Online Resources: Digests
Grammar and Its Teaching: Challenging the Myths
Diane Larsen-Freeman, School for International Training (VT)
Grammar is often misunderstood in the language teaching field. The misconception lies in the view that grammar is a collection of arbitrary rules about static structures in the language. Further questionable claims are that the structures do not have to be taught, learners will acquire them on their own, or if the structures are taught, the lessons that ensue will be boring. Consequently, communicative and proficiency-based teaching approaches sometimes unduly limit grammar instruction. Of the many claims about grammar that deserve to be called myths, this digest will challenge ten.
1. Grammar is acquired naturally; it need not be taught.
It is true that some learners acquire second language grammar naturally without instruction. For example, there are immigrants to the United States who acquire proficiency in English on their own. This is especially true of young immigrants. However, this is not true for all learners. Among the same immigrant groups are learners who may achieve a degree of proficiency, but whose English is far from accurate. A more important question may be whether it is possible with instruction to help learners who cannot achieve accuracy in English on their own.
It is also true that learning particular grammatical distinctions requires a great deal of time even for the most skilled learners. Carol Chomsky (1969) showed that native English speakers were still in the process of acquiring certain grammatical structures in English well into adolescence. Thus, another important question is whether it is possible to accelerate students' natural learning of grammar through instruction. Research findings can be brought to bear on this question from a variety of sources (see Larsen-Freeman & Long, 1991). Pienemann (1984) demonstrated that subjects who received grammar instruction progressed to the next stage after a two-week period, a passage normally taking several months in untutored development. While the number of subjects studied was admittedly small, the finding, if corroborated, provides evidence of the efficacy of teaching over leaving acquisition to run its natural course.
With regard to whether instruction can help learners acquire grammar they would not have learned on their own, some research, although not unequivocal, points to the value of form-focused instruction to improve learners' accuracy over what normally transpires when there is no focus on form (see Larsen-Freeman, 1995).
2. Grammar is a collection of meaningless forms.
This myth may have arisen because many people associate the term grammar with verb paradigms and rules about linguistic form. However, grammar is not unidimensional and not meaningless; it embodies the three dimensions of morphosyntax (form), semantics (meaning), and pragmatics (use). As can be seen in the pie chart in Figure 1, these dimensions are interdependent; a change in one results in change in another. Despite their interdependence, however, they each offer a unique perspective on grammar. Consider the passive voice in English. It clearly has form. It is composed minimally of a form of the be verb and the past participle. Sometimes it has the preposition by before the agent in the predicate: (1) The bank was robbed by the same gang that hijacked the armored car. That the passive can occur only when the main verb is transitive is also part of its formal description.
The passive has a grammatical meaning. It is a focus construction, which confers a different status on the receiver or recipient of an action than it would receive in the active voice. For example, the bank in sentence (1) is differently focused than it would be in the active sentence: (2) The same gang robbed the bank.
When or why do we use the passive? When the receiver of the action is the theme or topic, when we do not know who the agent is, when we wish to deliberately conceal the identity of the agent, when the agent is obvious and easily derivable from the context, when the agent is redundant, and so on.
To use the English passive voice accurately, meaningfully, and appropriately, English as a second language students must master all three dimensions. This is true of any grammatical structure.
3. Grammar consists of arbitrary rules.
While there is some synchronic arbitrariness to grammar, not all of what is deemed arbitrary is so. If one adopts a broad enough perspective, it is possible to see why things are the way they are. Consider the following sentences: (3) There is the book missing. (4) There is a book missing.
Grammar books will say that sentence (3) is ungrammatical because sentences with existential there almost always take an indefinite noun phrase in the predicate. Why? The reason is not arbitrary. There is used to introduce new information, and the preferred position for new information is toward the end of a sentence. A noun phrase that contains new information is marked by the use of the indefinite article, a or an, if it is a singular common noun, as in sentence (4).
4. Grammar is boring.
This myth is derived from the impression that grammar can only be taught through repetition and other rote drills. Teaching grammar does not mean asking students to repeat models in a mindless way, and it does not mean memorizing rules. Such activities can be boring and do not necessarily teach grammar. This does not mean there is no place for drills, but drills should be used in a meaningful and purposeful way. For example, to practice past-tense yes/no sentences in English, the teacher may ask her students to close their eyes while she changes five things about herself. She takes off one shoe, takes off her watch, puts on her glasses, puts on her sweater, and takes off her ring. Students are then asked to pose questions to figure out the changes she has made. Students may ask, "Did you take off a shoe?" or "Did you put on a sweater?" This kind of activity can be fun and, more importantly, engages students in a way that requires them to think and not just provide mechanical responses. Teaching grammar in a way that engages students may require creativity, but the teaching need not and should not be boring.
5. Students have different learning styles. Not all students can learn grammar.
Research shows that some people have a more analytical learning style than others. According to Hatch (1974), some learners approach the language learning task as "rule formers." Such learners are accurate but halting users of the target language. Others are what Hatch calls "data gatherers," fluent but inaccurate producers of the target language. This observation by itself does not address whether or not all students can learn grammar. While it may be true that learners approach language learning differently, there has been no research to show that some students are incapable of learning grammar. Students have different strengths and weaknesses. It is clear that all students can learn grammar as is evident from their mastery of their first language. As grammar is no different from anything else, it is likely that students will learn at different rates.
6. Grammar structures are learned one at a time.
This myth is demonstrably untrue. Teachers may teach one grammar structure at a time, and students may focus on one at a time, but students do not master one at a time before going on to learn another. There is a constant interaction between new interlanguage forms and old. Students may give the appearance of having learned the present tense, for example, but when the present progressive is introduced, often their mastery vanishes and their performance declines. This backsliding continues until the grammar they have internalized is restructured to reflect the distinct uses of the two tenses. We know that the learning curve for grammatical structures is not a smoothly ascending linear one, but rather is characterized by peaks and valleys, backslidings and restructurings.
7. Grammar has to do only with sentence-level and subsentence-level phenomena.
Grammar does operate at the sentence level and governs the syntax or word orders that are permissible in the language. It also works at the subsentence level to govern such things as number and person agreement between subject and verb in a sentence. However, grammar rules also apply at the suprasentential or discourse level. For example, not every choice between the use of the past and the present perfect tense can be explained at the sentence level. Often, the speaker's choice to use one or the other can only be understood by examining the discourse context. Similarly, use of the definite article with a particular noun phrase after the noun phrase has been introduced in a text is a discourse-governed phenomenon. It would be a mistake to teach students grammar only at the sentence and subsentence levels. Much of the apparent arbitrariness of grammar disappears when it is viewed from a discourse-level perspective.
8. Grammar and vocabulary are areas of knowledge. Reading, writing, speaking, and listening are the four skills.
While grammar can be thought of as static knowledge, it can also be considered a process. Language teachers would not be content if their students could recite all the rules of grammar but not be able to apply them. The goal is for students to be able to use grammar in an unselfconscious fashion to achieve their communicative ends. As with any skill, achieving this goal takes practice.
What sort of practice is warranted? Ellis (1993) postulates that structural syllabi work better to facilitate intake than to teach learners to produce grammatical items correctly. He suggests that grammar teaching should focus on consciousness raising rather than on the practice of accurate production. In support of this assertion is VanPatten and Cardierno's (1993) finding that students' experience with processing input data is more effective than giving students a grammatical explanation followed by output practice.
9. Grammars provide the rules/explanations for all the structures in a language.
Explaining why things are the way they are is an ongoing quest. Because languages evolve, linguists' descriptions can never be complete for all time; they have to accommodate the changing nature of language. For example, most grammar books make clear the fact that progressive aspect is not used with stative verbs; therefore, the following would be ungrammatical: (5) I am wanting a new car. For some English speakers, the sentence is not ungrammatical, and even those who find it so would be more inclined to accept progressive aspect when it co-occurs with perfective aspect, as in: (6) I have been wanting a new car (for some time now).
The point is, languages change, and any textbook rule should be seen as subject to change and non-categorical. Just as grammar learning is a process--witness the persistent instability of inter-languages--so is grammar itself. There is little static about either.
10. "I don't know enough to teach grammar."
Teachers often say this when they have opted to teach one of the other language skills, or when they choose to teach a low-proficiency class. While it is true that teachers can only teach what they know, teachers who articulate the above often know more than they think they do. The pie chart introduced earlier can be a useful tool for teachers to collect what they know about form, meaning, and use of a particular grammar structure. What they don't know will become apparent from the gaps on the chart and the gaps will nominate themselves as items for the teacher's agenda for further study. After all, what better way to learn something than to teach it?
If the goals of language instruction include teaching students to use grammar accurately, meaningfully, and appropriately, then a compelling case can be made for teaching grammar. Instead of viewing grammar as a static system of arbitrary rules, it should be seen as a rational, dynamic system that is comprised of structures characterized by the three dimensions of form, meaning, and use.
Chomsky, C. (1969). Linguistics and philosophy. In S. Hook (Ed.), Language and philosophy. New York: New York University Press.
Ellis, R. (1993). The structural syllabus and second language acquisition. TESOL Quarterly, 27, 91-113.
Hatch, E. (1974). Second language learning-universals? Working Papers on Bilingualism, 3, 1-17.
Larsen-Freeman, D., & Long, M. (1991). An introduction to second language acquisition and research. London: Longman.
Larsen-Freeman, D. (1995). On the teaching and learning of grammar: Challenging the myths. In F. Eckman et al. (Eds.), Second language acquisition theory and pedagogy. Mahwah, NJ: Lawrence Erlbaum.
Pienemann, M. (1984). Psychological constraints on the teachability of languages. Studies in Second Language Acquisition, 6, 186-214.
VanPatten, B., & Cadierno, T. (1993). Explicit instruction and input processing. Studies in Second Language Acquisition, 15, 225-44.
Larsen-Freeman, D. (Series Director). (1993; 1997). Grammar dimensions: Form, meaning, and use. Boston: Heinle & Heinle.
This report was prepared with funding from the Office of Educational Research and Improvement, U.S. Dept. of Education, under contract no. RR93002010. The opinions expressed do not necessarily reflect the positions or policies of OERI or ED.