Input processing revisited.
Current models of language pedagogy fail to provide with the abstract principles that a comprehensive theory of Second Language Acquisition (SLA) might require. All of these models acknowledge the important role of attention in any learning process, but they fall short of offering any linguistic model that could account for the psychological complexities involve in attention and perception. To exemplify this, the article puts into question (a) the internal validity of the principles stated in Van Patten's Input Processing Theory, and (b) the assumptions regarding language processing in Van Patten's language acquisition scheme. The article proposes an alternative scheme that redefines the set of cognitive processes involved in language acquisition and that incorporates the findings of Psychology and Pedagogy.
In spite of multiple research studies focusing on the ways in which learners process second-language input, it remains largely unknown how learners develop competence in that second language. VanPatten, a leading proponent of the input-processing approach to SLA, stated that Krashen (1982) provided "the strongest position on the role of comprehensible input" (VanPatten, 1995, p. 170). Even though critical evaluation of Krashen's model reveals that its premises are trivial at best, his work is still frequently cited in SLA research partly because criticism to the model "has served to underscore the need [...] to examine what learners do with and to input as part of the acquisition process" (VanPatten 1995:170). It is true that Krashen's model examined for the first time the interaction between learners and input as part of the acquisition process and its implications in teaching foreign languages, but it should be noted that the notion of comprehensible input is not new to general cognitive processes described in Psychology and Phenomenology.
Most research in SLA is informed by the presupposed existence of a language module that process information independently of other cognitive faculties of the mind. This paper investigates the psychological underpinnings of comprehensible input in SLA. Van Patten's theory of Input Processing will be examined in order to show that dealing with second language acquisition is far too complex a task to be reduced to a restrictive linguistic theory. Language is an organization of cognitive faculties that involve as much attention, conscious and unconscious, as memory. Language outcomes in adults differ from that in children, among other reasons because of their different metalinguistic knowledge. That by itself should constitute a counterargument against the idea of a cognitively independent language module. However, the fact that meaningful approaches have proven successful for adults as much as for children may be an indication that (a) the cognitive mechanisms of children and adults are the same and that (b) cognitive mechanisms in children are available to adults. This premise is not at all new to the field of SLA[I], but it takes a completely different direction when the approach shifts from generativist linguistics to lexical-functional linguistics. These cognitive mechanisms constitute what Lakoff refers to as conceptualizing capacity, that is, "People share a general conceptualizing capacity regardless of what difference they may have in conceptual systems" (Lakoff, in Albertazzi 2000, p. 77).
A theory of cognitive semantics, such as lexical-functional linguistics, links semantic structure to general cognitive processes described in Psychology. The construal operations, you may want to call it problem-solving strategies, that enter our conceptualizing capacity are not different from the gestalt schemes that enter the world of perception. Croft & Wood (2000) list these construal operations under four processes: Attention, Judgment, Perspective, and Constitution. They all offer accounts of how the mind extracts meaning creating a whole from a fragmented perception. In this paper, I will argue that Van Patten's Input Processing Theory articulates a general pedagogical principle but that it falls short of offering a coherent language-learning theory that incorporates the basic cognitive abilities we normally apply to the various realms of our experience.
Meaning and Cognition in Van Patten's Input Processing Theory
VanPatten cites evidence for a cognitive principle that serves as a foundation for communicative approaches: Attention capacity interacts with meaningful input. This is hardly a new established principle. In 1988, Andersen proposed a cognitive "Operating Principle" known as Formal Determinism.
When the form/meaning relationship is clearly and uniformly encoded in the input, the learner will discover it earlier than other form/meaning relationships and will incorporate it more consistently within his interlanguage system (Andersen, 1988, p. 16).
Finding out how form/meaning relationships may be encoded in the input is the basic tenet of Language Pedagogy. VanPatten (1995) proposes the Input Processing Theory, which he considers important for developing a theory of acquisition of a second language. In a first development of his theory, VanPatten (VanPatten 1995, p. 172) states in a set of hypotheses what the encoding of form/meaning relationships entails at the morphological level. It is noteworthy that these hypotheses do not take the role of conscious attention into account, even though VanPatten acknowledges that only conscious attention to input can be responsible for language acquisition: "You can't learn a foreign language [...] through subliminal perception" (Schmidt, 1990, in Van Patten 1995, p. 173 and 1998, p. 115). Input Processing states that learners process meaning over form and that conscious attention is subordinated to unconscious attention, but it does not offer explanations on how conscious attention extracts meaning and incorporates it into the speaker's linguistics system.
Input Processing Theory rests on (a) the presupposed validity of the separation of form and meaning and (b) the a-priori distinction between meaningful and non-meaningful morphology . Both assumptions imply an understanding of syntax and semantics as separate realms of linguistic knowledge, which then guides Van Patten's entire Input Processing Theory. According to VanPatten (1995:173, and 1998: 115), form and meaning often compete for cognitive resources. In other words, the learner is "driven to look for the message in the input [...] before looking for how the message is encoded." This statement seems to obviate the fact that meaning is an integral part of the information represented in lexical entries. Words create their own syntactic environment. They do not exist in the mind of the speaker by themselves, but in relationship to other words. In the process of form/meaning mapping in language acquisition, any new concept is incorporated into a language developing system according to a given encoding of linguistic information. In the case of learners of a second language, this encoding may have two sources of lexical interpretation: Speakers either map new entries to meanings developed in the process of L2 acquisition, or to the existing meanings in their L1 system.
VanPatten maintains that, from a universal standpoint of language acquisition, some morphological markers are more meaningful than others are. He cites an overwhelming amount of evidence regarding the order in which some English morphemes are acquired. Actually, VanPatten (1984) seems to offer a sound explanation for the order of acquisition of -ing forms, past tense, and third person markers based on their relative communicative value. However, his conclusions provide evidence for a different claim: No marker can be compared to another without considering the conceptual mapping of the structure to which they are associated. From a universal standpoint, the relative communicative value of a form could only explain patterns of acquisition of such form in comparable communicative environments. Moreover, we cannot presume that these three forms per se are conceptually comparable to one another in any given language.
Further developments of Van Patten's principles of Input Processing Theory (VanPatten, 2003, p. 420) seek to offer new insights into the relationship between grammar and meaning--or between grammar and cognition-. In all, the final formulation of the theory is little more than a restatement of the original, and leaves how the mind encodes meaning unexplained. Let me summarize some of the conceptual problems that the latest installment of the Input Processing Theory presents. The theory implies a distinction between the lexical and the grammatical systems. The distinction between the two systems not only makes impossible to distinguish content words from lexical items--both terms used distinctively in the theory-, but it presents a logical problem. According to Input Processing, the learner processes lexical items before grammatical items only "when they encode the same semantic information". This implies that there must be instances in which form guides meaning. However, the learner can only know if lexical items and grammatical forms encode the same semantic relations when he/she is able to identify both, content and grammar. In other words, an utterance is meaningful only when both systems, the lexical and the grammatical, encode the same information.
The theory also states that learners tend to process lexical items in a particular order. That particular order does not refer to a linguistic principle, but to what it seems a developmental principle: the learner will process the first noun or pronoun as a subject when lexical semantics, event probabilities, or contextual constraints allow for this interpretation. The assumption is that, for the learner to determine the reading of the first noun or pronoun as subject, constraints and probabilities have to be encoded in the lexical system. So the fundamental question remains unanswered: How do learners encode the L2 lexical system? Until explanations in terms of more abstract principles are given , generalizations of the type stated in Van Patten's Principles of Input Processing could contribute little to our understanding of the organization of linguistics systems in SLA. Of the principles governing Input Processing, the Principle of Primacy of Meaning reinstates a pedagogical principle underlying communicative approaches. As a pedagogical principle, it is external to the field of linguistics. A coherent theory of SLA must consider the inherent and fundamental connection between meaning and lexical categories--in which the grammatical system is encoded- in broader cognitive terms. The section below addresses the role of cognitive faculties in language processes.
Language Acquisition Scheme
VanPatten's Language Acquisition scheme distinguishes three sets of cognitive processes: Processing, Accommodating/Restructuring, and Monitoring/Accessing; and three distinctive informational systems: the Input System, the Developing System, and the Output System. He also includes a vaguely defined interim, the Intake System. In such scheme, L2 acquisition learners move from the Input System to the Intake System by virtue of the set of cognitive faculties involved in Processing; then, from the Intake System to the Developing System by Accommodating/Restructuring; and finally, from the Developing System to the Output System by Monitoring/Accessing. VanPatten fails to explain how the set of processes under Processing is distinctive from those under Accommodating/Restructuring. In other words, he does not explain how the Intake System differs from the Developing System. If intake is understood as "that part of the input that the learner notices" (Schmidt, 1990, p 139), which is the part of the input that has become meaningful, then the distinction between intake and developing system becomes irrelevant in the acquisition scheme.
The set of processes under Accommodating borrows the term from Piaget (see Beilin & Pufall, 1992). Assimilation and accommodation, as understood in developmental psychology, are two complementary ways to process stimuli. Assimilation implies the mapping of new entries over existing patterns in the developing system: Learners attend to input and incorporate it into an already existing system. As data enter the developing system, the system readjusts itself in order to incorporate newer data. Accommodation is the adjustment made in the existing patterns in the system to incorporate these data. Van Patten calls Restructuring to this adjustments. For him, the mechanisms involve in Restructuring are the mechanisms responsible for the acquisition of syntax according to Chomsky's Universal Grammar (Van Patten, 1995:171). This presents another logical problem: there cannot be Monitoring if such a self-sufficient data processing language module were solely responsible for Restructuring.
Monitoring, as defined, by VanPatten (2003, p. 415) incorporates Swain's (1985) output hypothesis, in which output fosters acquisition by promoting learners interaction with input . Thus, Monitoring Processes makes use of conscious channels of rule formation to increase the incidence of correct utterances. Using conscious channels implies that the function of language input system is affected by information in other systems. We may have to conclude that, as part of the conscious channels by which knowledge of any kind is attained, general cognitive principles such as memorization, induction, and deduction, among other, do play a role in language acquisition.
The role of instruction
All of the discussions above lead us to a revision of the role of instruction and of the theoretical underpinning of communicative approaches to teaching languages. Grammar-based methodologies, even if grammar instruction is limited to explain regularities in the target language, are disregarded because they assume that L2 learning mechanisms are not different from mechanisms in other domains of knowledge. However, and without advocating for any grammatical methodology, instruction is only possible if we give to L2 linguistic knowledge the same status as other kinds of knowledge.
The validity of any meaning-based approach to SLA cannot be based on the grounds that language is a system in its own right. By acknowledging this, we understand that SLA responds to a psychological complexity that extends beyond our current knowledge of any possible cognitively autonomous language acquisition device. Cognitive semantics proposes a paradigm that revives theories of descriptive psychology developed at the beginning of this century: Gestalt theory and experimentation (see Lakoft, 1977). Gestalt signifies a structured complex whose elements have value only in relation to the whole. Each element that enters our experience is determined by its relationship to other elements to which it is associated. Gestalt also signifies form, shape: A gestalt is a whole of relations that we can grasp only if we are able to unify everything in one intellectual glance. Intimately linked to the psychology of Gestalt, Graziano (1975) proposes a language construct known as Language Operational Gestalt (LOG), "a radically pragmatic and empirical way of experiencing that combines many of the contribution of Language Analysis [...] and Gestalt" which integrates isolated sectors of our experience. LOG is a theory of meaning: Only what can be experienced in the here and now has meaning or use. LOG is at the core of pedagogic proposals aimed to explore attention processes: input processing, input enhancement (Rutherford & Sharwood Smith, 1985; Sharwood Smith, 1993) and focus on form (Doughty, 2001; Doughty & Williams, 1998; Long, 1991; Long & Robinson, 1998). As Graziano says
The uses of a sound configuration are demonstrated by experiencing the same sound in connection with various situations to obtain different purposes. That sound is experienced as one configuration of many phenomena, a gestalt completed when the meaning is understood. (Graziano, 1975:1)
Only repeated experiences with comprehensible input can be responsible for language acquisition: Language generalizations require that learners have processed enough data in the context of meaningful communicative activity. Limitations in time and interaction in classroom instruction given, what seems to be at stake is how to reconcile input frequency with explicit methods of teaching. For Ellis the answer is straightforward: "language acquisition can be speeded by explicit instruction" (Ellis, 2002, p. 145). The effectiveness of one method over the other may be a futile argument until we know what is stored in the speaker's knowledge of the target language. This means that any implicit teaching, as well as any explicit teaching, assumes rules derived from a reference grammar of some kind with no certain knowledge of how a particular feature is represented in the mind of the learner. The fact that generalizations and categories about the target language may be implicitly taught does not take away from the fact that they have been formulated explicitly, at least to the extent to which it is necessary in order to structure the contents in the curriculum of any instructional setting.
How attention interacts with input is the point of departure for Van Patten's set of Input Processing principles, but none of these principles offers the basic tenets for the development of a theory of acquisition of a second language: essentially, they represent generalizations that lack explanatory value. Furthermore, the epistemology underlying the hypothesis, that is, the existence of a specific language domain for L2, undermines cognitive strategies that may enter the process of acquisition and that are external to any possible language module. If the language specific domain existed, then instruction would be completely irrelevant since what instruction offers is a shortcut to the process of language acquisition. It is without question that acquisition requires comprehensible input. However, how the learner moves from comprehension to production remains uncertain. As language educators, we should not discard any type of activity that involve the incorporation of non linguistic faculties as means of acquiring knowledge of a language, at least, "not until we are better informed by what we know about processes involved in acquisition" (VanPatten, 2003:419).
Andersen, R. (1988). Models, processes, principles, and strategies: Second language acquisition in and out the classroom. IDEAL, 3, 77-95.
Albertazzi, L. (Ed.). (2000). Meaning and cognition. Philadelphia: John Benjamins.
Beilin, H., & Pufall, P. (Eds.) (1992). Piaget theory: Prospects and possibilities. Hillsdale, N.J.: L. Erlbaum Associates.
Croft, W., & Wood, E. (2000). Construal operations in linguistics and artificial intelligence. In L. Albertazzi (Ed.), Meaning and cognition (pp. 51-78). Philadelphia: John Benjamins.
Doughty, C. (2001). Cognitive underpinnings of locus on form. In P. Robinson (Ed.), Cognition and second language instruction (pp. 206-257). New York: Cambridge University Press.
Doughty, C., & Williams, J. (1988). Pedagogical choices in focus on form. In C. Doughty & J. Williams (Eds.), Focus on form in classroom second language acquisition (pp. 197-262). New York: Cambridge University Press.
Ellis, N. (2002). Frequency effects in language processing. Studies in Second Language Acquisition, 24, 143-188.
Graziano, E. E. (1975). Language Operational Gestalt awareness: a radical empirical and pragmatical phenomenology of the process and systems of library experience. Tempe, Arizona: The Association for Library Automation Research Communications.
Krashen, S. D. (1982). Principles and practice in second language acquisition. Oxford: Pergamon Press.
Lakoff, G. (1977). Lingusitics Gestalts. In W.A. Beach & S.E Fox (Eds.), Papers from the Thirteenth Regional Meeting of the Chicago Linguistic Society (pp. 236-287). Chicago: University of Chicago Press.
Long, M. (1991). Focus on form: A design feature in language teaching methodology. In K. de Bot, R. Ginsberg & C. Kramsch (Eds.), Foreign language research in cross-cultural perspective (pp. 39-52). Amsterdam: John Benjamins.
Long, M., & Robinson, P. (1998). Focus on form: Theory, research, and practice. In C. Doughty & J. Williams (Eds.), Focus on form in classroom second language acquisition (pp. 15-41). New York: Cambridge University Press.
Rutherford, W., & Sharwood Smith, M. (1985). Consciousness raising and Universal Grammar. Applied Linguistics, 6, 274-282.
Schmidt, R. (1990). The role of consciousness in second Language learning. Applied Linguistics, 11, 129-58.
Sharwood Smith, M. (1993). Input enhancement in instructed SLA: Theoretical bases. Studies in Second Language Acquisition, 15, 165-179.
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 Language Acquisition (pp. 235-253). Rowley, MA.: Newbury House.
VanPatten, B. (1984). Communicative value and information processing in second language acquisition. In P. Larson, E. Judd, & D. Messerschmitt (Eds.), TESOL "84 (pp. 89-99). Washington: TESOL
VanPatten, B. (1995). Cognitive aspects of input processing in second language acquisition. In P. Hasehmipour, R. Maldonado, & M. Van Naerssen (Eds.), Studies in language learning and Spanish linguistics in honor of Tracy D. Terrell (pp. 170-183). New York: McGraw-Hill.
VanPatten, B. (1998). Cognitive characteristics of adult second language learners. In H. Byrne (Ed.), Learning foreign and second languages (pp. 105-127). New York: MLA.
Van Patten, B. (2003). The evidence is IN: drills are OUT. Foreign Language Annals, 36, 403-420.
 Two are the traditional competing models of SLA:
a) The Universal Grammar (UG) based model: The same language learning strategies are at work in L1 and adult L2 acquisition and
b) The problem-solving model: Non-U cognitive mechanisms determine the differences between L1 and L2 grammars
 Morphological morphemes refer to the open class categories that constitute the grammatical system according to Cognitive Linguistics.
 In terms of propositional logic alone, some principles stated in Input Processing Theory are irrelevant. For instance, if redundancy is subordinated to meaning--Principle P1.d-, stating that the learner processes non-redundant meaningful forms before redundant meaningful ones--Principle Pl .c- is a false statement. When an item considered redundant enters the learner's attention more easily than a non-redundant one, it is reasonable to believe that the learner may perceive the item as non-redundant.
 Van Patten argues that for Swain output plays three roles in promoting acquisition: Noticing linguistic features in the input, hypothesis testing, and promoting conscious awareness of language and language use.
Antonio Gragera, Texas State University San Marcos
Gragera, Ph.D., is Assistant Professor in the Department of Modern Languages.