Grammatical knowledge can be defined as a potentially dynamic, interactive knowledge system about what is to be preferred and what avoided in morphology and syntax. For me, the key words that relate to knowledge development in this definition are “potentially dynamic” and “interactive”. The more interactive a learner is with various mediums, the more dynamic his grammatical knowledge will be, and the more he will be able to understand what morphological and syntactic forms are required in a given situation. Thus, in my view, any approach that attempts to explain how L2 grammar develops needs to revolve around the concept of “interaction.”
The interactionist approach asserts that learner interaction with various mediums facilitates learning. In his Interaction Hypothesis, Long (1996) argues that when meaning is negotiated, it facilitates acquisition because “it connects input, internal learner capabilities, particularly selective attention, and output in productive ways” (pp. 451-452). So a learner, through focused negotiation work, can compare what they “know” grammatically with the target language through input and grammatical knowledge development can ensue as a result, either immediately or at a later time (interaction can “set the stage” for learning). Output can either be a reflection of grammatical knowledge development or a means to expand that knowledge. This essay will demonstrate that the major tenets of this theory are compelling, and this why the interactionist view of language, with a few additions from other theories and approaches, is the basis of my theory of second language grammar acquisition.
There are certain aspects of L2 learning, such as vocabulary, pragmatics and phonology that can be heightened when grammatical knowledge development occurs through interaction. With regard to vocabulary, a learner, through interaction with a textbook, can learn, for example, about the roles of noun phrases described by verbs. If a textbook has an exercise that explains the difference between the usage of “to” and “from”, it may use content that includes the following sentences:
(1-A) Jeff rents a VCR to Bill.
(1-B) Bill rents a VCR from Jeff.
In 1-A, Jeff (the subject) is the owner because rent is accompanied by to. In 1-B, when rent is accompanied by from, the subject is the person who had control of the VCR for a limited time (Bill). Learners, by studying the explanatory notes that would accompany this exercise, have the opportunity to be able to distinguish between the two sentences.
Pragmatics is one aspect where conversational interaction with a native speaker can be of great practical help to the learner. In a speech act, such as refusing, an answer can be grammatical but inappropriate, like in the following:
(2-A) Native speaker: Would you like to come to my party?
(2-B) L2 learner: No.
If a learner sees a frown on the face of the native speaker after his reply, which is likely, he will realize that he has said something inappropriate. This interaction can set the stage for learning because the learner can find out later what a more appropriate answer is, such as “I’m sorry I can’t. I have a previous engagement” and use it, or something similar, the next time he is in that kind of situation.
The relationship between phonology and grammatical knowledge development is perhaps more indirect than for the other two aspects already mentioned. One reason is because the native language often has more impact on phonology than grammatical knowledge does. A learner’s pronunciation is almost always clearly attributable to the NL because learners generally try to maintain their NL syllable structure. Still, there are opportunities for learners to potentially change pronunciation habits through interaction that encourages a focus on grammatical knowledge. When looking at the Markedness Differential Hierarchy (Dinnsen & Eckman, 1975), it can be seen that a Korean learning English has a lot to learn about voice contrast in all positions, because, in Korean, there are no voice contrasts in any positions. So, if he is in a classroom setting, interacting with the teacher and other students will allow him to gain the morphological knowledge necessary to pronounce words correctly. Also, as he learns more about syntax and creates longer sentences, he can start to practice using fast speech, as in 3-B:
(3-A) I am going to meet her later.
(3-B) I’m gonna meda layda.
In both cases, his output will provide a means to expand his knowledge because he will be able to receive feedback from the teacher.
Output is the evidence of a learner’s overall ability to use the L2, and the example above gives insight into the relationship it has with grammatical knowledge. An examination of the Korean’s learning process shows that output can be a reflection of grammatical knowledge, but conversely, grammatical knowledge can be a reflection of output (there are theorists who only subscribe to the first statement). With regard to the former, when a learner expands his knowledge through interaction, changes in the grammatical rule system often surface in output. For example, he might learn not to say “No” in refusing an invitation and use “No, thank you” or “Sorry, I have other plans” instead. Also, there can be changes to the underlying system that can lead to reanalysis or restructuring that can affect the potential for output. Orange juice may be learned as orangejuice, but later reanalyzed as orange+juice, and then apple juice and other similar forms are possible. With regard to the latter, the textbook looks at four ways in which output has a significant role in the development of syntax and morphology (Swain, 1995):
a) Testing hypotheses about the structures and meanings of the target language
b) Receiving crucial feedback for the verification of these hypotheses
c) Developing automaticity in IL production
d) Forcing a shift from more meaning-based processing of the second language to a more syntactic mode
Each of these is discussed in detail, but I will only briefly look at supporting arguments for each one. Swain (1995) showed how a student learning French was able to change a faulty hypothesis through interaction. White (1991) found that negative evidence promoted learning of adverb placement. Gass and Selinker argue that automatic processing results from consistent and successful mapping of grammar to output and output can initiate feedback, which focuses the learner’s attention on grammatical problems, which leads to reassessment.
As has already been shown, output influences feedback, one of the external conditions needed for grammatical knowledge to develop. However, the most significant external factor that influences the development of grammatical knowledge is input. There is some debate in linguistic circles about what input is and how it affects learning. The critical input for a learner is what I call apperceived comprehensible input (ACI). To utilize input, it needs to be apperceived – newly observed aspects of a grammatical form are related to past experiences. But input must also be comprehensible, or understood by the learner (my use of the word comprehensible is different to Krashen’s comprehensible input theory, due to aspects of his theory that are incompatible with my model of learning). Past experience with language does not ensure comprehension. When a learner receives ACI and recognizes differences in the input and his grammatical knowledge, this provides an impetus or motivation for modifying that knowledge through the processes of intake (the mental activity that mediates between grammatical knowledge and input) and integration (the development of grammatical knowledge or storage that can be accessed at a later time).
The learner can most effectively gain exposure to ACI through interaction. So far, we have seen examples of learner/content (in the discussion on vocabulary), learner/native speaker (in the discussion on pragmatics) and learner/classroom (in the discussion on phonology) interaction. Another example of interaction is learner/learner, and the following study by Gass and Varonis (1989) demonstrates how ACI gives the learner the type of exposure and experience that is needed to learn grammar. Two different pairs of non-native speakers talk about drinking coffee and getting married. As a result of negotiation, both speakers in the first conversation end up using the correct form “on his knee” (rather than “in his knee” or “in him knee”). In the second conversation, one of the speakers is confused about using the form “get married”. However, he was able to contrast his original grammatical hypothesis with negative evidence, and after 16 turns, the correct form is clear in his mind.
According to Long (1996), perhaps the most important element of the interaction hypothesis is selective attention; without it, grammar development will not take place (Gass, 1997; Schmidt, 1990, 1993 – three sources, 1994). Negotiation requires involvement and attentiveness and if this occurs, a learner will be in a position to notice differences between the ACI and his interlanguage grammar and make modifications. So, in my opinion, selective attention is the part of the learning process that a language teacher most needs to focus on. He can do so by incorporating motivational strategies into his lesson plans, finding ways to minimize negative affective factors and using material that is relevant and interesting to students so that they are attentive and get involved in classroom activities. He can also encourage metalinguistic awareness. For example, a teacher could make learners aware of speech errors (in a sensitive way) through the questioning and clarification that occurs in interaction as a way to make them question what they “know.”
Another important part of the learning process that a language teacher needs to focus on is how he will manipulate ACI. Naturally, he will use different types of interaction: learner/content, learner/instructor, and learner/learner. But in what ways? For example, teachers need to be ready to intervene in student interactions. Swain and Lapkin (1998) show that without teacher intervention, students can either be uncertain about the correct form or will learn something incorrect. Teachers also need to consider carefully what material they use to teach specific grammatical forms. Long (1991) distinguishes between focus on form (meaning-focused activities in which a certain grammatical form is embedded) and focus on forms (traditional explicit methods of teaching grammar). Long argues that the former is the better way of teaching, and his assertion is supported by the work of other linguists. Gas, Mackey, Alvarez-Torres, and Fernandez-Garcia (1999) support the idea that taking away the cognitive burden of focusing on both form and meaning from students allows them a greater opportunity to focus on form. A third way in which teachers can manipulate input is by making linguistic and interactional modifications to their speech to allow for better comprehension by students. Linguistic modifications, or foreigner talk, include slowing the rate of talking, using simplified vocabulary and syntax, and giving discourse “hints”, such as using tag questions and offering corrections (Hatch, 1983). Interactional modifications include confirmation checks (Is this what you mean?), comprehension checks (Do you understand?) and clarification requests (What?) (Long, 1980).
In conclusion, it is my view that the analyses presented here amply support the interactionist approach. Although other ideas and theories have been added, they haven’t detracted from the compelling nature of the approach as an explanation for L2 grammar development. Key points of this approach – input, selective attention and output have all been validated as essential components for learning. The two threads that join these components together are grammatical knowledge and interaction. It seems evident that any approach that tries separating grammar knowledge and interaction would result in a distorted picture of L2 development.