child language acquisition, and psychology. In more recent years, it has also developed ties with other disciplines, notably bilingualism, cogni-tive science, education, anthropology, and sociology. Given this vibrant disciplinary landscape, the second decade of the 21st century is an opportune time to reflect on the theories in SLA that offer the most viable explanations about humans’ capacity to learn additional languages (henceforth L2) later in life, after having learned—from birth to roughly age four—the first language (in the case of a monolingual upbringing) or languages (in the case of a bi/multilingual upbringing).

The editors of this collection, VanPatten and Williams, offer 10 observations based on well-established empirical findings in SLA (see Chapter 1 ). They reason that these agreed upon “observed phenomena” need to be explained by any viable theory of second language acquisition, or at least incorporated in them in some formal fashion. For the sake of economy, the ten facts can be combined into five central areas that have occupied the attention of most SLA researchers to date: (a) the nature of second language knowledge and language cognition, (b) the nature of interlanguage development, and the contributions of (c) knowledge of the first language (L1), (d) the linguistic environment, and (e) instruction. Table 13.1

13 SECOND LANGUAGE LEARNING EXPLAINED? SLA ACROSS 10 CONTEMPORARY THEORIES

Lourdes Ortega

246 Ortega

presents these areas and the ten associated observations that VanPatten and Wil-liams offer in their Chapter 1 .

In keeping with the overall purpose of the present collection to introduce unini-tiated readers to key SLA theories, I have two goals in this closing chapter. First, I hope to help readers review their understanding of the 10 contemporary theories of SLA gathered in this book. To do so, I will contrast and compare the position each theory takes with regard to the five key areas outlined in Table 13.1 . The comparison for each area is summarized in Figures 13.1 through 13.5 . In each figure, the theories or approaches are ordered slightly differently from their chapter order in the book, so as to more conveniently show commonalities and differences among them visually. My second goal is to pique readers’ intellectual curiosity and encourage them to pursue further study of SLA. In the hope of achieving this goal, I conclude the chapter with a small glimpse of some of the exciting but complex challenges that theories in SLA will likely have to tackle in the future.

The Nature of Language Knowledge and Language Cognition

Each of the theories featured in this book offers a different take on the nature of L2 knowledge, depending on the view of human language cognition that they espouse. Figure 13.1 offers a summary of the key differences and similarities in this area across the 10 theories.

The linguistic theory of Universal Grammar is affiliated with the field of generative Chomskyan linguistics and therefore adopts a linguistic view of lan-guage cognition. It offers the following logical argument. If L2 learners possess abstract knowledge of ambiguity and ungrammaticality that could have never been derived from the linguistic input available in the environment or from their

TABLE 13.1 Ten Observations That Every SLA Theory Needs to Explain

Construct Observations

Knowledge and cognition

# 2. A good deal of SLA happens incidentally . # 3. Learners come to know more than what they have been exposed to in

the input.

Interlanguage # 4. Learners’ output (speech) often follows predictable paths with predictable stages in the acquisition of a given structure .

# 5. Second language learning is variable in its outcome . # 6. Second language learning is variable across linguistic subsystems.

First language # 8. There are limits on the effect of a learner’s first language on SLA .

Linguistic environment

# 1. Exposure to input is necessary for SLA. # 7. There are limits on the effects of frequency on SLA. #10. There are limits on the effects of output (learner production) on language

acquisition.

Instruction # 9. There are limits on the effects of instruction on SLA.

Source: From Chapter 1.

FIG

URE

13.

1 T

he n

atur

e of

lang

uage

kno

wle

dge

and

lang

uage

cog

nitio

n in

10

SLA

theo

ries

.

Theo

ry

Cog

nitio

n L

earn

ing

Rep

rese

ntat

ion

Uni

vers

al G

ram

mar

The

ory

Ling

uist

ic:

nativ

ism, m

odul

arity

Im

plic

it ab

stra

ct d

educ

tion

Sym

bolic

and

unc

onsc

ious

Usa

ge-b

ased

app

roac

hes

Psyc

holo

gica

l: ge

nera

l cog

nitiv

e ar

chite

ctur

e

Impl

icit

indu

ctio

n, a

ssoc

iatio

n,

anal

ogy,

abst

ract

ion

Unc

onsc

ious

Skill

Acq

uisit

ion

The

ory

Exp

licit

dedu

ctio

n, a

utom

atiz

atio

n D

ecla

rativ

e, th

en p

roce

dura

l

Dec

lara

tive/

Proc

edur

al m

odel

N

euro

biol

ogic

al: g

ener

al

cogn

itive

arc

hite

ctur

e B

oth

expl

icit

dedu

ctio

n an

d au

tom

atiz

atio

n in

par

alle

l, w

ith in

vers

e de

gree

of

relia

nce

depe

nden

t on

age

and

con

text

Dec

lara

tive

and

proc

edur

al

mem

ory

repr

esen

tatio

ns a

t on

ce,

with

red

unda

nt c

ompe

titiv

e, an

d in

hibi

tory

eff

ects

Inpu

t Pr

oces

sing

The

ory

Und

ersp

ecifi

ed

Exp

licit/

impl

icit

inte

rfac

e?

UG

?

Proc

essa

bilit

y T

heor

y U

nspe

cifie

d Fu

nctio

nal u

nive

rsal

s

Con

cept

-ori

ente

d ap

proa

ch

Inte

ract

ion

fram

ewor

k E

xplic

it/im

plic

it in

terf

ace?

B

oth

cons

ciou

s an

d im

plic

it

Vyg

otsk

ian

Soci

ocul

tura

l T

heor

y So

cioc

ultu

ral:

rela

tiona

l, sit

uativ

e, m

edia

ted

Con

scio

us, g

oal-

driv

en, m

edia

ted

Con

scio

us, i

nter

men

tal,

then

in

tra-

men

tal

Com

plex

ity T

heor

y So

cioc

ogni

tive:

ord

erly

, dy

nam

ic, s

elf-

and

co

-ada

ptiv

e

Ada

ptat

ion

and

tran

sfor

mat

ion

driv

en; o

rien

ted

tow

ard

arisi

ng

affo

rdan

ces

Sens

itive

to

initi

al c

ondi

tions

, eve

r op

en-e

nded

, sub

ject

to c

hang

e th

at is

re

trod

icta

ble

rath

er t

han

pred

icta

ble

248 Ortega

L1 knowledge alone, we must assume that the knowledge was already there, in some initial form at least, independent from experience. That is, the theory is committed to nativism. Furthermore, it is also committed to the idea that learners are constrained in their learning task (in the positive sense of “guided”) by this preexisting initial grammatical knowledge they possess. Another theoretical com-mitment is to modularity. In other words, in this theory it is posited that language is distinct from other forms of cognition, a separate faculty, an organ of the mind. (In the specialized literature the terms language-specific and domain-specific are also used to refer to this same notion.) Finally, in Universal Grammar, language knowledge is thought to be symbolic, that is, rule based. This symbolic knowledge is posited to be formal, highly abstract, and unconscious or tacit, represented in the mind in the form of principles and parameters (in Universal Grammar) and features and functional categories (in more recent versions of Chomskyan theory). Consequently, the theory predicts that core grammatical knowledge (of a first or second language) unfolds incidentally by deduction from the innate abstract knowledge that predates any linguistic experience. The instantiated rules of the specific language, once acquired, remain implicitly represented.

In sharp contrast stand two theories that also make the nature of L2 knowl-edge central to their explanations: usage-based approaches and Skill Acquisition Theory. Both have their roots in the field of contemporary cognitive psychology and thus both offer a psychological view of cognition. In both theories, language is thought to be learned and used through the same cognitive architecture humans have at their disposal for the acquisition and use of other kinds of knowledge (e.g., knowing about history and biology; or knowing how to cook, how to play the piano, tennis, or chess; or knowing how to solve mathematical equations or do computer programming). Usage-based approaches explain language learning as, by and large, an implicit inductive task. Human language capacities are thought to result from the extraction of statistical patterns from the input. This extrac-tion is fueled by an innate general predisposition of the brain to learn and be shaped by experience, and it is further pushed by communicative needs as the organism interacts with the environment. The extraction of associative patterns is also driven implicitly and ineludibly by the human brain’s predisposition toward probabilistic, statistical learning:

Every time the language learner encounters an exemplar of a construction, the language system compares this exemplar with memories of previous encounters of either the same or a sufficiently similar exemplar to retrieve the correct interpretation . . . [and] the learner’s language system, processing exemplar after exemplar, identifies the regularities that exemplars share, and makes the corresponding abstractions. ( Chapter 5 )

Usage-based approaches are, therefore, committed to incidental learning and unconscious representations. Additional attention via conscious effort at explicitly

Second Language Learning Explained? 249

cracking the language code can help, but the bulk of language learning is statisti-cal and implicit. By contrast, Skill Acquisition Theory focuses on the prototypical case when a skill or expertise is approached through formal instruction as the starting point, for example when people avail themselves of a manual or a tutor to get started with learning tennis, computer programming, chess—or a foreign language. Therefore, the theory is committed to conscious processing, deliberate learning, and explicit representations, with an interface that allows for this explicit knowledge eventually to give rise to expert performance. As DeKeyser puts it in Chapter 6 , learning to become an expert (in language as in anything else) is viewed as a process of turning knowledge into behavior, or “turning ‘knowledge that’ into ‘knowledge how.’” It should be clear, then, that usage-based approaches and Skill Acquisition Theory, while sharing the same basic psychological view of language, differ greatly in the relative importance they accord to implicit and explicit knowledge in explaining SLA.

Together with usage-based approaches and Skill Acquisition Theory, the declarative/procedural model presents also a strong cognitive bent but has its roots in neurobiology rather than psychology. It therefore shares with the other two proposals several important premises about the nature of language knowledge and cognition, while at the same time making some distinct predictions. In agree-ment with usage-based approaches and Skill Acquisition Theory is this model’s orientation to study language as part of general cognition: Language processing and language learning are said to be served by the same neurobiologically based (declarative and procedural) memory systems that also serve memory for all other kinds of knowledge. Indeed, the declarative/procedural model is explicitly agnos-tic on the issue of modularity, because as Ullman concludes in Chapter 8 , at least provisionally on the basis of the available empirical evidence, “at this time there is no convincing evidence for domain-specific circuitry for language.” The recognition of important roles in language acquisition for both explicit and implicit learning modes and resulting explicit and implicit knowledge is also a shared position. However, the declarative/procedural model offers a much more detailed explana-tion for how implicit and explicit knowledge are served by different systems in the brain. Two predictions are particularly noteworthy for their empirical specific-ity. One prediction is that there will be differential relative involvement of both systems in language learning, with relative degree of reliance on the procedural memory system for younger ages and in contexts of naturalistic-immersive use, and on the declarative memory system for older ages and in contexts of instructed use. The second prediction is the posited existence of redundant, competitive, and inhibitory effects for the two memory systems on the resulting implicit and explicit knowledge of language. In the end, all three approaches (usage-based, Skill Acquisition, and declarative/procedural) allow for an interface position with regard to the explicit/implicit knowledge issue, but they posit different relation-ships and balances in that interface, with the declarative/procedural model offer-ing the greatest degree of specificity in this regard.

250 Ortega

Four other SLA theories in this book also hold a psychological view of cogni-tion, but they are much less specific about the assumptions they may make regard-ing the nature of language knowledge and the architecture of cognition. For one, they are ambiguous as to whether language cognition should be understood in psychological or linguistic terms. For example, Input Processing Theory assumes that Universal Grammar knowledge probably constrains learners’ hypotheses (see VanPatten, 1998), and both Processability Theory and the concept-oriented frame-work have explicitly tried to accommodate some version of linguistic nativism in their models, although they do so by drawing specifically on functional rather than formal linguistic constructs. In terms of the nature of knowledge representation, Processability Theory and the concept-oriented framework appear to side with implicit knowledge, and they remain silent as to the mechanisms that might make learning happen. On the other hand, Input Processing Theory and the interaction approach seem to give implicit knowledge a privileged place in acquisition while assuming an explicit/implicit knowledge and processing interface. However, nei-ther addresses directly the issue of how the interface may work.

Two theories, Vygotskian Sociocultural SLA Theory and Complexity Theory, stand out in that, in both, cognition is viewed as social and language is seen as emerging out of local, dynamic interactions. In both theories, the lines between environment and mind are blurred, in Sociocultural SLA Theory because the individual emerges from the social, which is the source for all learning, and in Complexity Theory because environment and agent are thought to continually interact and transform each other.

Sociocultural theorists posit that human cognition arises from the material, social, cultural, and historical contexts in which human experience is embedded. Learning (including language learning) is explained via mediation processes by which the mind appropriates and internalizes knowledge from the social world, whereby the dualism between individual mind and social environment is rejected: “the individual emerges from social interaction and as such is always fundamen-tally a social being” ( Chapter 11 ). Cognition, therefore, is fundamentally socio-cultural: It arises out of human relations to others, via cultural tools (including language) that mediate between us and our environment, and out of the specific events we experience. Indeed, in this theory the goal is to explain learning as a sociocultural accomplishment served by higher order cognition, whereby con-sciousness, agency, and intentionality are central to learning.

Complexity Theory intersects with Sociocultural Theory in its humanization of learners as conscious, intentional agents. But it also shares a good deal of usage-based tenets in that it precludes innate, built-in knowledge and attributes the emergence of linguistic knowledge to general cognitive mechanisms (e.g., analogy, statistical learning) that act on the socially gated ambient language. It is differ-ent from all other theories, however, in its emphasis on the orderly but dynamic interconnection among nested complex systems contributed by the learner and the environment, and in its concept of agentive self-adaptation and leveraging of

Second Language Learning Explained? 251

affordances by learners, who constantly and creatively co-adapt as they interact with other human, eco-social, and linguistic complex systems.

The Nature of Interlanguage

VanPatten and Williams ( Chapter 1 ) single out three agreed upon facts regard-ing interlanguage (cf. Table 13.1 ), in essence making systematicity and variability paramount in disciplinary understandings of interlanguage. How do the 10 SLA theories in this book address the systematicity and the variability observed in interlanguage? Figure 13.2 summarizes the range of positions on this issue.

Universal Grammar understands systematicity as a natural property of linguistic knowledge. Since all human languages are systematic, interlanguages must be too.

FIGURE 13.2 The nature of interlanguage in 10 SLA theories.

Theory Systematicity Variability

Universal Grammar Theory

Principles, parameters, features and functional categories

No theoretical status, only performance and processing effects

Usage-based approaches Systematicity, variability, and dynamicity expected in all complex systems

Skill Acquisition Theory

Taken for granted Experiential, cognitive, and developmental sources

Declarative/Procedural model

Not addressed Expected sources of variability: Age, context/input, linguistic phenomena, and individual differences in neuro-functioning

Input Processing Theory No theoretical emphasis on either

Processability Theory Two sides of the same phenomenon, both derived from functional constraints on processing and/or communication Concept-oriented

approach

Interaction framework Congruent with Processability and concept-oriented views

Congruent with Skill Acquisition views

Vygotskian Sociocultural theory

No theoretical emphasis

Central to activity and social cognition

Complexity Theory Focus on linguistic development via the study of longitudinal learner production but no theoretical emphasis on systematicity

Variability central to linguistic development. Inter-individual variability fully expected, which makes individual rather than group analysis a must. Intraindividual variability is precursor of all developmentally meaningful change.

252 Ortega

It is also natural to expect that there will be differences between the grammatical core of a language and everything else (pragmatics, vocabulary, and so on), since only certain properties of a language are thought to fall within the scope of the universal abstract linguistic knowledge that all humans share as a species. The the-ory leaves room for the possibility that certain variability is the result of a quality of indeterminacy that may be typical of L2 grammars (e.g., the fact that a learner may reject an ungrammatical string some times and accept it as grammatical other times). However, much of the variability is considered to be uninteresting theo-retically, on the grounds that it stems from simple shortcomings of performance, often due to processing effects that are thought to be unrelated to genuine L2 grammatical knowledge (e.g., the typical experience when a learner can remem-ber to use a rule while writing but forgets to use it in her speech).

Usage-based approaches take for granted that systematicity and variability are two properties of language just as they are of all complex, adaptive, emergent systems. In addition, they introduce a third construct, dynamicity (see Larsen-Freeman, 2002), to explain how systematicity is emergent and how systematicity and variability can co-exist and arise out of the brain’s interaction with its envi-ronment. By comparison, Skill Acquisition Theory focuses little on systematicity (or dynamicity, for that matter) and concentrates instead on explaining variability. Also noteworthy is that this theory finds the locus of variability in three sources that are external to the language system per se. A first source of variability is posited to be experiential. Namely, between-learner variability will arise from differing L2 experiences, as different learners are exposed to (or seek on their own to be exposed to) different amounts, qualities, and sequences of declarative knowl-edge and deliberate practice. A second source from which variability will arise is predicted to be psychological: Certain cognitive abilities differ greatly among people. (SLA research on this kind of variability is also known under the rubric of “individual differences”.) A third source of variability pertains to the same learner across contexts and conditions and can be considered cognitive-developmental. Namely, the same learner’s performance will vary depending on whether com-municative and cognitive stressors are present and overload the learner’s current performance capacities. That is, this kind of variability is an indication that per-formance has not yet become automatic in that particular area for that learner.

The declarative/procedural model does not directly investigate linguistic devel-opment. However, it predicts variability of L2 outcomes resulting from a number of conspiring factors that interact, including the timing of learning (younger ver-sus older), the context of L2 learning (naturalistic-immersive versus instructed) because of the concomitant differential quality and quantity of input each context implies, and the nature of the linguistic phenomena being learned (rule-based complex phenomena versus idiosyncratic and irregular phenomena).

Explaining interlanguage systematicity as well as variability is a major goal of both Processability Theory and the concept-oriented approach. Processability Theory, in particular, has been instrumental in establishing the basic findings for

Second Language Learning Explained? 253

developmental sequences in a number of L2 word order phenomena. This theory posits that learners are constrained (which in this theory is used in the sense of limited) in systematic ways by what grammatical information they can process syntactically at a given point in development. To process the L2 syntactically in this theory means to hold forms together in working memory for comparison and exchange of grammatical information. Variability is explained as the other side of the same coin, in that these same processing constraints will also determine the sets of alternatives (or variants) that are available to learners (their hypothesis space) at any given point in development. The concept-oriented approach, on the other hand, makes interesting but broader predictions about the interplay between systematicity and variability in interlanguage development, along the additive pro-gression from the realm of pragmatic resources (e.g., gesture, knowledge of the world and the context), onto lexical resources (e.g., adverbs as a means to convey time), and finally to the morphosyntactic or grammatical dimension (e.g., verbal morphology as a means to convey time). Existing nonlinguistic concepts and the need to express them linguistically pushes the system to arrive at increasingly more complex solutions along the pragmatic-lexical-grammatical cline. Input Process-ing Theory does not appear to make systematicity or variability central to its explanations, although it offers principles that are consistent with the prediction that interlanguage development will not be haphazard but systematic, in many instances predicted by the initial parsing strategies learners employ to comprehend the input.

The interaction approach makes broad use of the functionalist explanations for systematicity that the concept-oriented approach and Processability Theory have put forth, and is simultaneously interested in the second and third sources of individual variability posited in Skill Acquisition Theory. Proponents of the interaction approach find it theoretically important to understand the cognitive-developmental variability that is associated with communicative and cognitive stressors operating during communication (e.g., different interlocutors or interloc-utors of same versus different gender) and with requirements of task performance (e.g., a complex versus a simple task). The rationale is that such context and task factors might be manipulated externally to enhance processes during interaction that may eventually facilitate development. A second area of increasing theoretical importance in this approach is individual differences in cognitive resources. This focus is natural among interactionists, given that since the early 1990s they have viewed attention as a possible major explanatory construct for L2 learning, and given that it is well established in psychology that humans vary greatly in their attentional capacities.

In Sociocultural Theory, variability is a theoretically important phenomenon because actions and learning are thought to come about from situative engage-ment with others and out of affordances from specific contexts. Thus, it is thought that no universal cognitive abilities can be studied in disembodiment from the context and the people out of which they come about. The Zone of Proximal

254 Ortega

Development, the intentional goals that drive learners, and the type of mediation via tools (including language and others) available to them in any given event, all conspire to create variability. Moreover, this variability is theoretically interesting because it helps explains why some learners may acquire certain dimensions of language expertise but not others, and why some learners may be unsuccessful in their apparent efforts to learn the L2. Therefore, the design and method of what socioculturalists call the genetic method is favored in order to capture variability, as is the individualized measurement method of learner-tailored tests (Swain, 1993).

Finally, in the area of interlanguage as in most areas of examination, Complex-ity Theory (although placed at the bottom of Figure 13.2 ) occupies an interesting in-between position between usage-based approaches and Sociocultural Theory. The usage-based view of interlanguage development is by and large espoused by Complexity Theory as well. Indeed, these two theories make linguistic L2 development into a central disciplinary object of interest. For this reason, in both, but particularly in empirical work inspired by Complexity Theory, longitudinal designs and the quantitative-descriptive measurement of qualities of learner pro-duction are commonly found. However, an important difference is that emergent systematicity is emphasized by usage-based approaches, whereas in Complexity Theory, variability takes center stage in at least two ways. First, interindividual variability is considered normal and fully expected, because complex systems’ dynamic co-adaptation makes it theoretically untenable to expect that individual development will follow generalized group norms. Thus, interlanguage analyses focus on individual trajectories rather than groups. In the theorizing of inter-learner variability as a central phenomenon to be grappled with, Complexity Theory is most congruent with Sociocultural Theory. Second, periods of intense intralearner variability are thought to be precursors of developmentally meaning-ful change. Therefore, an extensive, theory-specific suite of nontraditional analyti-cal methods has been devised by the neighboring Theory of Dynamic Systems (Verspoor, de Bot, & Lowie, 2011), and Complexity Theory scholars make use of these new methods to capture, measure, and theoretically interpret intraindividual variability.

The Role of the First Language

The 10 theories presented in this book afford diverse roles to the L1 in their expla-nations of additional language learning, as depicted schematically in Figure 13.3 .

Three theories afford the L1 a privileged role in their explanations of SLA. Universal Grammar theory views the L1 as potentially the initial point of depar-ture for L2 acquisition. Indeed, some within this theory posit a large influence for the L1 in the early stages of L2 acquisition, although several other possibilities are also considered and empirically pursued (for an accessible explanation of the range of positions, see Mitchell, Myles, & Marsden, 2013, pp. 83–94). In the end, however, all Universal Grammar proponents agree that it is impossible to speak of

Second Language Learning Explained? 255

L1 influence as a wholesale phenomenon. As White puts it in Chapter 3 , in some areas the contribution of the L1 is fleeting, in others long-lasting, and yet in oth-ers it may be permanent. Since one important goal in this theory is to determine whether Universal Grammar knowledge still guides L2 acquisition in ways that are fundamentally similar to the ways in which it is posited to guide L1 acquisi-tion, it is imperative to tease out the relative contribution of L1 rules versus innate universal linguistic biases across core areas of linguistic knowledge. Thus, studies are set up to investigate groups of learners from carefully chosen L1 backgrounds and always by reference to baselines of monolingual native speakers of the L1 and L2 involved. Therefore, the L1 holds a privileged role in this theory not only in theoretical terms but also in terms of actual research practices.

Usage-based approaches also accord the L1 a privileged role in SLA, but they do so based on a different rationale from that of the Universal Grammar approach: “L1-tuned learned attention limits the amount of intake from L2 input, thus restricting the endstate of SLA” ( Chapter 5 ). In other words, as a result of early years of development, experience, and socialization, the brain’s neurons are tuned and committed to the L1, and any subsequent language learning (of a second, foreign, or heritage additional language beyond the first) is biased by this “learned atten-tion.” The framework posits that we humans are hard-pressed to change habits and routines that serve us well. It is as if with the flashlight of our L1 we were looking in the wrong L2 places for cues about what we are supposed to learn now. Certain cues will be frequent and salient, redundant, and meaningful enough that they will be attended to after sufficient L2 experience. More subtle features of the L2 input, however, may completely remain outside our flashlight beam, perhaps irreparably so.

FIGURE 13.3 The role of the first language in 10 SLA theories.

Theory Roles

Universal Grammar Theory Initial state? Central role in research designs L1-tuned learned attention that can bias/block L2 learning

Privileged role Usage-based approaches

Skill Acquisition Theory Taken for granted/not addressed

Declarative/Procedural model

Input Processing Theory UG/L1 ambivalence

Processability Theory Lesser influence

Concept-oriented approach

Interaction framework Taken for granted

Vygotskian Sociocultural Theory

Mediating role as cognitive tool for learning

Complexity Theory One of the resources that learners bring to the task, a complex system in and of itself

256 Ortega

The place of the L1 in the remainder theories of SLA is more modest. Input Processing Theory currently holds an ambivalent stance, since there is no theoreti-cal or empirical determination at this point in the development of the theory as to whether the strategies that learners employ to parse and comprehend the input ought to be considered L1-filtered or guided by linguistic universal knowledge (see Chapter 7 ). In Processability Theory and the concept-oriented framework, the L1 is thought to exert a lesser influence by comparison to robust functional and developmental universal forces. In Skill Acquisition Theory and the interac-tion approach, on the other hand, a selectively predictable influence is taken for granted, but without being crucial to any of the explanations proposed. Complex-ity theory agrees that the L1 is important because learners are not blank slates and come with prior knowledge of already known languages. Without making any specific predictions for how the L1 may influence development, it stipulates it is one of the resources that learners bring to the task, and itself a complex system. The declarative/procedural model does not make any predictions regarding roles that either declarative or procedural memory for L1 features might play when establishing and consolidating memories for L2 phenomena.

By contrast to all other theories in this book, in Vygotskian Sociocultural The-ory the L1 can take on a unique and positive role. The first language is a medi-ating tool, voluntarily used by learners to achieve self-regulation and to enable collaborative engagement in L2 learning events on occasions when using the L2 for higher level mental activity would be developmentally premature. In essence, the use of the L1 during L2 learning events is viewed not as a subconscious influ-ence that cannot be avoided, but as a strategy through which learners can achieve goals otherwise unavailable to them in the L2—for example, to discuss the L2 as an object of reflection, to understand an L2 grammar concept more deeply, or to clarify how to tackle a difficult L2 task. The L1, in these ways, can contribute to (rather than interfere with) L2 learning.

Contributions of the Linguistic Environment

What are the putative contributions to L2 learning of the linguistic environment to which learners are exposed and through which they interact with others? Each theory stipulates a different weight and role for input, input frequency, and output in explaining additional language learning. This is summarized in Figure 13.4 .

Let us first examine the roles each theory accords to L2 input. In Universal Grammar theory, input is thought to play only a limited, if necessary, part in acquisition: that of triggering values of knowledge that predate any experience with the linguistic environment. Once some relevant part of the linguistic input triggers, for example, our knowledge that the language we are learning is head-final (as in Japanese, where we say “the house to” instead of “to the house” or “I movies like” instead of “I like movies”), our preexisting knowledge should get reorganized in a domino effect, and a series of other knowledge pieces that

FIG

UR

E 13

.4 T

he r

ole

of t

he li

ngui

stic

env

iron

men

t in

10

SLA

the

orie

s.

Theo

ryIn

put

Freq

uenc

yO

utpu

t

Uni

vers

al G

ram

mar

The

ory

Trig

gers

ded

uctio

n of

kno

wle

dge

No

theo

retic

al s

tatu

s

Dec

lara

tive/

Proc

edur

al

mod

elD

iffer

ent

quan

titie

s an

d qu

aliti

es o

f in

put

are

expe

cted

in d

iffer

ent

cont

exts

ca

lling

for

diff

eren

tial r

elia

nce

on

decl

arat

ive

vs. p

roce

dura

l mem

ory

syst

ems

Rep

eate

d ex

posu

re m

ore

impo

rtan

t fo

r pr

oced

ural

mem

ory

and

singl

e ex

posu

re

fast

lear

ning

pos

sible

for

dec

lara

tive

mem

ory

No

theo

retic

al s

tatu

s

Usa

ge-b

ased

app

roac

hes

Ass

ocia

tive

lear

ning

is in

put-

driv

enU

tmos

t in

fluen

ce, s

ubco

nsci

ous

tally

ing

lead

s to

chu

nkin

g an

d sc

hem

atiz

atio

nC

onfin

ed r

oles

, ben

efic

ial o

nly

for.

. . .

. . .

sum

mon

ing

cons

ciou

snes

s

Skill

Acq

uisit

ion

The

ory

One

ingr

edie

nt o

nly,

nece

ssar

y bu

t no

t su

ffic

ient

Freq

uenc

y of

exp

osur

e/pr

actic

e is

impo

rtan

t in

aut

omat

izat

ion

. . .

deve

lopi

ng f

luen

cy

Inpu

t Pr

oces

sing

The

ory

How

lear

ners

pro

cess

es in

put

duri

ng

com

preh

ensio

n is

impo

rtan

tD

egre

e of

mea

ning

fuln

ess

mor

e im

port

ant

. . .

raisi

ng a

war

enes

s of

inpu

t m

isint

erpr

etat

ions

Proc

essa

bilit

y T

heor

yD

evel

opm

enta

l con

stra

ints

or

func

tiona

l pr

oces

sing

prin

cipl

es d

eter

min

e w

hat

can

get

proc

esse

d, w

hich

in t

urns

is

refle

cted

in p

rodu

ctio

n

Nat

ure

of t

he g

ram

mat

ical

info

rmat

ion

exch

ange

req

uire

d m

ore

impo

rtan

t. .

. on

ly a

ref

lect

ion

of w

hat

is pr

oces

sabl

e

Con

cept

-ori

ente

d ap

proa

chSu

bcon

scio

us t

ally

ing

of p

roto

typi

calit

y of

for

m–f

unct

ion

map

ping

s. .

. ge

nera

l pre

ssur

e to

co

mm

unic

ate

inte

nded

m

eani

ngs

driv

es a

cqui

sitio

n

Inte

ract

ion

fram

ewor

kO

ne in

gred

ient

onl

y, ne

cess

ary

but

not

suff

icie

ntIm

port

ant

but

in c

ombi

natio

n w

ith

othe

r fa

ctor

sC

ausa

l rol

e: c

an le

ad t

o de

velo

pmen

t

Vyg

otsk

ian

Soci

ocul

tura

l T

heor

yO

ne in

gred

ient

onl

yN

o th

eore

tical

sta

tus

Impo

rtan

t as

soc

ial p

artic

ipat

ion

(col

labo

rativ

e di

alog

ue,

lang

uagi

ng, p

riva

te s

peec

h)

Com

plex

ity T

heor

yA

ssoc

iativ

e le

arni

ng is

inpu

t dr

iven

, but

al

so d

rive

n by

lear

ner

perc

eptio

ns o

f af

ford

ance

s in

am

bien

t la

ngua

ge a

nd

agen

tive,

crea

tive

coad

apta

tion

Impo

rtan

t, bu

t ov

erri

dden

by

lear

ner

perc

eptio

ns a

nd a

gent

ive,

crea

tive

co-a

dapt

atio

n

Lang

uage

use

can

spu

r gr

owth

bu

t on

ly if

mea

ning

mak

ing

and

agen

tivity

are

eng

aged

258 Ortega

“cluster” together around the given value “head-final” should also get selected. Moreover, the input itself is thought to be underdetermined and impoverished by comparison to whatever L2 knowledge the language acquisition device ends up building. At the opposite extreme, usage-based approaches posit that language learning is input-driven. Every time constructions and exemplars in the linguistic input are experienced by the learner (through listening, reading, or both), neural connections are fired and strengthened, and memory traces are established until networks of associations emerge into a complex system. That is, of all ingredients of acquisition, input is posited to play the most central role in this framework.

Input is important for the declarative/procedural model, although only as a macrovariable seen as inherent to the two prototypical contexts for learning: nat-uralistic-immersive versus instructed. This theory predicts that exposure to input quantities and qualities typical of naturalistic-immersive contexts will engage pro-cedural memory system optimally, whereas input quantities and qualities typical of instructed contexts will lend themselves to optimal involvement of the declarative memory system.

In the remaining theories presented in this book, the linguistic input plays intermediate positions between the two extremes of minimal to maximal impor-tance. All of them maintain the need for exposure to L2 input, but each lends an increasingly larger role to other ingredients of the learning process.

In three theories—Input Processing Theory, the concept-oriented framework, and Processability Theory—it is how the learner processes the input, rather than the input per se, that is regarded as essential to explain acquisition. The specific theoretical details differ greatly among them, however. Input Processing Theory affords input a rather central role but, most importantly, exactly what part of it feeds into learning (i.e., becomes intake) is determined by comprehension, pro-cessing, and parsing strategies that the learner brings with her or him and through which the input is perceived. The concept-oriented framework predicts that two strategies brought to the task of input processing by learners, the one-to-one principle and the multifunctionality principle, figure prominently in shaping how learners are able or unable to use the L2 input for developing new resources for meaning-making during language production. Processability theory predicts that L2 learners’ limited capacity for what can be held momentarily in memory determines what abstract grammatical information in the input (as described in Lexical-Functional Grammar) can be held simultaneously and compared mentally in order to build a formal and meaningful representation of any utterance. Some-what ironically, despite the important role accorded to input in both the concept-oriented approach and Processability Theory, in the end both find their strongest evidence and make their most interesting predictions with regard to language production, not input: What gets processed or learned is best reflected in what can be generated in L2 production at a given time in development.

Skill Acquisition Theory, the interaction approach, and Sociocultural Theory go much further in construing the input as only one of several ingredients of SLA,

Second Language Learning Explained? 259

necessary but not sufficient, and perhaps not even the most crucial one. Thus, both skill learning and interactionist explanations lend more importance to other ingredients of the environment, such as explicit grammar explanation and delib-erate practice (in Skill Acquisition Theory), or interaction, feedback, and pushed output (in the interaction approach). In Sociocultural Theory, on the other hand, the input plays an important but general role for language learning insomuch as learners choose to engage with it actively, for example, through goal-oriented vicarious participation when they observe others using the language and through creative imitation of others’ utterances in private speech. However, social partici-pation in optimal learning events is thought to be more crucial for acquisition than the linguistic environment per se. These sociocultural views of the roles of the linguistic input resonate with the position espoused by Larsen-Freeman in her characterization of Complexity Theory in Chapter 12. Although with regard to the input, once again, Complexity Theory largely sides with the empirical pur-suit and interpretations made by usage-based approaches, Larsen-Freeman firmly rejects the computer metaphor of language learning implied in SLA discussions of input and output. In its place, she invokes two concepts: the “ambient language” and “affordances.” She argues that the input, better called ambient language, is not a static frame that simply provides learners with linguistic material for learning. While input is crucially important, it is not determining or defining of develop-mental trajectories. Instead, it is learner perceptions and creative co-adaptation to their ambient language that Larsen-Freeman views as critical for learning, for which the concept of affordances is needed, defined as “opportunities for action in the ecosocial environments (as perceived by learners) that can motivate agents to act and co-act.”

Naturally, since frequency is a feature of the linguistic input (referring in essence to the statistical properties of the input), only SLA theories that stipu-late a central role for input afford frequency a high explanatory power. All such theories, however, agree that the workings of frequency in L2 learning can only be understood as a force that affects acquisition in interacting with several others, rather than alone.

In usage-based approaches, the statistical properties of the input are of fore-most importance in explaining SLA. As humans process language input, they unconsciously compute the relative frequencies with which they encounter forms, constructions, and exemplars, the relative frequency of the surrounding linguistic contexts in which they appear, and the likelihood of the meanings they can refer to. Language knowledge gradually emerges in the learner by constantly tuning itself through every repeated experience to approximate the statistical properties of the experienced linguistic environment. However, different kinds of frequency are distinguished and their differential effects are investigated in dynamic interac-tion with other moderating factors, such as salience, prototypicality effects, and L1-tuned attention. To these factors that modulate frequency effects, Complexity theory adds that the learners’ own agentivity in selecting (or not) aspects of the

260 Ortega

ambient language as affordances overrides frequency. Once again, Complexity Theory essentially agrees with usage-based approaches but raises the same caveat for conceiving of roles for output in L2 learning as it does for theorizing input and frequency: Output may well be facilitative or even crucial, but far from its roles being autonomous, what is critical for learning is learner perceptions and adaptive productive use of language and their motivations as “agents to act and coact” in ways that involve meaningful language use. Learners are language users, that is, meaning makers by definition. When they use language productively, they do so agentively to “expand the meaning potential of a given language, not just to conform to a ready-made system.”

More than sheer frequency of occurrence of language forms in the ambient language, iteration, or the repetition of similar language usage events and mean-ings across changing tasks and activities, may be important for L2 development. Skill Acquisition Theory would approve of this notion of iteration, acknowl-edging as it does the importance of frequent exposure and repeated practice for enabling automatization and thus predicts that higher frequency forms in the input will be proceduralized earlier than rarely occurring ones. The declarative/procedural model posits different relative importance of frequency for the two different memory systems. Repeated exposure and thus high frequency is impor-tant in procedural memory, but fast learning out of single exposure is thought to be possible in declarative memory. Put differently, learning with involvement of the procedural memory system should be (gradual and) frequency-dependent, whereas learning with involvement of the declarative memory system should be (fast-paced and relatively) frequency-independent.

Work carried out within the concept-oriented framework also affords an important role to the statistical and distributional properties of the input in ways that are compatible with usage-based approaches. The relative frequency of form–function mappings in the input is predicted to influence the directions in which learners expand their linguistic repertoires. This is particularly well captured in the distributional bias hypothesis (e.g., Andersen, 1990), which posits that certain morphological markings (e.g., imperfective - ing ) are prototypically experienced in the input in combination with certain lexical meanings (e.g., actions that imply duration, such as run, walk, or sing ). The hypothesis predicts that this bias in the input will be reflected in learners’ development. For example, - ing may appear first in learner’s utterances containing activities like run (‘Look, a rabbit is running on the grass!’) and only later can spread to contexts containing accomplishments like run a marathon (‘Look, that man is running the marathon barefoot!’).

In Input Processing Theory and Processability Theory, objective frequency out there in the linguistic input is thought to explain less than other input features to which learner perception appears to orient, such as the semantic load or degree of meaningfulness (in Input Processing Theory) or the nature of the grammatical information exchange required (in Processability Theory). On the other hand, in the interaction approach, frequency is understood as only one of a number of

Second Language Learning Explained? 261

important influences, several outside the scope of input proper. In explanations proposed by Universal Grammar and Vygotskian Sociocultural Theory, frequency has no theoretical status.

Finally, let us review the place that L2 output holds across the theories. Most of the linguistically and psychologically oriented theories afford output a rather confined role. In Universal Grammar, learner output does not have a theoretical status, as it occupies no place in the explanations proposed or the evidence sought. On the cognitivist front, on the other hand, theories often construe output as beneficial for L2 learning, but without playing any major causal part in acquisi-tion processes. Thus, usage-based approaches hold that output is facilitative in promoting self-awareness and conscious processes that enhance learning or in fostering fluency and hence reinforcing chunking and automatization processes that normally happen tacitly during the processing of the input. Skill Acquisition Theory views output as deliberate practice, a special kind of language production activity in which explicit declarative knowledge about language is put to use, first effortfully and slowly, and later (with sufficient reiteration and deliberate effort) more fluently and accurately. Without being embedded in the right combina-tion and sequencing of explicit rule explanation and deliberate practice, however, output cannot contribute to learning. Output is afforded an even more limited role in Input Processing Theory, since comprehension (meaning extraction) is seen as the driving force in language learning. Output may promote fluency or at most allow for extra opportunities to realize that something in the input has been misinterpreted or misanalyzed (if, for example, a lack of understanding caused by such input miscue is revealed during interaction). Processability Theory, in turn, stipulates an even smaller role for output, in that it predicts that it simply mirrors development (hence the best evidence for acquisition can be gleaned from pro-duction data), but can never cause it or drive it. The declarative/procedural model does not make any theoretical predictions related to output.

Only 3 of the 10 SLA theories lend output some theoretical prominence. The concept-oriented framework explains language learning as driven by a general pressure to communicate intended meanings. Acquisition proceeds when the means available to the learner do not suffice for conveying the desired functions and concepts in their messages. Vygotskian Sociocultural Theory also affords an important role to output, albeit a more general one. This theory reconceptualizes output broadly as social participation and identifies an important learning poten-tial for the productive use of language through collaborative dialogue, languaging, and imitation in private speech.

The interaction approach is unique in that it accords to output the status of an acquisition catalyst, an acquisition-expanding force with interlanguage-stretching capabilites. When learners produce language for and with others, they can rely less on lexical and contextual cues (which often suffice during comprehension) and are forced to draw more on morphosyntactic cues. In addition, they may become more aware of gaps and holes in their linguistic resources, which may motivate them

262 Ortega

to look for solutions in the available input provided by others, either immediately upon experiencing difficulties or on a later occasion, when opportune timing and resources allow it. It is also through imperfect attempts to produce messages that mutual understanding may be obscured to the point that meaning needs to be nego-tiated (although of course, depending on the circumstances, mutual understanding could be faked or abandoned altogether!). If meaning is negotiated, language is often broken down into more manageable segments, new forms are offered, and implicit and explicit corrections are issued from well-intentioned (or alternatively, ill-predis-posed) interlocutors. In the interaction approach, output drives acquisition: We learn a language by speaking it, literally, and syntax emerges out of communication, a sug-gestion first made by Evelyn Hatch it in the beginnings of the field (Hatch, 1978).

The Role of Instruction

How does instruction interact with natural L2 learning processes? And what are the limits of what can be achieved, and what can be not, with instruction? Of the 10 SLA theories in this book, some make claims as to whether instruction is necessary or sufficient, beneficial or detrimental, whereas only a few go further to make specific proposals as to what features are needed for the design of optimal L2 instruction. The positions are presented schematically in Figure 13.5 .

In three theories, instruction plays no substantial role in L2 learning. Universal Grammar is consistent with this view. Much of a language may be teachable (e.g., vocabulary, stylistic choices, pragmatic preferences), but its morphosyntactic core (i.e., the facts of language that are most important for formal linguistic SLA) is not. The concept-oriented framework and Processability Theory also share the view that L2 instruction can play no large role, although the forces thought to overpower instructional influences are developmental-functional rather than for-mal. Given this view of instruction as peripheral, proponents within these three theories rarely pronounce themselves about the “how” of optimal instruction. However, in his work, Pienemann (see Lightbown & Pienemann, 1993; Piene-mann, 1984) has addressed the “what” of optimal instruction by proposing that instruction should target a level above learners’ current developmental stage, or else it can have negligible and possibly detrimental effects. In the view of Process-ability Theory, therefore, optimal instruction is instruction that is congruent with the current developmental level and readiness to learn of each individual learner.

In usage-based approaches, instruction can play a beneficial role, albeit one that is subordinated to input-driven, implicit statistical pattern induction. Instruction can be aimed at stimulating the dynamic interface between explicit and implicit learning and at destabilizing L1-tuned learned attention. While beneficial if it does so, however, it can never be considered “sufficient” for L2 learning. Propo-nents of this framework typically do not specify particular pedagogies, but they offer general principles for optimal instruction. One is that the L2 input that is brought into classrooms needs to be as abundant, rich, and authentic as possible.

FIG

UR

E 13

.5 T

he r

ole

of L

2 in

stru

ctio

n in

10

SLA

the

orie

s.

Theo

ry

Effe

cts

Opt

imal

feat

ures

In

struc

tiona

l des

ign

Uni

vers

al G

ram

mar

The

ory

No

effe

ct p

ossib

le o

n su

bcon

scio

us c

ore

know

ledg

e N

one

offe

red

Non

e of

fere

d

Usa

ge-B

ased

app

roac

hes

Ben

efic

ial b

ut n

ot s

uffic

ient

; sup

port

s co

nsci

ous

build

ing

of a

nalo

gies

and

cou

nter

s L1

-tun

ed

lear

ned

atte

ntio

n

Exp

licit

inst

ruct

ion

that

sum

mon

s co

nsci

ousn

ess

and

fost

ers

impl

icit

(bot

tom

up)

and

exp

licit

(top

do

wn)

lear

ning

inte

rfac

es, d

ynam

ic in

terf

ace

Non

e th

at is

the

ory-

spec

ific

Skill

Acq

uisit

ion

The

ory

Nec

essa

ry s

prin

gboa

rd a

nd s

tart

ing

poin

t pr

ior

to

auto

mat

izat

ion

Whe

n it

help

s ex

plic

it kn

owle

dge

to b

ecom

e pr

oced

ural

ized

C

ycle

s of

car

eful

ly s

eque

nced

ex

plan

atio

n an

d de

liber

ate

prac

tice

Dec

lara

tive/

Proc

edur

al

mod

el

Typ

ical

/exp

ecte

d ch

oice

for

adu

lts, i

t st

imul

ates

re

lianc

e on

exp

licit

lear

ning

at

the

expe

nse

of

impl

icit

lear

ning

Non

e of

fere

d N

one

offe

red

Inpu

t Pr

oces

sing

The

ory

Ben

efic

ial w

hen

it ta

rget

s un

prod

uctiv

e pr

oces

sing

stra

tegi

es t

hat

are

driv

en b

y co

mpr

ehen

ding

m

eani

ng a

nd b

ypas

s gr

amm

atic

al in

take

Com

preh

ensio

n ex

erci

ses

desig

ned

to s

hort

-ci

rcui

t un

prod

uctiv

e pa

rsin

g st

rate

gies

and

rep

lace

th

em w

ith p

rodu

ctiv

e on

es

Inpu

t Pr

oces

sing

Inst

ruct

ion

Proc

essa

bilit

y T

heor

y Li

mite

d ef

fect

s, ca

nnot

ove

rrid

e fu

nctio

nal-

univ

ersa

l for

ces

Con

sider

dev

elop

men

tal l

earn

er r

eadi

ness

whe

n ch

oosin

g ta

rget

s (t

he “

wha

t” o

f in

stru

ctio

n)

Non

e of

fere

d

Con

cept

-ori

ente

d ap

proa

ch

Non

e of

fere

d N

one

offe

red

Inte

ract

ion

fram

ewor

k B

enef

icia

l A

tten

tion

attr

acte

d to

lang

uage

for

m in

the

co

urse

of

mea

ning

ful t

ask

perf

orm

ance

Fo

cus

on f

orm

, tas

k-ba

sed

lang

uage

tea

chin

g

Vyg

otsk

ian

Soci

ocul

tura

l T

heor

y M

etal

ingu

istic

, met

acog

nitiv

e, ex

plic

it or

ient

atio

n: L

earn

ing

envi

ronm

ents

sho

uld

fost

er

mea

ning

ful e

vent

s an

d ot

her-

assis

tanc

e an

d de

ep

unde

rsta

ndin

g of

(gr

amm

ar)

conc

epts

med

iate

d by

visu

al a

rtif

acts

and

ver

baliz

atio

n or

lang

uagi

ng,

all a

ligne

d to

Zon

e of

Pro

xim

al D

evel

opm

ent

Syst

emic

-the

oret

ical

inst

ruct

ion

(aka

con

cept

-bas

ed in

stru

ctio

n)

Com

plex

ity T

heor

y B

enef

icia

l if

lear

ner-

cent

ered

Le

arne

r-ce

nter

ed, c

apita

lizin

g on

age

ntiv

ity,

crea

tivity

, mea

ning

mak

ing;

cul

tivat

e di

ffer

entia

ted

inst

ruct

ion

and

self-

refe

rent

ial g

oals;

att

end

to le

arne

r pe

rcep

tion

of a

ffor

danc

es; s

uppo

rt

lear

ners

’ aw

aren

ess

of d

iffer

ence

s; te

ach

adap

tatio

n by

iter

ativ

e la

ngua

ge u

se o

ver

chan

ging

act

iviti

es

Non

e th

at is

the

ory

spec

ific

264 Ortega

Another principle is to seed instruction with any kind of feedback (implicit or explicit) or other elements that help summon awareness, since conscious atten-tional control may help offset the effects of L1-learned attention. Grammar expla-nations can help, particularly when followed by strategically sequenced exemplars that make hidden patterns more salient to learners (see Ellis, 1993). Repetition and practice are thought to be beneficial, too.

In the declarative/procedural model, instruction is viewed as an unremarkably common choice for adults and one that stimulates reliance on explicit learn-ing, likely a good thing, although potentially at the expense of needed redundant implicit learning:

Explicit, classroom-like instruction of the grammar may encourage learn-ing in declarative memory, perhaps at the expense of learning in procedural memory. Conversely, the lack of explicit instruction, as often occurs in immersion contexts, may encourage learning in procedural memory. These predictions should hold for both L1 and L2 learners. ( Chapter 8 )

The remaining 5 theories of the 10 presented in this book take the position that instruction can optimize natural learning processes and may be even neces-sary when the goal is truly advanced levels of proficiency. Proponents in each have articulated full proposals for the design of optimal instruction and have addressed the “how” of instruction.

Proponents of Skill Acquisition Theory firmly believe that optimal instruc-tion should consist of cycles of explanation and deliberate practice of the various parts of language and language skills to be taught. Learners first need to be given explicit grammar explanations, often pedagogically simplified, and always accom-panied by good examples of the phenomenon being explained. This is because they must process this knowledge consciously, until they understand the rules well. This must be followed by carefully planned “deliberate practice” activities that enable learners to apply the rules they have newly committed to declarative memory to further examples and cases, first slowly and with high degrees of error, but gradually more fluently and accurately. Thus, learners who are found to make little progress in one (or all!) areas of the L2 may lack the relevant declarative knowledge or they may engage in insufficient, non-deliberate, or ill-sequenced practice for those areas. It is through the provision of relevant explanation and practice in the specific areas targeted by instruction that learners can be eventu-ally propelled to advanced levels of L2 competence. However, it is not grammar explanations alone or practice per se that fuels learning, but the fact that the two instructional elements are sequenced in specific ways and that the learners apply themselves in the conscious processing of knowledge and further practice the tar-get performances through deliberate and conscious efforts (see DeKeyser, 2007).

Input Processing theorists claim that optimal instruction should strive to alter how learners process the input during meaning-based comprehension. To this

Second Language Learning Explained? 265

end, they have developed a special type of instruction called Processing Instruction (see VanPatten, 2004), designed to afford high-quality opportunities to process certain aspects of the L2 input in the context of meaningful comprehension exer-cises, under conditions that short-circuit unproductive (L1 or universal) parsing tendencies. In this way, learners are primed to employ more appropriate parsing strategies in the target language. For example, L2 Spanish learners may be given practice with non-canonical (object-verb-subject) word order examples like “A Juan lo besó María” (“As for Juan, Maria kissed him”), immediately after being warned it would be misguided to assume that “Juan” is the doer of the kissing, just because their L1 English creates the expectation that the first noun in the string will most likely be the doer of any action. (It should be noted that explanations that draw explicit attention to L1-L2 parsing mismatches are thought to help, but they are not posited to be necessary.) With sufficient practice on how to parse strings using Spanish morphological cues rather than word order, learners’ internal parsing strategies are expected to attune themselves to the appropriate cues for extracting meaning and making Spanish-sensitive syntactic interpretations. (Far-ley, 2005, offers good suggestions for how to design a variety of input processing exercises across several target languages.)

In contrast to the sentence-level meaningful practice that is prioritized in peda-gogies based on Skill Acquisition Theory (via explicit instruction that provides declarative knowledge and opportunities for well-sequenced practice) and Input Processing Theory (via interventions that seek to affect implicit processing), the interaction approach favors task-based activities that afford learner practice with discourse-level language performance and subtly attract their attention to the spe-cific formal features that need to be learned. That is, this approach conceives of instruction as externally orchestrated opportunities to attend to relevant features of the target language in context, precisely when they are embedded unobtrusively in the task at hand, during meaningful comprehension and production activities, often in collaboration with peers or other interlocutors. A wide array of pedagogi-cal techniques are posited to be facilitative, ranging from most implicit (e.g., recasts) to most explicit (e.g., collaborative negotiation of language problems in group dictation exercises called dictogloss). Instruction is not expected to alter natural constraints and paths of development but to optimize them, and it is posited to be possibly necessary for the development of very advanced L2 capabilities. Currently debated within this approach is what counts as “unobtrusive” attentional manipu-lation and whether optimal instruction should seek changes in knowledge and processing at more explicit or more implicit levels (see Doughty & Williams, 1998).

Sociocultural Theory also construes instruction as clearly facilitative: “instruc-tional interventions [can be] designed to provoke development” ( Chapter 11 ). This theory specifies a type of educational design called Systemic Theoretical instruc-tion, which is unlike the instruction proposed by any other theory in that it decid-edly targets the metalinguistic, metacognitive, and explicit dimensions of learning. Instruction should promote rigorous understanding and internationalization

266 Ortega

of new concepts (language concepts, in the case of L2 learning) and should be designed to foster a social and material environment in which two additional things happen: (a) learners are encouraged to negotiate participation in meaning-ful activities by means of different kinds of mediation and (b) the quality of assis-tance from teacher and peers should be orchestrated so as to gauge the appropriate current level, aligned developmentally and contingent to individual learner needs. In this sense, appropriate L2 instruction should work within each learner’s Zone of Proximal Development and seek to expand it by enabling qualitative changes in the types of assistance and mediations called for. If these two conditions attain, learners can accomplish valued goals, first through assisted participation and later on their own, to the point that instruction will have helped individuals learn to use the L2 to self-regulate. Another unique aspect of instructional design in Sociocul-tural Theory is the “pedagogical imperative” (Lantolf & Poehner, 2014) to unite theory and praxis, that is, to pursue theoretical-conceptual understanding as well as meaningful transformation of learners’ material and symbolic worlds.

Finally, Complexity Theory might agree on many of the specifications for optimal instruction offered by usage-based approaches, but it resonates with Sociocultural Theory in adding the caveat that learners, and not any external instruction, are the agents of their own creative soft-assembling to the affordances they perceive in their environments, including the affordances they leverage from formal instruction. It can be surmised from this position that instruction will be good if it (a) capitalizes on learner agentivity and creativity, (b) considers meaning-making as central motivation for language, (c) addresses self-referential goals and is attuned to learners’ own perception of affordances, and (d) supports learners’ awareness of differences.

Some Future Challenges for SLA Theories

The 10 contemporary theories that readers find in this book (plus the early ones also reviewed by VanPatten and Williams in Chapter 2 ) are the most widely cited and discussed in the history of SLA to date. They attest to the three characteristics of the field mentioned earlier: youth, strength, and interdisciplinarity. In this final section, I forecast some areas that I believe will likely attract keen attention in future SLA work. In my opinion, future work that ventures in these directions has the best potential to improve disciplinary explanations about additional language learning.

First, I predict that in the future SLA theories will expend continued efforts in incorporating views of language cognition and L2 knowledge that are plausible, in the sense of compatible with cutting-edge knowledge about the workings of human cognition gleaned in the field of cognitive science. Some of the current theories in SLA fare better than others on this account. For example, usage-based approaches, Skill Acquisition Theory, and the declarative/procedural model offer fine-grained specifications of language cognition that draw on contemporary,

Second Language Learning Explained? 267

plausible models of cognition. Several alternatives to traditional Universal Gram-mar, such as Autonomous Induction Theory (Carroll, 2001; O’Grady, 2005), have originated precisely as attempts to accommodate cutting-edge knowledge of psy-cholinguistic processing into formal linguistic theories of language acquisition. However, in many SLA theories, a lack of specification of the assumed cognitive architecture is apparent. Indeed, this weakness was identified many years ago as a major obstacle for theory development, in a well known 1990 special issue of TESOL Quarterly devoted to exploring the proper scope and form of SLA theo-ries. In it, Schumann (1990) noted that claims like the “one-to-one principle” or the “noun-first principle” (p. 681) are useful in expressing observed, external behavior into predictive laws, but they are implausible direct descriptions of any underlying cognitive process or mechanism. Almost three decades later, the need for better specification of the cognitive architecture assumed in each SLA theory remains urgent.

Discussions regarding the contributions of the linguistic environment and instruction to L2 learning can be particularly muddled by the problem of under-specification regarding the nature of L2 knowledge that is posited in each theory. One illustrative case is the ongoing debate about recasts among proponents of the Interaction approach (see Chapter 10 ). It remains unclear in these exchanges whether the learning benefits of recasts that are under dispute (e.g., see exchange by Goo & Mackey, 2013; Lyster & Ranta, 2013) stem from metalinguistic (con-scious) or psycholinguistic (subconscious) levels of processing. If the latter case is to be assumed, valid evidence would have to come from two sources, online processing data and gains resulting from experimental manipulations. By contrast, a commitment to benefits that are metalinguistic and metacognitive in nature would demand that critical evidence be found in learner reports of awareness and documentation of incorporation of recasts in the immediate discourse. Without a theoretically guided agreement on what kinds of evidence can settle the debate, little progress, whether theoretical or empirical, can be made.

The psychologically oriented SLA theories presented in this book assume an explicit-implicit interface to some extent. Therefore, this area will likely attract much work in the future and stands to benefit greatly from interdisciplinary influ-ences from cognitive science. Several fundamental questions that need to be pur-sued are as follows: What constitutes explicit versus implicit knowledge of an L2? How does each type of knowledge originate? How and when do they interface with each other? What are the relative contributions of each to L2 learning? Inter-esting research has begun in recent years (e.g., see R. Ellis et al., 2009). However, in the future it will be important to investigate the nature and contributions to L2 learning of explicit and implicit knowledge from a wider range of theoretical SLA frameworks. The first challenge in this direction would be to specify appropri-ate empirical strategies for investigating the relative roles for implicit and explicit knowledge that can (or cannot) be postulated by a range of theoretical approaches beyond the ones currently engaged in this area. To be sure, each SLA theory

268 Ortega

will frame the questions regarding explicit and implicit knowledge differently, in ways that are congruent with the rest of the constructs and the view of cognition entailed in each. However, at this stage of our disciplinary knowledge it would be problematic for any theory of SLA to discard a priori one or the other type of knowledge as irrelevant for explaining L2 acquisition. For one, it is undeniable that L2 learners across formal and informal contexts encounter multiple oppor-tunities to learn from implicit, bottom up, and subconscious processing and they also seize opportunities for learning from explicit, top down, and conscious pro-cessing. In addition, all learners experience a “curious disjunction of knowledge” ( Chapter 2 ), so pervasive and striking that the phenomenon begs better theoretical understanding across all possible perspectives.

A second area that will hopefully attract future attention involves the need to theorize experience in explanations of SLA. As DeKeyser notes in Chapter 6 , different learners are afforded (or seek on their own to obtain) different amounts, qualities, and sequences of experience in and with the L2. Differential experience is thought to be connected to one of the most salient “facts” to be explained by any SLA theory, namely, the large variability and heterogeneity in L2 learning processes and outcomes. Of course, all theories of SLA acknowledge this fact and all admit that variable L2 outcomes are related in part to variable life experience for different learners across different contexts. However, most SLA theories are typically ill-equipped to deal with this reality in theoretically rigorous ways. As a consequence, they trivialize learner experience as anecdotal and outside the systematic scope of empirical documentation, divested as they see it from any theoretical status.

Among the 10 theories gathered in this book, there are two exceptions to this charge. Sociocultural Theory ( Chapter 11 ) is specifically designed to investigate cognition and learning as embedded in and taking its source from experience and context, not divested or abstracted from them. Complexity Theory ( Chap-ter 12 ) also shows clear leanings toward accounting for the social experiences of learners in a theoretic-central fashion. However, additional theories are available that have been adapted to SLA from the fields of anthropology, education, and sociology. Many of them have been termed “alternative” in the context of SLA, as the ones gathered in Atkinson (2011). Whether one considers them traditional or alternative, they draw on social understandings of cognition that are germane to the concerns in Sociocultural Theory ( Chapter 11 ) and Complexity Theory ( Chapter 12 ), and they hold great potential to help SLA researchers understand a range of social influences on L2 learning processes and outcomes, beyond the dimensions of cognition and language traditionally investigated in current SLA theories. The reason is that they have been designed specifically to deal with social experience as an object of study, rather than as random noise that needs to be eliminated from theory development. In other words, these theories were designed originally to theorize human social experience in their respective fields. In addition, they offer social respecifications of a number of constructs that are key

Second Language Learning Explained? 269

in SLA thinking, including grammar in systemic-functional linguistic theory (e.g., Schleppegrell, 2004), interaction in Conversation Analysis for SLA (e.g., Kasper & Wagner, 2011), and learning in Language Socialization Theory (e.g., Duff & Talmy, 2011) and Identity theory (Norton & McKinney, 2011).

Explanatory constructs that cut across this new family of SLA theories are agency, power, and identity. At the risk of simplifying grossly, just as certain SLA theories have helped us understand that the linguistic environment may be less important than how language usage events are processed and perceived by learn-ers, these other theories help us understand ways in which social experience may be important to understand, not as externally documented experience or fixed environmental encounters, but as experience that is lived, made sense of, negoti-ated, contested, and claimed by learners in their physical, inter-personal, social, cultural, and historical context. If in the future SLA researchers recognize the importance of theorizing learner experience, it may be possible to achieve a bal-ance between linguistic, cognitive, psychological, and social explanations in our theories. Encouraging in this regard are recent disciplinary discussions promoting epistemological bridges between cognitive and social orientations in the study of language learning and teaching (Hulstijn et al., 2014).

The final area for future theoretical development that I would like to forecast here pertains to the need for a complete reevaluation of SLA theories in light of what we know about bilingualism and the nature of bilingual competencies (Ortega, 2013, 2014). Ironically, the field of SLA takes as prototypical the idealized case of the individual who already possesses a mature monolingual grammar and subsequently begins learning an L2 with the goal to add on a monolingual-like command of the additional language. This bias is in part reminiscent of the same monolingual orientation in the field of child first language acquisition, which exerted a strong disciplinary influence on SLA during its formative years. Luckily, increasingly more empirical studies are available that illuminate the (very common case, worldwide) of bilingual first language acquisition from birth (see synthesis in De Houwer, 2009).

The burgeoning research in the neighboring field of child bilingual acquisition has exposed this monolingual bias and has left SLA theories vulnerable to serious critique on this count. In addition, Vivian Cook, one of the earliest SLA voices to raise these concerns (e.g., Cook, 1991), notes that the best psycholinguistic bilingual processing evidence tells us L2 competence is fundamentally different from the linguistic competence of a monolingual. To use Cook’s terminology, L2 users are not two monolinguals in one. Instead, they possess a psycholinguisti-cally distinct form of multicompetence. The validity of key notions in the field of SLA (e.g., interlanguage and ultimate attainment) is called into question when the monolingual native speaker is no longer held to be the norm. Consider, for exam-ple, how radically vacuous certain theoretical and empirical statements are, if we substitute the notion of “target-like” for the notion of “monolingual-like.” Thus, saying that “a given learner has failed to develop target-like competence in a given

270 Ortega

area of the L2” or that “most L2 learners fall short of the target norm” makes little sense, if what we think we are saying is that a given learner has failed to develop monolingual competence in a given area of the L2 or that most L2 learners fall short of the monolingual norm. After all, the impossibility for bilinguals to reach levels that are isomorphic with monolinguals would be a non-issue in a world in which bilingualism would be considered the default state of the human language faculty.

A bi/multilingual turn may be imminent in the near future. It would allow two fundamental reorientations in the field (Ortega, 2013). First, it would help recon-ceptualize L2 development as the development of late bilingualism, whose most natural comparison and counterpart is from-birth or very early-timed bilingual development. Second, it would make the field of SLA connect with the broader landscape of disciplines that study other kinds of language acquisition, all of which (together with L2 acquisition) vary along the two parameters of timing of learning and number of languages learned. If this sea change happens, I predict it will open up novel ways of framing the object of study and new premises from which to pose research questions, design studies, and interpret finding. A bilingual turn can thus only greatly enrich SLA and its future theory-building efforts.

Explaining how people learn languages later in life, above and beyond the mother tongue(s) they learn from birth, is the central task of the field of SLA. Scholarship in this area attracts keen interest and even fascination. This broad intellectual appeal is not surprising, given that speaking more than one language and being comfortable with more than one culture has become great personal and socio-economic assets to people from all walks of life and from all over the world. In close to 50 years of vibrant existence, SLA has produced a surprisingly varied and healthy number of theories that explain convincingly particular phe-nomena in L2 acquisition, sometimes in ways that are similar across theories, on occasion in ways that differ from (and sometimes even contradict) other theo-ries. Naturally, most progress has been made in the areas where most effort and attention has been directed to date, namely the acquisition of a linguistic system, traditionally defined as morphology and syntax of the L2. In the future, we can look forward to further theoretical development, innovation, and expansion. Future SLA thinking that continues to be interdisciplinary and reaps the ben-efits of advances in cognitive science, social theories, and bilingualism will be of essence in improving our explanations for the human capacity to learn additional languages later in life.

References

Andersen, R. W. (1990). Verbal virtuosity and speakers’ purposes. In H. Burmeister & P. L. Rounds (Eds.), Variability in second language acquisition: Proceedings of the 10th meeting of the Second Language Research Forum (Vol. 1, pp. 1–24). Eugene, OR: University of Oregon’s Department of Linguistics and American English Institute.

Atkinson, D. (Ed.). (2011). Alternative approaches in second language acquisition. New York, NY: Routledge.

Second Language Learning Explained? 271

Carroll, S. E. (2001). Input and evidence: The raw material of second language acquisition. Amster-dam, Netherlands: John Benjamins.

Cook, V. (1991). The poverty-of-the-stimulus argument and multicompetence. Second Lan-guage Research, 7, 103–117.

De Houwer, A. (2009). Bilingual first language acquisition. Bristol, England: Multilingual Matters.

DeKeyser, R. (Ed.). (2007). Practicing in a second language: Perspectives from applied linguistics and cognitive psychology. New York, NY: Cambridge University Press.

Doughty, C., & Williams, J. (1998). Pedagogical choices in focus on form. In C. Doughty & J. Williams (Eds.), Focus on form in classroom second language acquisition (pp. 197–261). New York, NY: Cambridge University Press.

Duff, P. A., & Talmy, S. (2011). Language socialization approaches to second language acquisition: Social, cultural, and linguistic development in additional languages. In D. Atkinson (Ed.), Alternative approaches to second language acquisition (pp. 95–116). New York, NY: Routledge.

Ellis, N. C. (1993). Rules and instances in foreign language learning: Interactions of implicit and explicit knowledge. European Journal of Cognitive Psychology, 5, 289–319.

Ellis, R., Loewen, S., Elder, C., Erlam, R., Philp, J., & Reinders, H. (2009). Implicit and explicit knowledge in second language learning, testing and teaching. Bristol, England: Multilingual Matters.

Farley, A. P. (2005). Structured input: Grammar instruction for the acquisition oriented classroom. Boston, MA: McGraw-Hill.

Goo, J., & Mackey, A. (2013). The case against the case against recasts. Studies in Second Language Acquisition, 35, 127–165.

Hatch, E. (1978). Discourse analysis and second language acquisition. In E. Hatch (Ed.), Second language acquisition: A book of readings (pp. 401–435). Rowley, MA: Newbury.

Hulstijn, J. H., Young, R. F., Ortega, L., Bigelow, M., DeKeyser, R., Ellis, N. C., . . . Talmy, S. (2014). Bridging the gap: Cognitive and social approaches to research in second lan-guage learning and teaching. Studies in Second Language Acquisition, 36 (3). doi:10.1017/S0272263114000035

Kasper, G., & Wagner, J. (2011). A conversation-analytic approach to second language acquisition. In D. Atkinson (Ed.), Alternative approaches to second language acquisition (pp. 117–142). New York, NY: Routledge.

Lantolf, J. P., & Poehner, M. E. (Eds.). (2014). Sociocultural theory and the pedagogical imperative in L2 education: Vygotskian praxis and the research/practice divide. New York, NY: Routledge.

Larsen-Freeman, D. (2002). Language acquisition and language use from a chaos/complex-ity theory perspective. In C. Kramsch (Ed.), Language acquisition and language socialization (pp. 33–46). New York, NY: Continuum.

Lightbown, P., & Pienemann, M. (1993). Comments on Stephen D. Krashen’s “teaching issues: Formal grammar instruction.” TESOL Quarterly, 27, 717–722.

Lyster, R., & Ranta, L. (2013). The case for variety in corrective feedback research. Studies in Second Language Acquisition, 35, 167–184.

Mitchell, R., Myles, F., & Marsden, E. (2013). Second language learning theories (3rd ed.). New York, NY: Routledge.

Norton, B., & McKinney, C. (2011). Identity and second language acquisition. In D. Atkin-son (Ed.), Alternative approaches to second language acquisition (pp. 73–94). New York, NY: Routledge.

O’Grady, W. (2005). Syntactic carpentry: An emergentist approach to syntax. Mahwah, NJ: Lawrence Erlbaum.

272 Ortega

Ortega, L. (2013). SLA for the 21st century: Disciplinary progress, transdisciplinary rel-evance, and the bi/multilingual turn. Language Learning, 63 (Suppl. 1), 1–24.

Ortega, L. (2014). Ways forward for a bi/multilingual turn in SLA. In S. May (Ed.), The multilingual turn: Implications for SLA, TESOL and bilingual education (pp. 32–53). New York, NY: Routledge.

Pienemann, M. (1984). Psychological constraints on the teachability of languages. Studies in Second Language Acquisition, 6, 186–214.

Schleppegrell, M. J. (2004). The language of schooling: A functional linguistics perspective. Mah-wah, NJ: Lawrence Erlbaum.

Schumann, J. (1990). Extending the scope of the acculturation/pidginization model to include cognition. TESOL Quarterly, 24, 667–684.

Swain, M. (1993). Second language testing and second language acquisition: Is there a con-flict with traditional psychometrics? Language Testing, 10, 193–207.

VanPatten, B. (1998). Cognitive characteristics of adult second language learners. In H. Byrnes (Ed.), Learning foreign and second languages (pp. 105–127). New York, NY: Mod-ern Language Association of America.

VanPatten, B. (Ed.). (2004). Processing instruction: Theory, research, and commentary. Mahwah, NJ: Lawrence Erlbaum.

Verspoor, M. H., de Bot, K., & Lowie, W. (Eds.). (2011). A dynamic approach to second lan-guage development: Methods and techniques. Amsterdam, Netherlands: John Benjamins.