Every linguist knows that language is changeable and elusive. Not only do languages
change over time, but they also vary on the horizontal planewe all speak differently
from one another and to one another. This makes our job as language teachers
extraordinarily difficult. We teach our students the "standard" way of saying
something, only to be confronted later by perplexed students wondering why native speakers
use a different or even aberrant version.
As it happens, we have in linguistics the means to explain this common phenomenon. When
a linguistic unit is realised by more than one norm, it is termed a linguistic variable.
If its use is sensitive to social context, it is called a sociolinguistic variable.
To further explain a sociolinguistic variable, we will center on one such variable,
namely, the form of the English future, "be going to + infinitive." This verb
tense has two norms: one, which we shall call the going form, and the second labeled the
gonna form. In norm one, we state "The government is going to suspend
negotiations...." But in norm two, we might say, "Yeah, Daves gonna fix
that. He said he was gonna do that early next week."
Together, the two norms constitute a linguistic variable. Although basically a
syntactic phenomenon, the occurrence of the two norms is sensitive to lexical choice. For
instance, norm two is rarely used by educated people in formal settings. We would probably
not say "The Australian National Youth Orchestra is gonna perform Bachs
Orchestral Suite No. 3 tonight."
From the viewpoint of an ESL teacher, the construction "be going to +
infinitive" is significant since it is taught fairly early. As one form of the future
in English, it carries a substantial functional load and is commonly used by native
speakers. A question arises then as to why so few nonnative speakers seem to have mastered
this form. It appears that this problem is related to the difficulty of learning English
rhythm and, thus, is a phonological problem. However, there is another possible
explanation, one that concerns the way teachers of English address their students.
"Teacher-talk" is defined as speech used by teachers that is
characteristically modified in four areas: phonology, lexis (consisting of morphology and
vocabulary), syntax, and discourse. The phenomenon of teacher-talk has captured the
attention of researchers for some time. Numerous studies (Chaudron 1988 cites 21) have
been devoted to its description and analysis.
According to some of these studies, the differences in teacher-talk as compared to a
typical sociolinguistic domain are not systematic or widespread enough, and they tend to
disappear as the class advances. Chaudron (1988) gives some of the typical modifications:
slower rate of speech
less reduction of vowels and consonant clusters
more standard "literary" pronunciation
more basic vocabulary
fewer colloquial expressions
fewer indefinite pronouns
fewer subordinate clauses
fewer words per clause
shorter length of utterance
higher proportion of simple present tense
higher proportion of well-formed sentences
delivery rate one-half to one-third slower
more first person reference
fewer functions per time unit
more teacher-initiated moves
more conversational frames
more verbalization per function
To summarize briefly, the speech that teachers use with their students is shorter,
simpler, and more carefully pronounced than typical speech.
Rate of speech
Somewhat surprisingly given the number of studies devoted to the description and
analysis of teacher-talk, only two characteristics of this phenomenon have been
investigated as to their efficacy: rate of speech and syntactic complexity. We will
restrict ourselves to the former since it is most appropriate to our subject.
Dahl (1981) investigated the relationship between the rate of speech and
comprehensibility. The subjects of the study (college students and nonnative speakers of
English) were exposed to several messages and asked to rate how understandable these were.
All of the subjects judged that the more comprehensible messages were those delivered more
slowly. Interestingly enough, however, these judgments did not correlate with the actual
measured rate of delivery. Dahl concluded that other factors such as the conciseness of
information and the clarity of articulation, which are both linked to the perceived rate
of speech, may have played a part in the students judgments.
Kelchs study (1985) also addressed the question of whether slowing speech
enhanced comprehensibility. University students who were nonnative speakers of English
were given dictations at varying speeds. It was found that the students performed
substantially better when the rate of delivery was slowed from about 200 words per minute
(normal speech) to 130 words per minute, which is the average rate of teacher-talk
directed toward beginning students.
Blau (1990) carried out two studies. One measured the effect of speed and syntactic
complexity on learner comprehension and the other measured the effect of pauses. Contrary
to Kelch, she found that slowing the rate of speech and simplifying syntax did not help
learner comprehension significantly; however, pausing at constituent boundaries did help.
At this point it is difficult to state with any scientific confidence that speaking
more slowly and in simpler sentences actually works. But, scientific claims aside, it is
intuitively clear to language teachers that teacher-talk does work, perhaps as a function
of all of its characteristics. In fact, it feels necessary, especially with
Returning to the sociolinguistic variable we were investigating earlier, "be going
to + infinitive," we begin to see why so few non-native speakers of English produce
the gonna norm of this form. Not only does its use imply close knowledge of phenomena of
English pronunciation such as reduction and a mastery of English rhythm, but also that
learners are unlikely to hear the reduced form gonna in the English language classroom.
When teachers slow their rate of speech and enunciate, gonna is impossible to say.
From the sociolinguistic point of view, teacher-talk, because it is slower and clearer,
unwittingly imitates formal English speech. The consequence is that without working on the
pronunciation of English (notably the rhythm) and exposure to the more usual pronunciation
of variables, learners of English learn and retain a relatively formal and sometimes
hypercorrect form of English.
A practical solution
We are thus presented with a classic problem. Teacher-talk, used judiciously, seems to
be effective, yet it can have lasting undesirable effects on a learners speech. But
there is an easy solution that allows students to hear both norms. First, when presenting
the formal form, that is, "be going to + infinitive," begin with careful
pronunciation to ensure that students understand the usage and can produce it with some
accuracy. Next, expose them aurally to the informal form, that is, gonna, and explain its
uses and constraints. Finally, have the students practice the reduced form, beginning with
simple question and answer drills and ending with role-play situations. The latter are
invaluable in pointing out such sociolinguistic facts as register variation and politeness
By exposing students to authentic speech and giving them the opportunity to use it, we
accomplish much. Richards (1983) considers both to be essential in developing listening
comprehension. If students master the various norms, their speech sounds more English.
Perhaps even more significantly, if they manipulate the norms properly, they are, in fact,
"acting" more English.
In a linguistic sense, we are also developing their sociolinguistic competence and in a
personal, affective sense, we are giving students the opportunity to seek status and/or
solidarity much as they do in their own languages (Gee 1988; Labov 1980; Milroy and Milroy
1985). To ignore language variation is to deprive and impoverish our students and
ultimately to do a disservice to the language that is in the words of Anthony Burgess,
always "gloriously impure."
Blau, E. 1990. The effect of syntax, speed and pauses on listening comprehension. TESOL
Quarterly, 24, 4, pp. 746-752.
Chandron, C. 1988. Second language classrooms: Research in teaching and learning. Cambridge:
Cambridge University Press.
Dahl, D. 1981. The role of experience in speech modifications for second language
learners. Minnesota Papers in Linguistics and Philosophy of Language, 7, 2, pp.
Gee, J. 1988. Dracula, the Vampire Lestat, and TESOL. TESOL Quarterly 22, 2, pp.
Kelch, K. 1985. Modified input as an aid to comprehension. Studies in Second
Language Acquisition, 7, 1, pp. 81-90
Labov, W. 1980. Locating language in time and space. New York: Academic Press.
Milroy, I. And L. Milroy. 1985. Authority in language: Investigation language
prescription and standardisation. London: Routledge and Kegan Paul.
Richards, J. 1983. Listening comprehension: Approach, design, procedure. TESOL
Quarterly, 17, 2, pp. 219- 240.