The Dynamics of Second Language Emergence: Cycles of Language Use, Language Change, and Language Acquisition

NICK C. ELLIS English Language Institute University of Michigan 500 E. Washington Street Ann Arbor, MI 48104 Email: [email protected] This article outlines an emergentist account whereby the limited end-state typical of adult second language learners results from dynamic cycles of language use, language change, language perception, and language learning in the interactions of members of language communities. In summary, the major processes are: 1. Usage leads to change : High frequency use of grammatical functors causes their phonological erosion and homonymy. 2. Change affects perception: Phonologically reduced cues are hard to perceive. 3. Perception affects learning : Low salience cues are difcult to learn, as are homonymous/polysemous constructions because of the low contingency of their formfunction association. 4. Learning affects usage: (i) Where language is predominantly learned naturalistically by adults without any form focus, a typical result is a Basic Variety of interlanguage, low in grammatical complexity but communicatively effective. Because usage leads to change , maximum contact languages learned naturalistically can thus simplify and lose grammatical intricacies. Alternatively, (ii) where there are efforts promoting formal accuracy, the attractor state of the Basic Variety can be escaped by means of dialectic forces, socially recruited, involving the dynamics of learner consciousness, form-focused attention, and explicit learning. Such inuences promote language maintenance. Form, user, and use are inextricable. LANGUAGE IS A DYNAMIC SYSTEM. IT COMprises the ecological interactions of many players: people who want to communicate and a world to be talked about. It operates across many different agents (neurons, brains, and bodies; phonemes, morphemes, lexemes, constructions, interactions, and discourses), different human conglomerations (individuals, social groups, networks, and cultures), and different timescales (evolutionary, diachronic, epigenetic, ontogenetic, interactional, neurosynchronic). Cognition, consciousness, experience, embodiment, brain, self, communication and human interaction, society, culture, and history are all inextricably intertwined in rich, complex, and dynamic ways in language. Yet despite this complexity, despite its lack of overt government, instead of anarchy and chaos, there are patterns everywhere, patterns not preordained by God, by genes, by school curriculum, or by other human policy, but patterns that emergesynchronic patterns of linguistic organization at numerous levels (phonology, lexis, syntax, semantics, pragmatics, discourse, genre, etc.), dynamic patterns of usage, diachronic patterns of language change

The Modern Language Journal, 92, ii, (2008) 0026-7902/08/232249 $1.50/0 C 2008 The Modern Language Journal

Nick C. Ellis (linguistic cycles of grammaticization, pidginization, creolization, etc.), ontogenetic developmental patterns in child language acquisition, global geopolitical patterns of language growth and decline, dominance and loss, and so forth. As a complex system, the systematicities of language are emergent and adaptive. Only by adopting an integrative, dynamic framework will we understand how they come about. Dynamic systems theory (de Bot, Lowie, & Verspoor, 2007; Herdina & Jessner, 2002; Thelen & Smith, 1994), complex systems theory (Holland, 1998; Kauffman, 1995), connectionism (Elman, Bates, Johnson, Karmiloff-Smith, Parisi, & Plunkett, 1996), chaos-complexity theory (Cooper, 1999; Larsen-Freeman, 1997, 2002; Larsen-Freeman & Cameron, in press; Waldrop, 1992), and emergentism (Bates & MacWhinney, 1987; N. C. Ellis, 1998; N. C. Ellis & LarsenFreeman, 2006a; MacWhinney, 1999) are general frameworks for investigating processes of emergence of systematicity and pattern from dynamic interactions. The goal of this article is to outline a program for considering second language (L2) acquisition in such light. In 8,000 words, this must necessarily be a broad-brushed, impressionistic sketch, one that is overly simplistic and lacking in detail, and where controversy and issues of current argumentation are brushed over for the sake of a general picture. At its end, there will be many specic areas requiring closer study, completion, and correction, if the outline is seen to be useful at all, that is. Language learning and language use are dynamic processes in which regularities and systems arise from the interaction of people, brains, selves, societies, and cultures using languages in the world. Analyses of diachronic processes of grammaticization (Hopper & Traugott, 2003), complexication (Dahl, 2004), and language change (Bybee & Hopper, 2001; Croft, 2000; Mufwene, 2001), have always focused on dynamic usage. Usage-based theories have become increasingly inuential in the study of language acquisition (Barlow & Kemmer, 2000; Goldberg, 2006; Tomasello, 2003), centering on how children learn constructions while engaging in communication, the interpersonal communicative and cognitive processes that everywhere and always shape language (Slobin, 1997, p. 267). So too are input and interaction at the core of accounts of L2 learning (Gass, 1997; Larsen-Freeman & Long, 1991; Robinson & Ellis, 2008). Yet despite their similarities, rst language (L1) and L2 acquisition differ in signicant ways, most obviously on the level of ultimate attainment. Even the most dili-

233 gent L2 learner usually achieves a prociency considerably below what a child L1 acquirer achieves, with some naturalistic L2 acquirers acquiring only a Basic Variety characterized by pragmatic word order and minimal morphology (Klein & Purdue, 1992).1 Is there sense to be made of the limitations of L2 acquisition, and the structure of languages themselves, by considering the broader integrated dynamics of language change, acquisition, and use? One proposal is as follows. An emergentist account of second language acquisition (SLA) views the limited end-state typical of many naturalistic adult L2 learners as the result of dynamic cycles of language use, language change, language perception, and language learning in the interactions of members of language communities. The major processes include the following: 1. Usage leads to change : High frequency use of grammatical functors causes their phonological erosion and homonymy. 2. Change affects perception: Phonologically reduced cues are hard to perceive. 3. Perception affects learning : Low salience cues are difcult to learn, as are homonymous/polysemous constructions because of the low contingency of their formfunction association. 4. Learning affects usage: (i) Where language is predominantly learned naturalistically by adults without any form focus, a typical result is a Basic Variety of interlanguage, low in grammatical complexity but communicatively effective. Because usage leads to change , maximum contact languages learned naturalistically can thus simplify and lose grammatical intricacies. Alternatively, (ii) where there are efforts promoting formal accuracy, the attractor state of the Basic Variety can be escaped by means of dialectic forces, socially recruited, involving the dynamics of learner consciousness, form-focused attention, and explicit learning. Such inuences promote language maintenance. And so on, always, language shifts and changes, and cycles of usage evolve dynamically onward. This article outlines each of these processes in turn. USAGE LEADS TO CHANGE Languages change over time. They change as a result of usage. The diachronic study of language demonstrates characteristic processes of change that occur across languages, continents, and time, universal processes so general as to be recognized as laws. For example, in sound change, Grimms

234 law describes how Proto-Indo-European stop consonants developed into Proto-Germanic stops and other consonants, with voiceless stops changing into voiceless fricatives, voiced stops becoming voiceless, and voiced aspirated stops losing their aspiration and changing into plain voiced stops. Grimms law was a turning point in the development of linguistics. The realization of such regularities in the historical replacement of one phonetic feature by another was the foundation of the rigorous scientic study of language change. There are many such regularities in processes of phonological reduction (assimilation, lenition, elision, apocope, and syncope) and elsewhere in historical sound change. But their ubiquity and law-like systematicity have never prompted any theories of diachronic change in terms of their being rule-drivenwithout any conceivable mechanism, any such suggestion could be little more than a distraction. Rules do not govern these universals from above, instead such regularities are emergent (Bybee, 2000, 2005, 2007; Croft, 2000; Dahl, 2004; N. C. Ellis, 1998, 2003; N. C. Ellis & Larsen Freeman, 2006a; MacWhinney, 1998, 1999), evolving, bottom-up, from language as it is used. Like other patterns in natural and social science, they emerge in ecological and social interactions. The same is likely true for ontogenetic changefor language as it is acquired by children and adults from naturalistic usage (Robinson & Ellis, 2008; Tomasello, 2003). Usage Leads to Erosion Frequency is the driving force of phonological change. Its hard to keep languages from getting blurry: speakers tend to smudge phonology wherever possible, to delete and contract surface forms, and so forth. (Slobin, 1992, p. 191). The basic principles of automatization that apply to all kinds of motor activities and skills (like playing a musical instrument, a sport, or cooking) are that, through repetition, sequences of units that were previously independent come to be processed as a single unit or chunk (N. C. Ellis, 1996). Considerable practice with a particular token results in automaticity of its production and in sound reduction, assimilation, and lenition the loss and overlap of spoken gestures. On a variant of Hebbs (1949) learning rule later encapsulated in the paraphrase Cells that re together wire together, Bybees (2003) maxim is that Items that are used together fuse together (p. 112). The phenomenon is entirely graded the degree of reduction is a continuous function of the frequency of the target word and the condi-

The Modern Language Journal 92 (2008) tional probability of the target given the preceding word and following words (Bybee & Hopper, 2001; N. C. Ellis, 2002a; Jurafsky, Bell, Gregory, & Raymond, 2001). High token frequency also leads to autonomy whereby creative constructions learned by rote may never be analyzed into their constituent units, for example, learners may never have considered the literal roots of a dicey situation or that gimme consists of give + me. Such changes underpin grammaticalization in language change (Bybee, 2000; Croft, 2000). These separate processes of sound reduction themselves conspire in a more general law still, that of Zipf (1935), which describes the distributional properties of the lexicon whereby the frequency of a word decays as a (universal) power function of its rank. The more common words account for many more tokens of our language than do the less common ones (consider the at more than 60,000 tokens per million words, of at 29,000 tokens, and at 27,000 tokens, a at 22,000 tokens, etc.). The more frequent words are also the shorter ones in the language. Zipf (1949) summarized this in the principle of least effort speakers want to minimize articulatory effort and hence encourage brevity and phonological reduction. They tend to choose the most frequent words, and the more they use them, the more likely it will be that automatization of production will cause their shortening. Frequently used words become shorter with use. Erosion Leads to Homophony The most frequent words of the language tend also to be the most ambiguous (K ohler, 1986; Polikarpov, 2006). Many of the most frequently used words are ambiguous in their homophony and polysemy (e.g., to, too, two; there, their, theyre; I, eye, aye ). Ambiguity is a result of frequent usage, too: in polysemy from the grammaticalization processes of desemanticization (weakening, broadening, or abstraction of semantic force by habituation) and extension (use in new contexts; Hopper & Traugott, 2003); in homonymy from the fact that there are an awful number of meanings to hang onto a limited number of short sounds. This pattern generalizes across languages: The greater the number of monosyllabic words in the lexicon of a language, the greater the degree of homophony (Ke, 2006). Ambiguity is a loss of communicative capacity that arises if individual sounds are linked to more than one meaning as in homophony and polysemy. If the absence of word ambiguity is a mark of its evolutionary tness, then word formation provides an exponential

Nick C. Ellis increase in tness with length (Nowak, Komaraova, & Niyogi, 2002). Listeners also want to minimize their efforts in interpreting and determining what a word actually means. So, least effort for the speaker is in tension with least effort for the listener. As these forces play out in symbolic usage, languages evolve to have scaling properties that follow Zipfs (1935) law (Ferrer i Cancho & Sol e, 2003). Grammatical Functors Are the Most Frequently Used Forms The most frequently used words of the language are the closed-class words, the grammatical functors. The top 20 mostly frequently used words in English are the, of, and, a, in, to, it, is, to, was, I, for, that, you, he, be, with, on, by, and at (Leech, Rayson, & Wilson, 2001). More than half of spontaneous English speech consists of functors such as these. These are the little words of the language because of their frequency of usage; they are ambiguous in their interpretations; they are abstract and semantically light; they are syntagmatically highly connected. High frequency of use has caused their abbreviation over time. Indeed, eventually some grammatical markers wear away completely, creating a pressure for the development of others to replace them. Thus in French, negative statements were originally formed by the use of ne before the verb. For emphasis, ne often came to be reinforced by particles that once had been independent nouns (e.g., pas step, Je ne vais pas Im not going, Il ne marche pas hes not walking; goute drop, Je nai goute dargent I have no money, etc.). These particles underwent grammaticalization, with pas assuming special status as the default neutral obligatory negative adverb (e.g., Je ne pense pas I dont think so), though before the 12th century it was used with verbs of motion where its semantic connection is clear. In modern French, the ne is as often as not omitted entirely (e.g. Je suis pas all ee I did not go), its use dependent on sociolinguistic factors such as age, gender, style of speech, phonology, and clause type (Dewaele, 2004), again, the general pattern emergent from subpatterns of regularities of usage. A generation that grows up hearing a sound produced less distinctly than it had been pronounced for the previous generation gradually comes to take this lesser rendition as the default. In following the general tendency to pronounce unaccented sounds less distinctly, they, in turn, pronounce their default version of the sound, already less distinct than the last genera-

235 tions, even less distinctly. Eventually the default is no sound at all in that position. This erosion has a particularly dramatic effect in sounds such as sufxes or prexes that perform important grammatical functions. In this way, whereas Latin had different forms for all six combinations of person and number in the present tense, French has just three different forms for the present tense of er verbs (four for ir , re , and oir type verbs), and modern English has just two different forms (McWhorter, 2002). The ambiguity of the most frequent grammatical functors of a language and their lack of clear semantic reference is readily evident from the number of pages devoted to their explanation in dictionaries and grammars. Because of their range and frequency of use, these functors also have the highest connectivity or degree . When the sequential co-occurrence of words in discourse is described in terms of graphs of word connections, mapping the interactions like social networks, the World Wide Web, or other complex systems, these graphs show so-called small-world properties of being highly clustered and richly interconnected (Ferrer i Cancho & Sol e, 2001; Ferrer i Cancho, Sol e, & K ohler, 2004). Despite having many thousands of nodes (the >450,000 words populating a language), the average number of jumps in the path needed to get from any word to any other in this graph is remarkably small at less than three. A small number of highly connected words allows these properties. And it is the function words, the prepositions, pronouns, determiners, and so forth, that connect like this, having both high token frequency and a high degree of connectivity. The 10 most connected words of English are and, the, of, in, a, to, s, with, by, and is (Ferrer i Cancho & Sol e, 2001). The grammatical functors of a language, the closed-class words and morphological inections, are the milestones; they mark the structure of language. In language usage, they cue the combinatorial interpretation of words in phrases and sentences. In growth, they are the high degree nodes of the kernel lexicon of the language network to which new subunit constructions are preferentially attached, allowing scale-free growth distribution according to the so-called Barbar asi-Albert model (Barbar asi & Albert, 1999; Ferrer i Cancho & Sol e, 2001). Gathering the strands of this section together, we see too that grammatical functors are the most frequent tokens of the language. As a result of this high frequency of usage, they are the shortest, most phonologically reduced, least salient aspects of linguistic form. And, as a result of their

236 high frequency of usage and shortness, they are also the most ambiguous forms of the language they have many functions, they are abstract and semantically light, and they are often homonymous, too. CHANGE AFFECTS PERCEPTION In informal and rapid speech, the tendency to give short shrift to function words and bound morphemes, exploiting their frequency and predictability, deforms their phonetic structure and blurs the boundaries between these morphemes and the words that surround them. Clitics, accentless words, or particles that depend accentually on an adjacent accented word and form a prosodic unit together with it, are the extreme examples of this blurring: the /s/ of hes , /l/ of Ill , and /v/ of Ive can never be pronounced in isolation. Thus, grammatical function words and bound inections tend to be short and low in stress, even in speech that is produced slowly and deliberately (Bates & Goodman, 1997) and in speech directed at children (Goodman, Nusbaum, Lee, & Broihier, 1990), with the result that these cues are difcult to perceive. When grammatical function words (by, for, no, you, etc.) are clipped out of connected speech and presented in isolation at levels where their open-class equivalents (buy, four, know, ewe , etc.) are perceived 95% correctly, adult native speakers can recognize them only 50% of the time (Herron & Bates, 1997). Grammatical forms are of low salience. PERCEPTION AFFECTS LEARNING Low Salience Cues Are Poorly Learned The Rescorla-Wagner (1972) model, a formula summarizing the results of thousands of psychological investigations of animal and human associative learning, states that the amount of learning induced from an experience of a cue-outcome association depends crucially on the salience of the cue and the importance of the outcome. Lowsalience cues are poorly learned. They are also readily affected by overshadowing and blocking. Overshadowing describes how, when two cues are presented together and they jointly predict an outcome, the strength of conditioning to each cue depends on their salience, with the most salient cue becoming associated with the outcome and the less salient one being overshadowed so that on its own it evinces little or no reaction (Kamin, 1969). Blocking is a learned attention effect (N. C. Ellis, 2006c, p. 164) whereby a cue that is experienced in a compound along with a known strong

The Modern Language Journal 92 (2008) predictor is blocked from being seen as predictive of the outcome; once a cue has been blocked, further subsequent learning about that cue is attenuated/inhibited (Kamin, 1968; Kruschke & Blair, 2000; Mackintosh, 1975). In earlier work (N. C. Ellis, 2006b, 2006c), I discussed how these selective attention effects and the low salience of grammatical functors conspire in making their acquisition more difcult for adult L2 learners. The salience of these forms correlates 0.63 with their difculty of acquisition in the morpheme order studies (Goldschneider & DeKeyser, 2001). What of the other factor in the Rescorla-Wagner model, the importance of the outcome? Grammatical morphemes are often redundant and overshadowed by more salient lexical cues to tense or number (e.g., Tomorrow , Ill do the shopping; Yesterday I walked ; Seven boys ; Pica, 1983; Terrell, 1991). If a learner knows these lexical cues and has processed them, then subsequent processing of the morphological cues in these contexts affords no further information. Blocking affects L2 tense much more than L1 tense. Infants learn meanings at the same times as words, and children learning their L1 only acquire the meanings of temporal adverbs quite late in development. But adults, with their experience of the world and of their L1, know a variety of pragmatic and lexical means for expressing temporal reference (serialization: presenting events in their order of occurrence; adverbials, e.g., soon , now ; prepositional phrases, e.g., in the morning ; calendric reference, e.g., May 12, Monday, etc.; Schumann, 1987). Thus, adult language learners expression of temporality exhibits a sequence from pragmatic to lexical to grammatical devices, and the earlier other means block the acquisition of the later morphosyntactic ones: Whereas all learners apparently achieve the pragmatic and lexical stages of development, fewer learners achieve the morphological stage of development (BardoviHarlig, 2000, p. 415). Lexical and serialization strategies for expressing temporal reference are salient, constant, and simple to apply. Morphological cues to tense are nonsalient, they often vary by person and number, and typically there are additional irregularities. If, in expression, adult learners can get their message across by using these simpler strategies, they have achieved their goal. In the words of Simon (1962, p. 167), they have satisced rather than optimized, using the minimum necessary level of formal accuracy to achieve their communicative intention, whereas optimizing on native-like accuracy would be beyond their current cognitive bounds. Good enough (for the

Nick C. Ellis naturalistic world), but not perfect enough (for the more formal criteria of schooling). Homophonous (Low Form-Function Contingency) Forms Are Poorly Learned Learning the associations between cues and outcomes is a function of their contingency (Rescorla, 1968). The more reliably a cue predicts an outcome, the better the association is learned. The contingency as measured using, for example, P, the one-way dependency measure of the directional association between a cue and an outcome, predicts difculty of learning (Shanks, 1995) in a wide variety of human and animal learning. This relationship is at the core of connectionist (Chater & Manning, 2006; Christiansen & Chater, 2001) and competition (MacWhinney, 1987) models of language learning. Consider an English as a second language (ESL) learner trying to learn from the naturalistic input the interpretation of s at the ends of words. Plural s , third-person singular present s , and possessive -s , are all homophonous with each other as well as with the contracted allomorphs of copula and auxiliary be . Therefore, if we evaluate s as a cue for one particular of these functional interpretations, it is clear that there are many instances of the cue being there but that outcome not pertaining and thus P is accordingly low. Consider the mappings from the other direction as well: plural s , third-person singular present s , and possessive s all have variant expression as the allomorphs [/s/, /z/, /ez/]. Therefore, if we evaluate just one of these, say /ez/, as a cue for one particular outcome, say, plurality, it is clear that there are many instances of that outcome in the absence of the cue; P is concomitantly reduced. Thus, a contingency analysis of these cueinterpretation associations suggests that they will not be readily learnable (N. C. Ellis, 2006b, 2006c; Goldschneider & DeKeyser, 2001). This is just one particular illustration of the general case. The ambiguity of grammatical functors, their homophony and polysemy that result from high frequency of usage, erosion, desemanticization and extension, entail that they are lowcontingency constructions that are difcult to learn. Compounded Prejudices: Low Salience and Low Contingency These simple analyses have profound consequences. If, as Herron and Bates (1997) demonstrated, uent native speakers can only perceive

237 grammatical functors from the bottom-up evidence of input 50% of the time compared to open-class words, how can language learners hear them thence to learn their function? Fluent language users perceive these elements in continuous speech because their language knowledge provides top-down support. Knowledge of the constructional patterns of the language also helps them to disambiguate some interpretations from context. Although homophony abounds in language, homophonous forms rarely exist within the same grammatical class. Fluent speakers do not seem to have any difculty in disambiguating homophones from different syntactic categories; and as languages evolve, tness for usage entails that they do so to allow the possible interpretations of homophonic form to be limited usually to one by syntactic constraints (Lyon, Nehaniv, Bailey, & Warren, May 2004). But these top-down constraints are exactly the knowledge that learners lack. It is not surprising, therefore, that in L1 acquisition young children are unable to acquire grammatical forms until they have a critical mass of content words, providing enough top-down structure to permit perception and learning of those closed-class items that occur to the right or left of real words (Bates & Goodman, 1997, pp. 5152). Nor is it surprising that it is these elements that are difcult for L2 learners, with the order of acquisition of these morphemes being much the same among L2 as among L1 learners (Bailey, Madden, & Krashen, 1974; Brown, 1973; Dulay & Burt, 1973; LarsenFreeman, 1976). Grammatical L2 Acquisition Is Harder Still Fluent L1 users produce grammatical functors (because of their high frequency) that are smoothly automatized, eroded, and nonsalient. They also perceive these grammatical functors uently by expectation-driven means, supplementing the weak data-driven source and its low contingency with outcome by means of topdown support. Children nd grammatical functors difcult to perceive (low salience) and hard to learn (low salience plus low contingency). They have little language knowledge to provide top-down support in their perception because it is this language knowledge that they are trying to acquire. But they slowly accrete the strings, formulas, and more memorized forms (Bates & Goodman, 1997; N. C. Ellis, 1996; Peters, 1983; Peters & Menn, 1993) that provide the top-down support to allow the construction of language (Croft, 2001; Goldberg,

238 2006; Tomasello, 2003). They do so from plastic neural foundations that provide an optimal rational analysis of the problem space (N. C. Ellis, 2006b). The very things that make language use easy for a uent speaker (automaticity and top-down knowledge) make language acquisition hard (nonsalient cues, no top-down support). Adult L2 learners also nd grammatical functors difcult to acquire for all these reasons. But they are hampered, in addition, by learned attention to language. The language foundation of an adult learner is not a tabula rasa, but a tabula repleta (N. C. Ellis, 2006a, 2006c). Adult learners have various cognitive biases, as a result of their L1 experiences, which tune their attention away from these cues. LEARNING AFFECTS USAGE The consequences of the low salience and low learnability of grammatical functors by adult acquirers depends on their learning contexts, their motivations for learning, and the social and educational reactions to their language usage. Communicative bias in naturalistic contexts where language learning is predominantly implicit results in outcomes that differ from more explicit, formfocused interactions either in the classroom or in the feedback from accuracy-minded discourse partners. The extremes are so different that I will characterize them here as two possible parallel futures.2 This is a bifurcation point (Holland, 1998; Scott Kelso & Engstrm, 2006) in the dynamical system, and as such it is a point of leverage (Holland, 2006), where biasing one way or the other has the potential to switch or shift the future phase. Naturalistic L2 Acquisition Consider the following snippet of ESL, a classic piece from Lightbown and Spada (1999) that has introduced many students to the study of SLA. It is writing by an ESL French speaking secondary school pupil describing the cartoon lm The Great Toy Robbery: During a sunny day, a cowboy go in the desert with his horse. he has a big hat. His horse eat a our. In the same time, Santa Clause go in a city to give some surprises (pp. 74 75). It illustrates a classic ESL difcultythe omission of the third-person singular present tense s , that same s which was used previously for illustration of the low salience and low P of grammatical functors. Third-person present s and possessive s are the last acquired functors in the morpheme

The Modern Language Journal 92 (2008) order studies (Bailey, Madden, & Krashen, 1974; Goldschneider & DeKeyser, 2001). This is a particular example of the more general phenomenon that, although naturalistic L2 learners are surrounded by language, not all of it goes in, and L2 acquisition is typically much less successful than L1 acquisition. This is Corders (1967) distinction between input , the available target language, and intake , that subset of input that actually gets in and that the learner utilizes in some way. Schmidt (1984) described a naturalistic language learner, Wes, as very uent, with high levels of strategic competence, but low levels of grammatical accuracy: using 90% correct in obligatory contexts as the criterion for acquisition, none of the grammatical morphemes counted has changed from unacquired to acquired status over a ve year period (Schmidt, 1984, p. 5). The European Science Foundation crosslinguistic and longitudinal research project (Perdue, 1993) examined how 40 adult learners picked up the language of their social environment (Dutch, English, French, German, and Swedish) by means of everyday communication. Analysis of the interlanguage of these L2 learners resulted in its being described as the Basic Variety. All learners, independent of source language and target language, developed and used it, with about one third of them fossilizing at this level in that although they learned more words, they did not further complexify their utterances in respects of morphology or syntax. In this Basic Variety, most lexical items stem from the target language, but they are uninected.
There is no functional morphology. By far most lexical items correspond to nouns, verbs and adverbs; closed-class items, in particular determiners, subordinating elements, and prepositions, are rare, if present at all . . . . Note that there is no functional inection whatsoever: no tense, no aspect, no mood, no agreement, no casemarking, no gender assignment. (Klein, 1998, pp. 544545)

These grammatical functors abound in the input, but, as a result of their low salience, the low contingency of their form-function mappings, and adult acquirers learned attentional biases and L1-tuned automatized processing of language, they are simply not implicitly learned by many naturalistic learners whose attentional focus is on communication. Naturalisic L2 Use Causes Language Simplication . Linguistic evolution proceeds by natural selection from among the competing alternatives made available from the idiolects of individual speakers that vary among them (Croft, 2000;

Nick C. Ellis Mufwene, 2001). Given that adults are typically less successful than children at language learning, language use by a high proportion of adult language learners typically means simplication, most obviously manifested in a loss of redundancy and irregularity and an increase in transparency (Trudgill, 2002a, 2002b). The Basic Variety of interlanguage shows similarities with pidgins (Schumann, 1978) because pidgins are the languages that result from maximal contact and adult language learning (McWhorter, 2001). Veronique (1999, 2001) and Becker and Veenstra (2003) detailed many parallels between the grammatical structures of French-based creoles and the Basic Variety of interlanguage of learners of French as an L2, particularly in the 1:1 iconicity of their mapping of function and form (Andersen, 1984), their controller-rst, focus-last constituent ordering principles, their lack of verbal morphology, and the order of development of their means of temporal reference. Some creoles evolve as the complexication of pidgins resulting from the habitual use by children learning it as their L1. Other creoles, such as the Atlantic and Indian Ocean French-related creoles, developed from the interactions of adult speakers of nonstandard varieties of the target language and nonnatives (Mufwene, 2001). Creoles have systematic grammar, but not as many syntactic features as languages like West African Fula, which has 16 grammatical genders, or as many morphophonological features as the complex system of consonant mutations of Welsh, or as many phonological features as the tonal languages of South East Asia. All these are languages that have had much longer than the creoles to evolve their grammatical elaborations and diachronically motivated but synchronically obscure irregularities. Creoles typically have little or no inection, they have little or no tone to distinguish words or express grammar, and their prex/sufx + root combinations are semantically predictable (McWhorter, 2001, 2002). The older and more isolated a language, the more complexity it has, that is, the more it overtly signals distinctions beyond strict communicative necessity. The most elaborate languages in these respects are those older, more isolated languages that are spoken by groups of people whose interactions are primarily with other speakers of the language and that thus are learned as L1s by children whose plastic brains are ready to represent them optimally. But their linguistic complexities pose great difculties to L2 learners, who are prejudiced by L1 transfer and entrenchment. So some languages are easier for adults to learn, in an ab-

239 solute sense, than others: If one were given a month to learn a language of ones choice, I think one would select Norwegian rather than Faroese, Spanish rather than Latin, and Sranan rather than English (Trudgill, 1983, p. 106). It is no accident that Faroese, as a low-contact language not subject to adult language learning, has maintained a degree of inectional complexity that Norwegian has lost. Stasis allows a language, left to its own devices, to develop historical baggage linguistic overgrowths that, however interesting, are strictly incidental to the needs of human exchange and expression (McWhorter, 2001, 2002, 2004). In the same way that in nature, nichestability during the at periods of punctuated evolution allows the continuation of elaborate vestigial forms while competition selects them out, so in language, isolation allows the slow accretion of complexity and its maintenance, while large amounts of external contact and adult language learning select out the less functional linguistic overdevelopments. Consider again the case in point of the English third-person present tense s . It weaves through this stream like a yellow rubber duck, illuminating the ow of the English language and L2 acquisition wherever it bobs. English is no longer a language spoken primarily as an L1. The 375 million L1 speakers are in a very denite minority compared to the 750 million English as a foreign language (EFL) and 375 million ESL speakers (Graddol, 2000). This preponderance of adult language learning of English is changing the nature of the language. Seidlhofer (2004) described these changes as English is used across the world as a lingua franca. First and foremost on her list of observables is dropping the third person present tense -s (as in She look very sad) (p. 236). Languages are streamlined when history leads them to be learned more as second languages than as rst ones, which abbreviates some of the more difcult parts of their grammars (McWhorter, 2004, p. 51). As complex, adaptive systems, languages emerge, evolve, and change over time (N. C. Ellis & Larsen Freeman, 2006a, 2006b; Larsen-Freeman, 1997; Lee & Schumann, 2003, 2005; Schumann et al., 2006). Just as they are socially constructed, so too are they honed by social discourse. They adapt to their speakers. Because children are better language learners than adults, languages that adults can learn are simpler than languages that only children can learn. Thus the circle is unbroken. L2 acquisition by adults changes the very nature of language itself.

240 Form-Focused L2 Acquisition Socialinteractional or pedagogical reactions to nonnative-like utterances can serve as dialectic forces to pull L2 acquisition out of the attractor state of the Basic Variety. When an interaction partner or instructor intentionally brings additional evidence to the attention of the learner by some means of focus on form (Doughty & Williams, 1998), form-focused instruction, or consciousness raising, this focusing of attention can help the learner notice aspects of linguistic form or formfunction mapping (Schmidt, 2001). Terrell (1991) characterized explicit grammar instruction as the use of instructional strategies to draw the students attention to, or focus on, form or structure, or both, with instruction targeted at increasing the salience of inections and other commonly ignored features, rst by pointing them out and explaining their structure, and second, by providing meaningful input that contains many instances of the same grammatical meaningform relationship. Processing instruction (VanPatten, 1996) similarly aims to alter learners default processing strategies, to change the ways in which they attend to input data, thus to maximize the amount of intake of data to occur in L2 acquisition. L2 acquisition can thus be freed from the bounds of L1-induced selective attention by some means of focus on form that is socially provided (Lantolf & Thorne, 2006) and that recruits the learners explicit conscious processing. In this way, L2 acquisition involves learners in a conscious dialectic tension (Kramsch, 2002; Lantolf & Pavlenko, 1995; Lantolf & Thorne, 2006; Larsen-Freeman, 2002; Swain, 2000) between the conicting forces of their current interlanguage productions and the evidence of feedback, either linguistic, pragmatic, or metalinguistic, that allows socially scaffolded development. Form-focused instruction of this type results in more accurate L2 acquisition than does instruction without focus on form. Reviews of the experimental and quasi-experimental investigations into the effectiveness of explicit learning and L2 instruction (N. C. Ellis & Laporte, 1997; Spada, 1997), particularly the comprehensive meta-analysis of Norris and Ortega (2000) that summarized the ndings from 49 unique sample experimental and quasi-experimental investigations into the effectiveness of L2 instruction, demonstrated that form-focused L2 instruction results in large, target-oriented gains, that explicit types of instruction are more effective than implicit types, and that the effectiveness of L2 instruction is durable.

The Modern Language Journal 92 (2008) Consciousness and Learning. Form-focused instruction pulls learners out of their implicit habits, their automatized routines, by recruiting consciousness. Habits are implicitly controlled attractor states. We never think of walking, until it breaks down; as we start to stumble, then the feeling of falling is the negative evidence that recruits conscious control. We rarely think about driving, until it breaks down; as the clutch grinds, or the child runs into the road, these are the times when we become aware of the need to escape automatized routines. The more novelty we encounter, the more conscious involvement is needed for successful learning and problem-solving (Baars, 1997a, p. 303). So too for language: At each point in our history of language usage, the sample of language to which we have been exposed serves as the database from which we have induced our current model of how language operatesour modus operandi is based on estimates of the workings of the whole that we have determined from analysis of our sample of usage (N. C. Ellis, in press a). We operate according to these hypotheses until we receive negative evidence that we have erred in our analysis. Our consciousness is raised, and the tension between our implicitly controlled system and the evidence of overgeneralization to which we have been made aware serves as the interface allowing system change (N. C. Ellis, 2005). The last 10 years have seen signicant advances in our understanding of consciousness, its dynamics, and its roles in learning and memory. There have been three main developments to the scientic study of consciousness (Baars, Banks, & Newman, 2003): (a) cognitive neuroscientic investigation of the neural correlates of consciousness (see Koch, 2004, for review), (b) cognitive analysis of consciousness (particularly global workspace theory; see Baars, 1988, 1997b), and (c) computational modeling of the events underlying the emergence of self-amplifying resonances across a global network of neuronal coalitions, the dynamic competition among the massively parallel constituency of the unconscious mind that elects (Koch, 2004) the current oneness of the eeting stream of conscious experience (Dehaene & Changeux, 2004; Dehaene, Sergent, & Changeaux, 2003). The neural correlates of consciousness involve a coalition of forebrain neurons implicated in working memory and planning, interconnected via widespread corticocortico and corticothalamic feedback loops with sets of neurons in sensory and motor regions that code for particular features. Any one perceptreal or imagined corresponds to a winning coalition of the essential

Nick C. Ellis features coded by these different but related regions. Thus, for example, even though different areas of the brain code the heat, the liquidity, the aroma, the sweetness, and the color of a mouthful of coffee, activation in these nodes is simultaneously synchronized into a winning coalition that reinforces the ring activity of its member neuronsprobably by synchronizing their spiking dischargeand suppresses competing ones in a winner-takes-all fashion. Consciousness gives clout : When processes compete for ongoing control of the body, the one with the greatest clout dominates the scene until a process with even greater clout displaces it (Dennett, 2001). At any one moment, the winning coalition, expressed in the content of consciousness, is briey sustained for a discrete epoch of somewhere between 20 and 200 ms before it is replaced by another coalition in the ongoing stream of snapshots of consciousness. Stabilization of the coalition seems to be achieved by massive feedback known as reentrant signaling , perhaps involving thalamocortical loops, which is synchronized in rhythmic action potential discharge in the 3060 Hz gamma band of electroencephalography frequency. There is considerable ongoing research into this gamma band activity, both as an index of attentive awareness and as a mechanism for solving the binding problem (Crick & Koch, 2003; Dehaene & Changeux, 2004; Edelman, 1989; Edelman & Tononi, 2000; Koch, 2004; Singer, 1999). The dynamics of election to consciousness have profound implications for the nature of the lexicon and of grammatical categories as well. Lexical meaning is not static and compartmentalized as in a dictionary. Lexicography demonstrates that any instantiation of a words use represents an activation of only selected components of a words fuzzy meaning, largely indicated by the context (Hanks, 2000). Words do not have meaning; instead, they provide cues to meaning. They are stimuli that affect mental states. Their phonological, syntactic, and semantic properties are revealed by the effects they have on those states (Elman, 2004). The same is true for grammar. There are numerous places in grammar where it is necessary to recognize categories with a clear prototypical core but a somewhat fuzzily delimited periphery (Huddleston, 1984). A distributional analysis of nouns demonstrates that different nouns vary as to their morphosyntactic behavior, and that some nouns display more behavior characteristic of nouns than others (N. C. Ellis, 2002b). Knowledge of language is knowledge (Goldberg, 1995, p. 5), hence linguistic categories should be of

241 the same type as other categories in our linguistic system (Lakoff, 1987, p. 58). Elmans (1990) Serial Recurrent Network connectionist models of grammatical category activation shows how network state does not correspond to a word per se, but rather to the outcome of processing a word within a particular context. The contextual nature of interpretation suggests that it is better to look to the phrase or to the construction than to words as the basic level of language representation where form and meaning come together with greatest reliability (Chafe, 1994; Croft, 2001; N. C. Ellis, 1996, 2007b; Goldberg, 1995, 2003; Peters, 1983; Sinclair, 1991, 2004). Language, like the rest of cognition, is dynamical, situated, and embodied (Beer, 2000). What is elected to consciousness affects learning. Consciousness is the publicity organ of the brain. It is a facility for accessing, disseminating, and exchanging information and for exercising global coordination and control. Consciousness is the interface (N. C. Ellis, 2005). Paying attentionbecoming conscious of some materialseems to be the sovereign remedy for learning anything, applicable to many very different kinds of information. It is the universal solvent of the mind (Baars, 1997b, p. 301). So learning is dynamic; it takes place during processing, as Hebb (1949), Craik and Lockhart (1972), Pienemann (1998), and OGrady (2003) have all reminded us from their neural, cognitive, and linguistic perspectives. There are different forms of language learning. Broadly, these forms are (a) the implicit tallying and chunking that take place during usage (N. C. Ellis, 2002a, 2002b), and (b) the explicit learning in the classroom and that follows communication breakdown (N. C. Ellis, 2005). Implicit learning from usage occurs largely within modality and involves the priming, or chunking, of representations or routines within a module, with abstract schema and constructions emerging from the conspiracy of memorized instances. It is the means of tuning our zombie agents, the menagerie of specialized sensory-motor processors that carry out routine operations in the absence of direct conscious sensation or control. It is largely automatized. It operates in parallel. In contrast, conscious processing is spread wide over the brain and unies otherwise disparate areas in a synchronized focus of activity. Conscious activity affords much more scope for focused long-range association and inuence than does implicit learning. It brings about a whole new level of potential associations. It operates serially. Consciousness too is dynamic. It is perhaps the prototype example of an emergent phenomenon:

242 The stream of consciousness is one of ever-changing states, each cued by prior state and perceptual context, the units of consciousness being identiable as patterns of brain synchrony in time (Spivey, 2006). The dynamics of language learning are inextricably linked to the dynamics of consciousness, in neural activity and in the social world as well. Social Interaction and Dialectics. Language use and consciousness are both socially emergent, too. Language use, social roles, language learning, and conscious experience are all socially situated, negotiated, scaffolded, and guided. They emerge in the dynamic play of social intercourse. Our expectations, systematized and automatized by prior experience, provide the thesis, our model of language, and we speak accordingly. If we speak intelligibly and appropriately, we get one type of social reaction, and conversation focuses further on the intended message, meaning, and communication. If not, we may get another type of social reaction that helpfully focuses our attention on what we do not yet know how to do. Through the provision of negative feedback, be it a clarication request or possibly a recast, some dialectic (an antithesis that contradicts or negates our thesis, our model of language, and the tension between the two, which is resolved by means of synthesis) promotes the development of our language resources (Kramsch, 2002; Lantolf, 2006; Lantolf & Thorne, 2006). The usual socialinteractional or pedagogical reactions to non-native-like utterances involve an interaction partner (Gass & Varonis, 1994) or instructor (Doughty & Williams, 1998) intentionally bringing additional evidence to the attention of the learner by some clarication request, or negative feedback, or correction, or focus-on-form, or explicit instruction, recruiting consciousness to overcome the implicit routines that are nonoptimal for the L2. Analyses of classroom, motherchild, and native speaker nonnative speaker interactions demonstrate how conversation partners can scaffold the acquisition of novel vocabulary and other constructions by focusing attention on perceptual referents or shades of meaning and their corresponding linguistic forms (R. Ellis, 2000; Gass, 1997; Long, 1983; Mackey & Gass, 2006; Oliver, 1995; Tomasello, 1999; Tomasello & Akhtar, 2000). An interlocutor has various means of making the input more comprehensible: (a) by modifying speech, (b) by providing linguistic and extralinguistic context, (c) by orienting the communication to the here and now and, (d) by modifying the in-

The Modern Language Journal 92 (2008) teractional structure of the conversation (Long, 1982). Of course, learners are not passive recipients of modied input, but rather are agents of their own learning, playing an active role in negotiating meaning and selective attending (Larsen-Freeman, 1985). Interaction in which the participants attention is focused on resolving a communication problem, and the consequent negotiation of form and meaning connects input, internal learner capacities, particularly selective attention, and output in productive ways (Long, 1996, p. 452). Learning is ever thus. It takes place in a social context, involving action, reaction, collaborative interaction, intersubjectivity, and mutually assisted performance (Donato, 1994; Lantolf, 2006; Lantolf & Appel, 1994; Lantolf & Pavlenko, 1995; Lantolf & Thorne, 2006; Ricento, 1995; van Geert, 1994). Speech, speakers, and social relationships are inseparable (Norton, 1997). Activity theory emphasizes how individual learning is an emergent, holistic property of a dynamic system comprising many inuences, both social, individual, and contextual (Lantolf & Appel, 1994). Action provides a context within which the individual and society, mental functioning and sociocultural context can be understood as interrelated moments (Wertsch, 1998; Wertsch, Del Rio, & Alvarez, 1995). Uttering invokes feedback that is socially provided (Tarone, 1997) and that recruits the learners consciousness. Indeed consciousness itself is an emergent end product of socialization (Vygotsky, 1980; Wertsch, 1985). Just as L1 acquisition is socially gated (Kuhl, 2004), so too is SLA (N. C. Ellis, in press b). Cultural Forces: Written Language, Instruction, and Language Maintenance. Social reactions conspire in cultural dynamics. Setting a language into written form affects both language change and language learning. Language change occurs much faster in languages that are spoken but not written (only approximately 5% of the roughly 6,000 languages of the world are written; McWhorter, 2002). Written language affects L2 acquisition (Tarone & Bigelow, 2005). Educational policies and national curricula can focus on grammar to greater or lesser degrees. Language use is directly and indirectly forged with politics, government, nationhood, and identity. Language change can be challenged with prescriptive grammars; language loss can be reversed through language maintenance, revitalization, and reversal policies (Baker & Prys Jones, 1998; Fishman, 1991; Skutnabb-Kangas, Phillipson, & Rannut, 1994), and so on. In these ways, the dynamics of language

Nick C. Ellis learning are inextricably linked with the dynamics of the sociopolitical world as well. Language usage involves consciousness and learning and dialogue and dialectics, and it is motivated by cultural forces whether it occurs in naturalistic or formal contexts. Socially guided consciousness is the motivator for growth and change in all contexts and all cognitive domains. So why do I put these sections here in Parallel Future 2 where language structure is preserved or complexied, rather than in Parallel Future 1 where things tend toward simplication? Simply because in naturalistic environments these forces tend more toward meaning, whereas in formal environments, their bias is more toward form. EMERGENCE AND PANCHRONY There are many notable and interesting regularities of language structure, of L1 and L2 acquisition, of language production and language comprehension, and of language change. Despite considerable individual variation, where we are unable to predict precisely what a particular learner will say next, how he or she will phrase it, or what he or she will pick up from the social or pedagogical reactions to the utterance (de Bot, Lowie, & Verspoor, 2007; N. C. Ellis & Larsen Freeman, 2006b; R. Ellis, 1985, 1999; Larsen-Freeman, 2006; Tarone, 1988; Verspoor, Lowie, & van Dijk, this issue), or, in advance, how French, Spanish, Italian, Portuguese, and Romanian were to diverge from Latin, there are general patterns of interlanguage developmental sequences, of psycholinguistic processing, of native norms, of L1 transfer effects, of native-like selection, of different types of instruction, of language change, and of the rest across the wide range of phenomena of language. There is a lot of patterning to be explained (N. C. Ellis, 2007a; Long, 1990). Yet these regularities are not rule-driven; there are no mechanisms for such top-down governance. Instead, they emerge from the dynamics of language usage (Bybee, 2005, in press; N. C. Ellis, 2003; N. C. Ellis & Larsen-Freeman, 2006a; Elman et al., 1996; MacWhinney, 1998, 1999). Much comes from usage frequency (Bod, Hay, & Jannedy, 2003; Bybee & Hopper, 2001; N. C. Ellis, 2002a). This article has outlined effects of frequent production on phonological erosion, both ontogenetically and diachronically, and also on the language changes involved in grammaticalization (Hopper & Traugott, 2003). Words found together with a high frequency come to be cognitively processed as single chunks, which then evolve as individual words. As they are fre-

243 quently used, so four processes of change typically occur: desemanticization , the broadening or abstraction of meaning or content; extension , or use in new contexts; decategorialization , the loss of morphosyntactic properties; and erosion , the loss of phonetic substance. The result is a very 1971; Hodge, predictable linguistic cycle (Givon, 1970) whereby a notion that is rst expressed in discourse > syntax > morphology > morphophonemics > zero (> discourse.>.>, and round again). The mechanisms and principles involved in grammaticalization conform to a complex process of coding and organization of language which is universally applicable to describe the evolution of grammatical forms (Wischer & Diewald, 2002, p. 425). These universal processes emerge from dynamic processes of cognition and diachrony: For a theory of grammaticalization, it is both unjustied and impractical to maintain a distinction between synchrony and diachrony (Heine, Claudi, & Hunnemeyer, 1991, p. 248), and of usage, discourse, and social interaction:
Grammar is not absolutely formulated and abstractly represented, but always anchored in the specic form of an utterance . . . . Its forms are not xed templates, but are negotiable in face-to-face interaction in ways that reect individual speakers past experience of these forms, and their assessment of the present context. (Hopper, 1998, p. 142)

Grammar is always emergent and never present (Hopper, 1998, p. 148). So too are the regularities of language production and language reception, so too are the regularities of L2 acquisition. We cannot separate language from discourse, language from usage, language from cognition, language from society, or language from brain. Human thinking, like nature, appears to partition things, events, and ideas into pairs. But these pairs too are emergent, and they are complementary, more mutually dependent than mutually exclusive. They drive change, with the action taking place in between in complex coordination dynamics (Scott Kelso & Engstrm, 2006). From dynamic processes over all diachronic timescales and all synchronic states, there emerge what Saussure (1916) termed Panchronic principles (p. 135), generalizations of language that exist independently of time, of a given language, or of any concrete linguistic facts. There are indeed laws of L1 and L2 acquisition, comparable to the laws of the natural sciences. They are emergent too.
NOTES
1 This is not invariably the case, there are documented cases of adults acquiring nativelike skills (Ioup, Boustagoui, Tigi, & Moselle, 1994). Nevertheless, it is an

244
outcome typical enough to be regarded as an accepted fact of L2 acquisition (Long, 1990). 2 Of course, this characterization of two black-andwhite extremes (naturalistic acquisition as implicit leading to some kind of pidgin or basic variety vs. formal acquisition as conscious leading to a grammatical variety) is a gross simplication. Consciousness and social support play roles in naturalistic contexts because there is always learning in formal contexts that is implicit and data-driven. There are numerous gradations between these two extremes. Thanks to Kathleen Bardovi-Harlig for encouraging me to make this point explicit, and therefore, for encouraging the realization that every interaction is a point of possible bifurcation. Some causal factors combine additively, some are nonlinear in their combinations, and some are more catastrophic in their interactions causing marked qualitative or topological changes in the systems long-term dynamical behavior. The problem is knowing which, when, and why.

The Modern Language Journal 92 (2008)

Bates, E., & MacWhinney, B. (1987). Competition, variation, and language learning. In B. MacWhinney (Ed.), Mechanisms of language acquisition (pp. 157 193). Hillsdale, NJ: Erlbaum. Becker, A., & Veenstra, T. (2003). Creole prototypes as basic varieties and inectional morphology. In C. Dimroth & M. Starren (Eds.), Information structure and the dynamics of language acquisition (pp. 235 264). Amsterdam: John Benjamins. Beer, R. D. (2000). Dynamical approaches to cognitive science. Trends in Cognitive Sciences, 4, 9199. Bod, R., Hay, J., & Jannedy, S. (Eds.). (2003). Probabilistic linguistics . Cambridge, MA: MIT Press. Brown, R. (1973). A rst language: The early stages . Cambridge, MA: Harvard University Press. Bybee, J. (2000). Mechanisms of change in grammaticalization: The role of frequency. Unpublished manuscript. Bybee, J. (2003). Sequentiality as the basis of constituent & B. F. Malle (Eds.), The structure. In T. Givon evolution of language out of pre-language (pp. 109 132). Amsterdam: John Benjamins. Bybee, J. (2003). Mechanisms of change in grammaticalization: The role of frequency. In B. D. Joseph & R. D. Janda (Eds.), Handbook of historical linguistics (pp. 602623). Oxford: Blackwell. Bybee, J. (2005). From usage to grammar: The minds response to repetition . Paper presented at the Linguistic Society of America, Oakland, CA. Bybee, J. (2007). Usage-based grammar and second language acquisition. In P. Robinson & N. C. Ellis (Eds.), A handbook of cognitive linguistics and SLA (pp. 216236). New York and London: Routledge. Bybee, J., & Hopper, P. (Eds.). (2001). Frequency and the emergence of linguistic structure . Amsterdam: John Benjamins. Chafe, W. (1994). Discourse, consciousness, and time: The ow and displacement of conscious experience in speaking and writing . Chicago: University of Chicago Press. Chater, N., & Manning, C. (2006). Probabilistic models of language processing and acquisition. Trends in Cognitive Science, 10, 335344. Christiansen, M. H., & Chater, N. (Eds.). (2001). Connectionist psycholinguistics . Westport, CO: Ablex. Cooper, D. (1999). Linguistic attractors: The cognitive dynamics of language acquisition and change. Amsterdam: John Benjamins. Corder, S. P. (1967). The signicance of learners errors. International Review of Applied Linguistics, 5, 161 169. Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11, 671 684. Crick, F. C., & Koch, C. (2003). A framework for consciousness. Nature Neuroscience, 6, 119126. Croft, W. (2000). Explaining language change: An evolutionary approach . London: Longman. Croft, W. (2001). Radical construction grammar: Syntactic theory in typological perspective. Oxford: Oxford University Press.

REFERENCES Andersen, R. W. (1984). The one to one principle of interlanguage construction. Language Learning, 34, 7795. Baars, B. J. (1988). A cognitive theory of consciousness . Cambridge: Cambridge University Press. Baars, B. J. (1997a). In the theater of consciousness: The workspace of the mind . Oxford: Oxford University Press. Baars, B. J. (1997b). In the theatre of consciousness: Global workspace theory, a rigorous scientic theory of consciousness. Journal of Consciousness Studies, 4, 292309. Baars, B. J., Banks, W. P., & Newman, J. B. (Eds.). (2003). Essential sources in the scientic study of consciousness . Cambridge, MA: MIT Press/Bradford Books. Bailey, N., Madden, C., & Krashen, S. (1974). Is there a natural sequence in adult second language learning? Language Learning, 24, 235 243. Baker, C., & Prys Jones, S. (1998). Encyclopedia of bilingualism and bilingual education . Clevedon, UK: Multilingual Matters. Bakhtin, M. M. (1986). Speech genres and other late essays (V. W. McGee, Trans.). Austin, TX: University of Texas Press. Barbar asi, A.-L., & Albert, R. (1999). Emergence of scaling in random networks. Science, 286, 509511. Bardovi-Harlig, K. (2000). Tense and aspect in second language acquisition: Form, meaning, and use . Oxford: Blackwell. Barlow, M., & Kemmer, S. (Eds.). (2000). Usage based models of language . Stanford, CA: Center for the Study of Language and Information. Bates, E., & Goodman, J. C. (1997). On the inseparability of grammar and the lexicon: Evidence from acquisition, aphasia and real-time processing. Language and Cognitive Processes, 12, 507586.

Nick C. Ellis
Dahl, O. (2004). The growth and maintenance of linguistic complexity . Amsterdam/Philadelphia: John Benjamins. de Bot, K., Lowie, W., & Verspoor, M. (2007). A dynamic systems theory to second language acquisition. Bilingualism: Language and Cognition, 10, 721. Dehaene, S., & Changeux, J.-P. (2004). Neural mechanisms for access to consciousness. In M. Gazzaniga (Ed.), The cognitive neurosciences (3rd ed., pp. 11451158). Cambridge, MA: MIT Press. Dehaene, S., Sergent, C., & Changeaux, J.-P. (2003). A neuronal network model linking subjective reports and objective physiological data during conscious perception. Proceedings of the National Academy of Science USA, 100, 85208525. Dennett, D. C. (2001). Are we explaining consciousness yet? Cognition, 79, 222237. Dewaele, J.-M. (2004). Retention or omission of the ne in advanced French interlanguage: The variable effect of extralinguistic factors. Journal of Sociolinguistics, 8, 433450. Donato, R. (1994). Collective scaffolding in second language learning. In J. Lantolf & G. Appel (Eds.), Vygotskian approaches to second language research (pp. 3356). Norwood, NJ: Ablex. Doughty, C., & Williams, J. (Eds.). (1998). Focus on form in classroom second language acquisition . New York: Cambridge University Press. Dulay, H. C., & Burt, M. K. (1973). Should we teach children syntax? Language Learning, 23, 245 258. Edelman, G. M. (1989). The remembered present: A biological theory of consciousness . New York: Basic Books. Edelman, G. M., & Tononi, G. (2000). A universe of consciousness . New York: Basic Books. Ellis, N. C. (1996). Sequencing in SLA: Phonological memory, chunking, and points of order. Studies in Second Language Acquisition, 18, 91126. Ellis, N. C. (1998). Emergentism, connectionism and language learning. Language Learning, 48, 631 664. Ellis, N. C. (2002a). Frequency effects in language processing: A review with implications for theories of implicit and explicit language acquisition. Studies in Second Language Acquisition, 24, 143188. Ellis, N. C. (2002b). Reections on frequency effects in language processing. Studies in Second Language Acquisition, 24, 297339. Ellis, N. C. (2003). Constructions, chunking, and connectionism: The emergence of second language structure. In C. Doughty & M. H. Long (Eds.), Handbook of second language acquisition (pp. 33 68). Oxford: Blackwell. Ellis, N. C. (2005). At the interface: Dynamic interactions of explicit and implicit language knowledge. Studies in Second Language Acquisition, 27, 305 352. Ellis, N. C. (2006a). Cognitive perspectives on SLA: The associative cognitive CREED. AILA Review, 19, 100121.

245
Ellis, N. C. (2006b). Language acquisition as rational contingency learning. Applied Linguistics, 27, 1 24. Ellis, N. C. (2006c). Selective attention and transfer phenomena in SLA: Contingency, cue competition, salience, interference, overshadowing, blocking, and perceptual learning. Applied Linguistics, 27, 164194. Ellis, N. C. (2007a). Dynamic systems and SLA: The wood and the trees. Bilingualism: Language & Cognition, 10, 2325. Ellis, N. C. (2007b). Phraseology: The periphery and the heart of language. In S. Granger & F. Meunier (Eds.), Phraseology in foreign language learning and teaching (pp. 113). Amsterdam: John Benjamins. Ellis, N. C. (in press, a). Optimizing the input: Frequency and sampling in usage-based and form-focussed learning. In M. H. Long & C. Doughty (Eds.), Handbook of second and foreign language teaching . Oxford: Blackwell. Ellis, N. C. (in press, b). The psycholinguistics of the Interaction Hypothesis. In A. Mackey & C. Polio (Eds.), Multiple Perspectives on Interaction in SLA. Mahwah, NJ: Erlbaum. Ellis, N. C. & Laporte, N. (1997). Contexts of acquisition: Effects of formal instruction and naturalistic exposure on second language acquisition. In A. M. DeGroot & J. F. Kroll (Eds.), Tutorials in bilingualism: Psycholinguistic perspectives (pp. 5383). Mahwah, NJ: Erlbaum. Ellis, N. C., & Larsen-Freeman, D. (Eds.). (2006a). Language emergence: Implications for applied linguistics [Special issue]. Applied Linguistics, Introduction to the Special Issue, 27, 558559. Ellis, N. C., & Larsen-Freeman, D. (2006b). Language emergence: Implications for Applied Linguistics. Introduction to the Special Issue. Applied Linguistics, 27, 558589. Ellis, R. (1985). A variable competence model of second language acquisition. IRAL: International Review of Applied Linguistics in Language Teaching, 23 (1), 4759. Ellis, R. (1999). Item versus system learning: Explaining free variation. Applied Linguistics, 20, 460480. Ellis, R. (2000). Learning a second language through interaction . Amsterdam: John Benjamins. Elman, J. L. (1990). Finding structure in time. Cognitive Science, 14, 179211. Elman, J. L. (2004). An alternative view of the mental lexicon. Trends in Cognitive Science, 8, 301306. Elman, J. L., Bates, E. A., Johnson, M. H., KarmiloffSmith, A., Parisi, D., & Plunkett, K. (1996). Rethinking innateness: A connectionist perspective on development . Cambridge, MA: MIT Press. Ferrer i Cancho, R., & Sol e, R. V. (2001). The small world of human language. Proceedings of the Royal Society of London, B ., 268, 22612265. Ferrer i Cancho, R., & Sol e, R. V. (2003). Least effort and the origins of scaling in human language. Proceedings of the National Academy of Sciences, 100, 788791.

246
Ferrer i Cancho, R., Sol e, R. V., & K ohler, R. (2004). Patterns in syntactic dependency networks. Physical Review, E69, 05191510519158. Fishman, J. A. (1991). Reversing language shift . Clevedon, UK: Multilingual Matters. Gass, S. (1997). Input, interaction, and the development of second languages . Mahwah, NJ: Erlbaum. Gass, S., & Varonis, E. (1994). Input, interaction and second language production. Studies in Second Language Acquisition, 16, 283302. T. (1971). Historical syntax and synchronic morGivon, phology: An archeologists eld trip. Chicago Linguistic Society, 7, 394415. Goldberg, A. E. (1995). Constructions: A construction grammar approach to argument structure. Chicago: University of Chicago Press. Goldberg, A. E. (2003). Constructions: A new theoretical approach to language. Trends in Cognitive Science, 7, 219224. Goldberg, A. E. (2006). Constructions at work: The nature of generalization in language . Oxford: Oxford University Press. Goldschneider, J. M., & DeKeyser, R. (2001). Explaining the natural order of L2 morpheme acquisition in English: A meta-analysis of multiple determinants. Language Learning, 51, 150. Goodman, J. C., Nusbaum, H. C., Lee, L., & Broihier, K. (1990). The effects of syntactic and discourse variables on the segmental intelligibility of speech . Paper presented at the Proceedings of the 1990 International Conference on Spoken Language Processing, Kobe, Japan. Graddol, D. (2000). The future of English . London: The British Council. Hanks, P. (2000). Do word meanings exist? Computers and the Humanities, 34, 205215. Hebb, D. O. (1949). The organization of behaviour . New York: Wiley. Heine, B., Claudi, U., & Hunnemeyer, F. (1991). Grammaticalization: A conceptual framework . Chicago: University of Chicago Press. Herdina, P., & Jessner, U. (2002). A dynamic model of multilingualism: Perspectives of change in psycholinguistics . Clevedon, UK: Multilingual Matters. Herron, D., & Bates, E. (1997). Sentential and acoustic factors in the recognition of open- and closed-class words. Journal of Memory and Language, 37, 217 239. Hodge, C. (1970). The Linguistic Cycle. Linguistic Sciences, 13, 17. Holland, J. H. (1998). Emergence: From chaos to order . Oxford: Oxford University Press. Holland, J. H. (2006). Studying complex adaptive systems. Journal of Systems Science and Complexity, 19, 18. Hopper, P. J. (1998). Emergent grammar. In M. Tomasello (Ed.), The new psychology of language: Cognitive and functional approaches to language structure (pp. 155176). Mahwah, NJ: Erlbaum. Hopper, P. J., & Traugott, E. C. (2003). Grammaticalization (2nd ed.). Cambridge: Cambridge University Press.

The Modern Language Journal 92 (2008)

Huddleston, R. (1984). Introduction to the grammar of English . New York: Cambridge University Press. Ioup, G., Boustagoui, E., Tigi, M., & Moselle, M. (1994). Reexamining the critical period hypothesis: A case of successful adult SLA in a naturalistic environment. Studies in Second Language Acquisition, 16, 7398. Jurafsky, D., Bell, A., Gregory, M., & Raymond, W. D. (2001). Probabilistic relations between words: Evidence from reduction in lexical production. In J. Bybee & P. Hopper (Eds.), Frequency and the emergence of linguistic structure (pp. 229254). Amsterdam: John Benjamins. Kamin, L. J. (1968). Attention-like processes in classical conditioning. In M. R. Jones (Ed.), Miami symposium on the prediction of behavior: Aversive stimulation (pp. 931). Miami, FL: University of Miami Press. Kamin, L. J. (1969). Predictability, surprise, attention, and conditioning. In B. A. Campbell & R. M. Church (Eds.), Punishment and aversive behavior (pp. 276296). New York: AppletonCentury-Crofts. Kauffman, S. A. (1995). At home in the universe: The search for the laws of self-organization and complexity. Oxford: Oxford University Press. Ke, J.-Y. (2006). A cross-linguistic quantitative study of homophony. Journal of Quantitative Linguistics, 13, 129159. Klein, W. (1998). The contribution of second language acquisition research. Language Learning, 48, 527 550. Klein, W., & Perdue, C. (1992). Utterance structure: Developing grammars again . Amsterdam: John Benjamins. Koch, C. (2004). The quest for consciousness: A neurobiological approach . Englewood, CO: Roberts and Company. K ohler, R. (1986). Zur linguistischen Synergetik: Struktur und Dynamik der Lexik [On linguistic synergetics: Structure and dynamics of the lexis]. Bochum, Germany: Brockmeyer. Kramsch, C. (Ed.). (2002). Language acquisition and language socialization: Ecological perspectives . London: Continuum. Kruschke, J. K., & Blair, N. J. (2000). Blocking and backward blocking involve learned inattention. Psychonomic Bulletin & Review, 7, 636645. Kuhl, P. K. (2004). Is speech-learning gated by the social brain? Developmental Science, 10, 110120. Lakoff, G. (1987). Women, re, and dangerous things: What categories reveal about the mind . Chicago: University of Chicago Press. Lantolf, J. (2006). Sociocultural theory and L2: State of the art. Studies in Second Language Acquisition, 28, 67109. Lantolf, J., & Appel, G. (Eds.). (1994). Vygotskian approaches to second language research . Norwood NJ: Ablex. Lantolf, J., & Pavlenko, A. (1995). Sociocultural theory and second language acquisition. Annual Review of Applied Linguistics, 15, 3853.

Nick C. Ellis
Lantolf, J., & Thorne, S. (2006). Sociocultural theory and the genesis of second language development . Oxford: Oxford University Press. Larsen-Freeman, D. (1976). An explanation for the morpheme acquisition order of second language learners. Language Learning, 26, 125134. Larsen-Freeman, D. (1985). State of the art on input in second language acquisition. In S. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 433444). Rowley, MA: Newbury House. Larsen-Freeman, D. (1997). Chaos/complexity science and second language acquisition. Applied Linguistics, 18, 141165. Larsen-Freeman, D. (2002). Language acquisition and language use from a chaos/complexity theory perspective. In C. Kramsch (Ed.), Language acquisition and language socialization (pp. 3346). London: Continuum. Larsen-Freeman, D. (2006). The emergence of complexity, uency, and accuracy in the oral and written production of ve Chinese learners of English. Applied Linguistics, 27, 590619. Larsen-Freeman, D., & Cameron, L. (2008). Complex systems and applied linguistics . Oxford: Oxford University Press. Larsen-Freeman, D., & Long, M. (1991). An introduction to second language acquisition research . New York: Longman. Lee, N., & Schumann, J. (2003). The evolution of language and of the symbolosphere as complex, adaptive systems . Paper presented at the American Association for Applied Linguistics, Arlington, VA. Lee, N., & Schumann, J. (2005). The interactional instinct: The evolution and acquisition of language. Paper presented at the Congress of the International Association for Applied Linguistics, Madison, WI. Leech, G., Rayson, P., & Wilson, A. (2001). Word frequencies in written and spoken English: Based on the British national corpus . London: Longman. Lightbown, P. M., & Spada, N. (1999). How languages are learned . Oxford: Oxford University Press. Long, M. H. (1982). Native speaker/non-native speaker conversation in the second language classroom. In M. Long & J. Richards (Eds.), Methodology in TESOL: A book of readings (pp. 339354). New York: Newbury House. Long, M. H. (1983). Linguistic and conversational adjustments to non-native speakers. Studies in Second Language Acquisition, 5, 177193. Long, M. H. (1990). The least a second language acquisition theory needs to explain. TESOL Quarterly, 24, 649666. Long, M. H. (1996). The role of linguistic environment in second language acquisition. In W. Ritchie & T. Bhatia (Eds.), Handbook of second language acquisition (pp. 413468). New York: Academic. Lyon, C., Nehaniv, C., Bailey, J., & Warren, S. (2004). Evolutionary tness, homophony and disambiguation through sequential processes . Paper presented at the 1st International Workshop on Emergence and Evolution of Linguistic Communication, Kanazawa, Japan.

247
Mackey, A., & Gass, S. (2006). Introduction to the special issue on interaction research. Studies in Second Language Acquisition, 28, 169178. Mackintosh, N. J. (1975). A theory of attention: Variations in the associability of stimuli with reinforcement. Psychological Review, 82, 276298. MacWhinney, B. (1987). The competition model. In B. MacWhinney (Ed.), Mechanisms of language acquisition (pp. 249308). Hillsdale, NJ: Erlbaum. MacWhinney, B. (1998). Models of the emergence of language. Annual Review of Psychology, 49, 199 227. MacWhinney, B. (Ed.). (1999). The emergence of language . Hillsdale, NJ: Erlbaum. McWhorter, J. (2001). The worlds simplest grammars are creole grammars. Language Typology, 5, 125 166. McWhorter, J. (2002). The power of Babel: A natural history of language . San Franciso: W. H. Freeman. McWhorter, J. (2004). The story of human language: Lecture 24: Language Interrupted. On: The great courses : The Teaching Company Limited Partnership. Mufwene, S. S. (2001). The ecology of language evolution . Cambridge: Cambridge University Press. Norris, J., & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis and quantitative meta-analysis. Language Learning, 50, 417528. Norton, B. (1997). Language, identity, and the ownership of English. TESOL Quarterly, 31, 409430. Nowak, M. A., Komaraova, N. L., & Niyogi, P. (2002). Computational and evolutionary aspects of language. Nature, 417, 611617. OGrady, W. (2003). The radical middle: Nativism without universal grammar. In C. Doughty & M. H. Long (Eds.), Handbook of second language acquisition (pp. 4362). Oxford, UK: Blackwell. Oliver, R. (1995). Negative feedback in child NS/NNS conversation. Studies in Second Language Acquisition, 18, 459481. Perdue, C. (Ed.). (1993). Adult language acquisition: Crosslinguistic perspectives . Cambridge: Cambridge University Press. Peters, A. M. (1983). The units of language acquisition . New York: Cambridge University Press. Peters, A. M., & Menn, L. (1993). False starts and ller syllables: Ways to learn grammatical morphemes. Language, 69, 742777. Pica, T. (1983). Adult acquisition of English as a second language under different conditions of exposure. Language Learning, 33, 465497. Pienemann, M. (1998). Language processing and second language development: Processability theory. Amsterdam: John Benjamins. Polikarpov, A. A. (2006). Cognitive model of lexical system evolution and its verication . Retrieved 15 May, 2006, from http://www.philol.msu.ru/lex/ articles/cogn_ev.htm Rescorla, R. A. (1968). Probability of shock in the presence and absence of CS in fear conditioning. Journal of Comparative and Physiological Psychology, 66, 15.

248
Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and nonreinforcement. In A. H. Black & W. F. Prokasy (Eds.), Classical conditioning II: Current theory and research (pp. 6499). New York: Appleton-Century-Crofts. Ricento, T. (1995). Considerations of identity in L2 learning. In E. Hinkel (Ed.), Handbook of research on second language teaching and learning (pp. 895 910). Mahwah, NJ: Erlbaum. Robinson, P., & Ellis, N. C. (Eds.). (2008). A handbook of cognitive linguistics and SLA. London: Routledge. Saussure, F. d. (1916). Cours de linguistique g en erale [Course in General Linguistics] (R. Harris, Trans.). London: Duckworth. Schmidt, R. (1984). The strengths and limitations of acquisition: A case study of an untutored language learner. Language, Learning, and Communication, 3, 116. Schmidt, R. (2001). Attention. In P. Robinson (Ed.), Cognition and second language instruction (pp. 3 32). Cambridge: Cambridge University Press. Schumann, J. H. (1978). The pidginisation process: A model for second language acquisition . Rowley, MA: Newbury House. Schumann, J. H. (1987). The expression of temporality in basilang speech. Studies in Second Language Acquisition, 9, 2141. Schumann, J. H., Favareau, D., Goodwin, C., Lee, N., Mikesell, L., Tao, H., et al. (2006). Language evolution: What evolved? Marges Linguistiques, 11, 167 199. Scott Kelso, J. A., & Engstrm, D. A. (2006). The complementary nature . Boston: MIT Press. Seidlhofer, B. (2004). Research perspectives on teaching English as a lingua franca. Annual Review of Applied Linguistics, 24, 209239. Shanks, D. R. (1995). The psychology of associative learning . New York: Cambridge University Press. Simon, H. A. (1962). The sciences of the articial . Cambridge, MA: MIT Press. Sinclair, J. (1991). Corpus, concordance, collocation . Oxford: Oxford University Press. Sinclair, J. (2004). Trust the text: Language, corpus and discourse . London: Routledge. Singer, W. (1999). Neural synchrony: A versatile code for the denition of relations? Neuron, 24, 49 65. Skutnabb-Kangas, T., Phillipson, R., & Rannut, M. (1994). Linguistic human rights . New York: Mouton de Gruyter. Slobin, D. I. (1992). Psycholinguistics (2nd ed.). Glenview, IL: Scott, Foresman and Company. Slobin, D. I. (1997). The origins of grammaticizable notions: Beyond the individual mind. In D. I. Slobin (Ed.), The crosslinguistic study of language acquisition (Vol. 5, pp. 265323). Mahwah, NJ: Erlbaum. Spada, N. (1997). Form-focused instruction and second language acquisition: A review of classroom and laboratory research. Language Teaching Research, 30, 7387.

The Modern Language Journal 92 (2008)

Spivey, M. (2006). The continuity of mind . Oxford: Oxford University Press. Swain, M. (2000). The output hypothesis and beyond: Mediating acquisition through collaborative dialogue. In J. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 97114). Oxford: Oxford University Press. Tarone, E. (1988). Variation in interlanguage . London: Edward Arnold. Tarone, E. (1997). Analyzing IL in natural settings: A sociolinguistic perspective of second-language acquisition. Communication and Cognition, 30, 137 150. Tarone, E., & Bigelow, M. (2005). Impact of literacy on oral language processing: Implications for SLA research. Annual Review of Applied Linguistics, 25, 7797. Terrell, T. (1991). The role of grammar instruction in a communicative approach. Modern Language Journal, 75, 5263. Thelen, E., & Smith, L. B. (1994). A dynamic systems approach to the development of cognition and action . Cambridge MA: MIT Press. Tomasello, M. (1999). The cultural origins of human cognition . Boston: Harvard University Press. Tomasello, M. (2003). Constructing a language . Boston: Harvard University Press. Tomasello, M., & Akhtar, N. (2000). Five questions for any theory of word learning. In R. Golinkoff (Ed.), Becoming a word learner: A debate on lexical acquisition (pp. 179186). Oxford: Oxford University Press. Trudgill, P. (1983). On dialect: Social and geographical perspectives . Oxford: Basil Blackwell. Trudgill, P. (2002a). Linguistic and social typology. In J. K. Chambers, P. Trudgill, & N. Schilling-Estes (Eds.), The handbook of language variation and change (pp. 707728). Oxford, UK: Blackwell. Trudgill, P. (2002b). Sociolinguistic variation and change . Edinburgh, Scotland: Edinburgh University Press. van Geert, P. (1994). Vygotskian dynamics of development. Human Development, 37, 346 365. VanPatten, B. (1996). Input processing and grammar instruction in second language acquisition . New York: Ablex. Veronique, D. (1999). L emergence de cat egories grammaticales dans les langues cr eoles: Grammaticalisation et r eanalyse [The emergence of grammatical categories in creole languages: Grammaticization and reanalysis]. In J. Lang & I. Neumann-Holzschuh (Eds.), Reanalyse und Grammatikalisierung in den romanischen Sprachen (pp. 187209). Tubingen, Germany: Max Niemeyer Verlag. Veronique, D. (2001). Gen` ese(s) et changement(s) grammaticaux: Quelques modestes lec ons tir es de l emergence des cr eoles et de lacquisition trang` des langues e eres [Grammatical genesis and change: Some small lessons taken from the emergence of creoles and the acquisition of

Nick C. Ellis
foreign languages]. In M. Matthey (Ed.), Le changement linguistique. Evolution, variation, h et erog en eit e. Travaux neufch atelois de linguistique 34/35 (pp. 273303). Neuch atel, Switzerland: Publications de l Universit e de Neuch atel. Verspoor, M., Lowie, M., & van Dijk, M. (2008). Variability in second language development from a Dynamic Systems perspective. Modern Language Journal, 92, 214231. Vygotsky, L. S. (1980). Mind in society: The development of higher mental processes . Boston: Harvard University Press. Waldrop, M. (1992). Complexity: The emerging science at the edge of order and chaos . New York: Simon & Schuster. Wertsch, J. V. (1985). Culture, communication, and cog-

249
nition: Vygotskian perspectives . Cambridge: Cambridge University Press. Wertsch, J. V. (1998). Mind as action . Oxford: Oxford University Press. Wertsch, J. V., Del Rio, P., & Alvarez, A. (Eds.). (1995). Sociocultural studies of mind . Cambridge: Cambridge University Press. Wischer, I., & Diewald, G. (Eds.). (2002). New reections on grammaticalization . Amsterdam: John Benjamins. Zipf, G. K. (1935). The psycho-biology of language: An introduction to dynamic philology. Cambridge, MA: MIT Press. Zipf, G. K. (1949). Human behaviour and the principle of least effort: An introduction to human ecology. Cambridge, MA: Addison-Wesley.