Principle 5: Basic to use of language are language knowledge and language control
Basic to use of a language is a mental representation of how that language works. We need a certain basis of systematic knowledge in order to be able to operate in the language, no matter how minimally. We cannot learn everything we wish to express in a language one thing at a time; there is an infinite number of potential sentences we may wish to utter, yet each has a structural framework that can be used to convey many other messages. The human mind systematizes and organizes material to make it manageable, and this systematization is basic to our mental representation of the language. We cannot operate in a language without such a mental representation because comprehensible language use at any level is rule-governed; in other words, to understand and be understood we cannot use elements of the new language haphazardly. We need a succinct, internalized structural model as a plan to direct our tactics (in this instance, expression of personal meaning).
All languages are organized at several levels (phonological, syntactic, semantic, and pragmatic), and these various subsystems interact within the mental representation. Grammatical structure and vocabulary, which are interrelated in their functioning, provide the tools for expressing semantic and pragmatic meaning. As Halliday expresses it: “Reality consists of ’goings-on’: of doing, happening, feeling, being. These goings-on are sorted out in the semantic system of the language, and expressed through the grammar of the clause.” Once we have internalized the fundamentals of this organization for our new language (linguists are working continually at systematizing the details and even native speakers are still learning its potential), we are liberated to express a multiplicity of meanings.
A ballet dancer learns basic steps and then can fly. Scientists learn basic principles and can then build new knowledge: experimenting, applying, thinking creatively. Scientists may not fully understand the principles (research is always continuing), and their findings may even cast doubt on some of these principles or even cause them to restructure their concept of the basic framework, but the basic framework is there to be reinterpreted. Linguistic scientists continue to research linguistic structure. They may propose new models or variations of existing models to explain what is there, but they cannot restructure the object of their study: they can only reinterpret the operation of the basic framework or bone structure of the language.
Is the problem for language learning an inadequate or incomplete description of the linguistic structure of the language? People have learned languages for millennia while awaiting discovery and description of the ultimate adequate model of what they are learning, just as people have grown crops and lit fires, without a complete understanding of the processes underlying these phenomena. What language learners need is a functional mental model of the linguistic structure of a particular language that works for them in producing at a basic level speech that communicates their meanings}; their present needs obviate their waiting for final decisions by linguistic researchers on the structural model that will best describe the inner workings of the basic framework. Chomsky speaks of language behavior as being governed by an innate knowledge of a system of rules of universal grammar, with values of parameters set according to the language within which one is operating; this systematic framework consists of a core or basic skeleton on which the language is constructed and a periphery of marked exceptions that are added to the core on the basis of specific experience . This is one description of a mental representation; there are others. The real problem for the language learner is the setting of the values of the parameters for the mental representation of the new language. How this can be achieved most effectively and economically is a subject for research for applied linguists _ research to which classroom teachers, who are observing students daily can contribute valuable insights.
We cannot use a language without some sort of mental representation of the basic framework or mechanism, no matter how personal and idiosyncratic; this we shall call knowledge of language . Learners pass through a succession of interim grammars at various stages in the learning process, as they work to construct a functional model, or establish values of parameters for the new language, in the Chomskyan sense. Teachers can help students acquire an understanding of this basic mechanism that is sufficient to enable them to use it at their particular stage of development, and to further refine this understanding as they progress. Without expert help, students will acquire some form of mental representation, often incorporating elements of the mental representation of their first or another language; with help they will acquire more rapidly one that is closer to the native model. Years of experience with learners of many languages in the federal government’s language institutes, which comprise the Interagency Language Roundtable, have demonstrated that, when language learners try to express their meanings freely without a firm structural framework, “incorrect communication strategies … fossilize prematurely,” and “their subsequent modification or ultimate correction is rendered difficult to the point of impossibility, irrespective of the native talent or high motivation that the individual may originally have brought to the task.” Yet precision of construction of meaningful phrases is important for fluent expression of meanings that will be understood by native speakers.
Students acquire this precision of expression through performing rules, not through memorizing or discussing them. Knowledge of the systematic interrelationships that constitute the structure of the language is acquired actively through use in communicative contexts. In this way it becomes part of the learner’s mental equipment (it becomes internalized) and can be called upon readily, even if more and more without conscious focus, to express personal meaning or to comprehend and recreate the meanings others are trying to convey. It can also be re-examined consciously, should there be a need to reinterpret its potential. As William James observed: “Experience is never yours merely as it comes to you, facts are never mere data, they are data to which you respond , your experience is constantly transformed by your deeds.”
Sharwood Smith expresses succinctly this necessity for performing rules when he says: “Whatever the view of the underlying processes in second language learning … it is quite clear and uncontroversial to say that most spontaneous performance is attained by dint of practice. In the course of actually performing in the target language, the learner gains the necessary control over its structures such that he or she can use them quickly without reflection.”  Experimentation in Sweden has demonstrated that this performance practice is most effective when it involves student-initiated utterances, which constitute student response to the data in James’s sense and gradually transform the student’s mental representation.
Performing rules, then, provides the natural bridge to using these rules in creating personal messages, which we shall call control of language. Language control necessarily implies the ability to understand messages and their full implications in the context, social and cultural, interpreting tone of voice, stress, intonation, and kinesics, as well as actual words and structures. In expression, it implies more than syntactic accuracy; it requires also syntactic appropriateness in contexts of use and in culturally determined relationships. Once some degree of language control has been attained, language is used “as a medium which will engage the thought, perception, and imagination of the learner.”