PRODUCT AND PROCESS APPROACHES TO THE TEACHING OF GRAMMAR

Among the various ways of exploiting grammar in the classroom, product and process
approaches have been enormously influential in language teaching.
Product Approach
Whether the focus is form, function or skills, product approaches segment the
target language into separate linguistic items for presentation one at a time. The assumption
behind such approaches seems to be that language is analyzable into a finite set of rules which
can be combined in various ways to make meaning. Moreover, these approaches rely "on
learners' assumed ability to learn a language in parts . . . which are independent of one another,
and also to integrate, or synthesize, the pieces when the time comes to use them for
communicative purposes" (Crookes and Long 1992: 28). Rutherford (1987: 4) calls this the
"accumulated entities" view of language learning. Product approaches focus on what is to be
learned (i.e. L2), on the knowledge and skills which learners should gain as a result of
instruction.
However, product approaches have been the object of the following criticisms:
a. Samples of the target language lack authenticity. It is also difficult to isolate and present
one separate item at a time, especially if teachers want to provide some sort of context for
the language.
b. SLA research offers little evidence to support the model of language acquisition that
product approaches assume. Learning does not occur in simple additive fashion. Nor
could immediate target like mastery of the linguistic form be attained. However, progress
in one area depends on progress in the others. Studies of language acquisition
demonstrate beyond reasonable doubt that learning is an active process of gradually
working things out. The L2 learner progresses along the interlanguage L1 continuum,
which consists of formulation and reformulation of hypotheses about how language
works. Learners do not jump from one stage to the next. Rather, learning grammatical
forms involves a gradual revision and re-structuring of the interim hypotheses to
accommodate new systems about the target language. Furthermore, "progress is often not
even unidirectional. SLA frequently involves temporary 'deterioration' in learner
performance (so called backsliding), giving rise to U-shaped and zigzag developmental
curves" (Crookes and Long 1992: 31).
c. The previous analysis of the target language seems to have priority over considerations
about the psychological processes involved in learning. As Crookes and Long argue,
"language learning is a psycholinguistic process, not a linguistic one"; however, product
approaches "consistently leave the learner out of the equation" (1992: 34).
But product approaches to grammar teaching also have their strengths. They can
provide the learner with a strong sense of direction. This fact can contribute positively to the
learner's feeling of security and purpose. Such an approach, it is argued, facilitates the processes
of noticing and structuring grammar. Specific grammatical structures are made as salient and
noticeable as possible. This perspective can also provide the learner with repeated opportunities
to structure and re-structure target forms.
 In traditional models of language teaching, considerations of syllabus design
(what to teach) have tended to be kept separate from methodology (how to teach). White's (1988)
Type A or Crookes and Long's (1992) Synthetic syllabuses will see the syllabus as primary,
since the syllabus designer or teacher interferes in the language learning process through the
selection, ordering and presentation of the discrete linguistic items. Methodology is put to the
service of syllabus, "directed at facilitating this internalization process" (Widdowson 1990: 119).
Nevertheless, one can argue that the syllabus itself is a framework, a resource, a guide.
Consequently, it can only give you ideas; the syllabus cannot prescribe the methodology.
Teachers can always find some room for manoeuvre, even in cases when syllabus and
methodology seem to be incompatible. As Widdowson puts it, the syllabus itself is a slow abstract
object . . . , a set of bearings for teacher action and not a set of instructions for learner activity. What
learners do is not directly determined by the syllabus but is a consequence of how the syllabus is
methodologically mediated by the teacher. (1990: 129)

Product syllabuses provide teachers and learners with a clear outline of what should be
taught. But they need not determine the methodology. Consider, for example, a case where the
syllabus is structural; however, a teacher can exploit it in an interactive fashion and the syllabus
may be as "communicative" as a notional/functional or a task-based one.
Process Approach
In process approaches, there is a shift in emphasis from the outcomes of instruction to the
learning experiences themselves. Therefore, the concern is with the learning process and
pedagogical procedure rather than with the content. Accordingly, in such an approach, "there is
little or no attempt to interfere in the language learning process through the selection, ordering
and presentation of content by the syllabus designer or teacher" (White 1988: 47). Instead of
thinking in terms of grammar for the learner (as in product-oriented approaches), we can now
think of grammar by the learner. Grammar, then, can be seen as a device which language users
call upon when motivated by a communicative need to make their meanings clear. In this sense,
as Widdowson concludes, "grammar is not a constraining burden but a liberating force: it frees
us from a dependency on context and the limitations of a purely lexical categorization of reality"
(1990: 86).
The focus will be on communication and the aim is to have learners communicate
effectively. Learners will be given the opportunity for productive and creative language and the
cooperation of meaning in the classroom. In normal language use, competent speakers organize a
highly complex number of skills and subskills simultaneously. The learners' language has to be
appropriate, relevant to the occasion, sensitive to the other discourse participants. when under
pressure. Learners do not generally attain the same kind of competence as native language
speakers, but they too manage to operate the systems of communication as a whole and develop
the skills and strategies of the discourse process. Teachers will have to shift their learners to a
situation where they use the various skills they have already learnt automatically. They have to
learn to be able to engage, assemble and have access to their knowledge of the system, so that
they can free their resources and perform automatically when focusing on meaning.
Process approaches, therefore, provide opportunities for learners to exploit strategies for
negotiation and interaction in language use. It is argued that grammatical forms are unlikely to
become internalized unless proceduralization can take place. Noticing and structuring grammar
are not enough. This is where the third dimension of language learning comes in. Following Ellis
(1985), a distinction can be drawn between two types of L2 knowledge: declarative and
procedural. The former refers to what the learner knows about the language, i.e. adopted L2 rules
and memorized chunks of speech. The latter is "how to" knowledge; that is, it comprises "the
strategies and procedures employed by the learner to process L2 data for acquisition and for use"
(Ellis 1985: 164). Proceduralizing refers to the mental organization of knowledge, so that
accessing is made easy. Competent language users' knowledge of the language seems to be
stored in the mind ready for use, already assembled for immediate access. This prefabricated
speech has both the advantage of more efficient retrieval from memory and of permitting
speakers to devote their attention to the larger structure of the discourse (see Bolinger 1976,
Pawley and Syder 1983). In short, the mechanism of proceduralization "refers to the embedding
of factual knowledge into productions so that the products of frequently executed productions
can be retrieved directly from memory and declarative knowledge need not be activated in
working memory for their execution" (Schmidt 1992: 363).
However, there is a second dimension which is sometimes referred to as "procedural
skill" (Batstone 1994; Schmidt, 1992). If proceduralized knowledge is concerned with the
formation and storage of knowledge, procedural skill relates to the accessing and efficient
performance in language use (e.g. being economical and avoiding undue repetition or excessive
pausing, controlling pace, engaging strategies to be fluent). So learners need plenty of practice to
proceduralize grammar in real-world language use. Without regular opportunities to put their
grammar into action more or less automatically when negotiating meanings, much of the
grammar learners may have noticed and structured through product work will gradually
disappear. Furthermore, learners are likely to revert to a more lexical language system, which
predisposes to fossilization and lack of intellectual development.
There is evidence, therefore, that process teaching encourages both the proceduralization
of knowledge and procedural skill. However, as Batstone (1994) points out, process work may
be more effective in developing procedural skill than in promoting proceduralized knowledge. In
a similar vein, Skehan suggests that "there is the possibility that communicating meaning
predominates as an aim, and that learners achieve this aim by using strategies which by-pass the
underlying language system, so that they may not be driven to develop their interlanguage
systems" (1992: 185).
We see, then, that there are dangers with both product and process approaches. As Skehan clearly
puts it, excessive priority given to analysis will compromise the process of synthesis and the acquisition
of a memory-based fluency in performance. Too much emphasis on synthesis may well detract from the
learner's ability to be accurate and restructure. (1992:193)

It would appear that not too strict an adherence to either product or process perspectives will
prove satisfactory. We need to reconcile these opposites and admit there is no single effective
methodology for the learning of grammar. Consequently, in the day-to-day world of teaching, an eclectic
position will probably emerge. My position, as I hope will become clear, would be in favour of a mixed
methodology, with a variable balance between product and process. Both approaches may be combined in
different ways according to varying circumstances. In view of this, product and process should best be
thought of as points on a pedagogic continuum, rather than extremes. Once again the issue is how to
achieve a balance between a controlled approach to language development and the learners' direct
involvement in the discourse process