Although research on second language acquisition and second language curriculum development has stressed the importance of grammar recycling in language materials, there’s still a crucial issue as to what grammatical structures should be taught and in what order they should be taught. Several criteria for
determining the selection and sequence of grammatical structures have been suggested and they might be summarized as: their inherent learning difficulty, their relations to learners’ native languages, their markedness, their relations to learner errors, and their utility and frequency in the input (Doughty, 1998; Doughty &
Williams, 1998; Ellis, 1993, 2006; Larsen-Freeman, 1991, 2003; Savignon, 1983;
Yalden, 1983).
The first criterion for selecting and sequencing structures for instruction was the inherent learning difficulty of different grammatical structures. According to Ellis (2002b, 2006), two different learning difficulties needed to be distinguished: (1) the difficulty learners have in understanding the grammatical structure, and (2) the difficulty learners have in internalizing the structure. Krashen (1982) also made a distinction between these two as he believed that there are rules that are easy to learn but hard to acquire and rules that are easy to acquire but hard to learn. Structures that are easy to learn but hard to acquire are the best choices for instruction. One example, as he mentioned, is the rule of third person-s in English, which is normally taught very early in a course.
Another criterion was the relations between learners’ L1 and the target language. It was believed target structures that are similar to those in the native language should be taught first as learners transfer at least some of the features of their L1 into L2 (Lado, 1957). Therefore, a contrastive analysis of structures may be conducted so as to identify the troublesome areas for learners (Larsen-Freeman, 1991).
Regarding the markedness of structures, Richard, Platt, and Weber (1985) noted that a grammatical structure which is in some sense infrequent, unnatural, and deviant from a regular pattern is usually more marked than the other ones. Therefore,
identified as marked. In general, teachers should teach the marked forms and leave the learners to learn the unmarked ones naturally by themselves (Ellis, 1993).
Although this criterion can be used as a basis for determining which linguistic forms to introduce early and which late, the concept of markedness is still opaque.
Another alternative for organizing structures was to base on the learner errors.
Teachers may need to observe the learners’ errors in order to further draw learners’
attention to errors and then help them compare the errors with the correct target forms (Ellis, 1993, 2006).
Utility and frequency of structures were also important criteria for choosing and grading structures (Mackey, 1965). ‘Utility’ refers to the relevance of structures to learners’ daily life or the coverage value of the structures. Usually when the communicative utility of the structures was concerned, structures such as the possessive determiners or the verb to be and basic word order were usually chosen to be taught earlier so that students might be able to introduce themselves and others at the beginning of the class, saying “My/his/her name is...” Besides, when certain structures were considered to be needed to form more complex structures, they might be put earlier or be put together with the complex one in a course (Larsen-Freeman, 2003). The verb be, which is basic to the language must be presented early and recycled often (McIntosh, 1979).
Frequency was helpful not only for deciding whether an item shall be selected for inclusion but also for deciding the sequence of the items for teaching (Halliday, McIntosh, & Strevens, 1964; White, 1998). Frequency was related to learner attention and was claimed that when a form appeared more frequently in the input, there’s more likely that learners would acquire and notice it (Bardovi-Harlig, 1987).
George (1963, cited in White, 1998) and his colleagues did several studies on frequency of structures based on their chosen English corpus. They found that the
most frequently occurring tenses in the corpus were the Simple Past narrative, the Simple Present actual, the Simple Past actual and the Simple Present neutral. As he defined, tenses ‘actual’ refers to at this or that time, ‘neutral’ refers to no specific time reference, and ‘narrative’ refers to a sequence of events. George’s finding (1972) also revealed that Simple Present tense referring to ‘now’ was much more frequent than the Present Progressive. This was a conflict with people’s expectation. In fact, high frequency did not necessarily lead to acquisition of all structures. According to Doughty and Williams (1998), articles and third person-s in English were the examples of frequently encountered structures that students often failed to acquire.
This might because that they are less salient or because that they carry less important communicative functions.
To sum up, no one actually knows what the most ideal way is for selecting and sequencing structures as there are still other factors, such as learnability or student readiness, need to be considered. Larsen-Freeman (2001, 2003), therefore, emphasized that it might be helpful to see structures in a checklist, not a sequence.
As she stated, “it would be a teacher’s responsibility to see that students learn certain grammatical items by the end of a given course or period of time, but not by following a prescribed sequence” (Larsen-Freeman, 2001, p. 263).
In the context of Taiwan, no fixed selection or sequence of English grammatical structures has been provided for junior high school textbook publishers or textbook writers. Mostly, they followed the principle suggested in the Curriculum Guidelines, that is, to cover basic structures in textbooks and to teach grammatical items in order from easy to difficult. Nevertheless, this still caused divergences among different sets of textbooks in their syllabus design.