Lifelong Learning: Proceedings of the 4th Annual JALT Pan-SIG Conference.
May 14-15, 2005. Tokyo, Japan: Tokyo Keizai University.

Defining "Communicative"

by Jim Smiley (Tohoku Bunka Gakuen University)

(Japanese Abstract)


As a key term in modern language theory, the meaning of 'communicative' is assumed to be fairly stable. However, this article suggests that there is no generally accepted meaning of that term by critically comparing its use by various writers in the field. The conclusion is that each individual teacher needs to develop their own sense of what 'communicative' means to them based on a thorough knowledge of how it is used differently in the literature. To facilitate this more critical understanding, the author presents four methods of discovering meaning and concludes by presenting his own working definition of this concept.

This essay developed out of a growing sense of confusion, a feeling that those involved with language education were discussing different things when they used the term 'communicative', even though they were supposedly connected through a common belief about prioritising communication as a primary vehicle of language learning. The range of teaching acts allowed under this banner range from activities which are conducted entirely in the target language, to those which expect learners to provide content in the lesson, to a more restrictive interpretation in which output constraints are placed on some sections of the lesson. These output constraints might justify completely controlled activities or even first-language instruction based on the rationale that the intent of some activities is to provide the basis for future communicative possibilities. A simple conclusion to the problem of categorisation would be to defend one's particular teaching practices as being communicative using justifications as needed, according to one's own criteria. Another would be to critically analyse the boundaries of the term 'communicative'. The value in such an activity lies in the information uncovered and in the application of that information to a deeper understanding of one's teaching methodology.

[ p. 122 ]

Let us begin by looking at the following imaginary yet possible dialogues. At a cocktail party, one person engages another in introductory chit-chat, and the following statement is produced, "I'm very interested in Japanese culture." To which the interlocutor replies, "So you like noh, zen and hougaku?" "Actually, not so much. Everyday things the Japanese do interest me far more, like how they deal with each other and how they think." Later on the second person offers this: "I'm interested in classical music". His unfortunate conversation partner replies with, "Do you like Bach?" to receive the curt response, "He's baroque, not classical." These exchanges failed because the key term in each statement was ambiguous. 'Culture' may refer to high art or to the sum of a society's actions, and 'classical' music may be the rococo period or a genre distinct from pop, rock, folk or jazz. However problematic these terms are, context and feedback possibilities usually help the reader/listener disambiguate the author/speaker's intent.
"In the world of language teaching, however, the concept of 'communicative' is, at best, vague."

Academic writers try to clarify which meaning is being referred to in their professional discourse. In the world of language teaching, however, the concept of 'communicative' is, at best, vague.
The term 'communicative' serves a dual role as a signifier of a particular approach to language education and a more general term which covers a number of related ideas about language and language education, which in turn inform teaching practices based on those ideas. The indicative meanings of these words carry two functions; a more precise, limited definition and a broad ideology. However, in the case of 'communicative', unlike the examples from music and culture discussed above, the context often impedes a clear view of which indicative meaning is being referred to.
A typical description of the term ' communicative' comes from Stern (1992, p177) who asks, "What is meant by a 'communicative activity'?" His answer centres on involving learners in "authentic communication" and in "' real' communication". The discussion following this statement attempts to separate 'real' language from language which is typically used to teach and learn language. In other words, "The focus is not on learning specific language features but on putting the language to use as the circumstances require". Stern contrasts the concepts of message-oriented language and medium-oriented language to drive a further wedge between language code and communicative intent, supporting this premise by claiming that, "children learn their first language in the process of communication without focusing on the code". Stern continues to define communicative activities by contrasting them with classroom ones. "Even if she has had a year or two of instruction, she is likely to find that a natural target language environment is very different from the conventional language class".
Stern's strong view of communicative, that is having the requirement of authenticity, may be contrasted with that of Morrow (cited in Larsen-Freeman, 2000, p129) who has no such requirement. For Morrow, communicative activities have three salient features: (1) some type of information gap, (2) a degree of choice and (3) the existence of feedback. Stern suggests that all of these are possible within the classroom environment. A close look at his activities reveals many instances where information gap, choice and feedback are not present. For example, when a teacher asks a student to open a window in a warm room and a student obeys, Stern would claim that the sequence of events was genuinely communicative. I suspect that Morrow would do the same although not under the banner of the communicative method. Conversely, when Morrow presents students with contrived activities, Stern's authenticity requirement is missing. One way of avoiding this problem is to ignore the distinction or requirement for authenticity. Nunan (1999) devotes an entire chapter of 'Language Teaching Methodology' "Communicative Approaches to Listening Comprehension" and studiously avoided any direct reference to what Stern would call authentic language, at the same time as strongly implying that the language chosen was indeed created for purposes other than language learning. Do we take this to mean that Nunan sides with Stern and implicitly requires authenticity? In the same book's chapter on materials development, he outlines a procedure for teachers to work with "authentic sources of data".

[ p. 123 ]

An earlier attack on the lack of clarity in defining each individual position in regard to 'communicative' is neatly summed up by Richards (2002, p36), who noted that "The 1970s was a period when everyone was 'going communicative', although precisely what was meant by that varied considerably". For Richards, Communicative Language Teaching (CLT) stands opposed to the systems of teaching where grammatical competence is the main aim with "a focus on communication as the organising principle for teaching". Richards described the emergence of Communicative Language Teaching as a historical movement challenging syllabuses based on the division of language into grammatical units. An additional irony concerning the term 'communicative' was the subsequent analysis of communicative functions and the presentation of those analyses into each equally strict and structured syllabuses. All that was accomplished by this, "was to take traditional items of grammar and restate them in terms of concepts or notions" (p37). In keeping with this selection and structured presentation of communicative concepts within a syllabus, Brown (1995, p7) categorises six ways of organising concepts on top of the traditional grammar or structural syllabus. Brown foregoes any discussion on authenticity per se but simply registers that in the communicative approach "students must be able to express their intentions. They must learn from the meanings that are important to them".
"Made-up language may be more beneficial for learners as it is easier to procure . . . and can reflect far more precisely the particular language input needs of any learner group at any particular time."

From the previous discussion we can isolate a few key strands: authentic versus made-up language for class use; systems concentrating on the code or on the message, inside or outside the classroom; and the structuring of language code for pedagogic purposes. A simple analysis of each of these strands quickly reveals semantic problems which serve to undermine any definitive treatment of each term. Perhaps one of the reasons Nunan and Brown avoid any discussion of authentic is because of the earlier controversy this word started. Widdowson summed up the discussion by challenging the relevancy of the term, "I am not sure that it is meaningful to talk about authentic language as such at all. I think it is probably better to consider authenticity not as a quality of residing in instances of language, but as a quality which is bestowed upon them, treated by the response of the receiver" (Cited in Lewis, p28). Lewis (1993, p28) further comments that, "artificial activities such as games may produce a more authentic experience in this sense than genuine language data". Made-up language may be more beneficial for learners as it is easier to procure (it can be made up on the spot) and can reflect far more precisely the particular language input needs of any learner group at any particular time. There is no theoretical basis for the supremacy of authentic language, yet unauthentic language, which is meaningful, is more likely to be more useful in the classroom. However, this previous sentence itself is open to criticism in that, for some, the criterion here of 'meaningful' validates the language as authentic.
Stern's (1992, p178) contention that native speaking children do not grapple with code is dubious. Many first language researchers have noted that children babble, which often serves no communicative purpose and is seen as children's play with language. Stern's unstated argument questions the degree to which language is recognised to be language by the child rather than some unnamed device for getting things done. Certainly children do not have metalanguage, which would allow them to cognitively and deliberately focus on the language units, yet at the theoretical level it is possible to question whether or not adults actually have appropriate metalanguage facilities. If, according to Prahu (1987, p21), "language is too complex to be handled at a conscious level", the language of traditional grammar classification seems to be insufficient to function as a competent metalinguistic system. It may well be that the level of children's understanding of their babbling or of their early inter-language system might be very similar to that of adults learning a second language.

[ p. 124 ]

It would be very difficult to claim that the simple distinction between inside and outside the classroom is enough to qualify or disqualify an activity as being communicative. I would argue that for a student who spends around eight hours in a classroom each day and whose very reason for life is to collect grades and credits, classroom language is very real indeed. Furthermore, if students can learn mathematics, geography and so on acceptably reasonably well in school, the reason for the failure to learn language might well lie in a different place. The reasons may well be to do with the type and the amount of language exposure in the classroom, not simply the classroom itself. Of course, this is an oversimplification of Stern's argument. He seems to use the term 'classroom' as a metaphor for structure-graded language classes and other courses where the content has been selected and graded according to traditional principles, even if the language units selected are not themselves grammar items. In other words, Stern disallows product teaching from his classification of 'communicative'. Against this view, Ur (1997) outlines the possibility of using progressively freer answer types in a scale running from fixed to free within a general framework of grammar instruction. The product emphasis placed here is still traditional, but it allows for some degree of choice. Ellis (2001) tries to structure this choice in communicative language practice design by creating activities that are language necessary (they require certain language to be used to complete the activity), language essential (they allow for some variation, but still limited, answer or response types), or language possible (they allow for the intended target language to be just one of the possible ways an activity can be completed). The implication is that, for Ur and Ellis, even strictly controlled language practice activities may be considered possible within the limits of methodologies informed by communicative teaching principles.

Working Towards a Definition

This essay could stop at this point by simply saying that everyone in language education should define what they mean by 'communicative' and fall back on the practical position of Prabhu (1990), who asserted that all teachers can do is to develop their own personal sense of 'plausibility' (p161). Yet, if we are to claim to be communicative teachers using communicative methods, but at the same time our own understanding of what that actually means differs from everyone else's, we run the risk of being charged with inefficiency: we might mislead or confuse our students; we might misrepresent our professional intentions; and we might generally be working towards what are discovered to be diverse goals when the purpose was a unitary one. Furthermore, we can hardly wear the badge of professionalism if the very flag under which we charge is effectively meaningless. One might counter-argue by questioning if there really is a problem requiring attention, suggesting that it is acceptable to use any meaning of the term as long as the actual pedagogic activity supports the general aim of communicative competence. However, this is a circular argument as only those activities that foster communicative competence may be considered communicative, and all others will fall out with these definitional boundaries. Practically, if we are more aware of our purpose, the chances are that we will be more exact in our actions. I would like to offer a few suggestions about how teachers might define their own boundaries more critically. Following which, I will offer my own working definition.
Four sources were used: a standard dictionary, a corpus database, a specialised linguistics dictionary and usages taken from within language teaching theory. In a dictionary, we should expect to find the core meanings listed. The exclusionary/ inclusionary nature of standard dictionaries provides valuable boundaries for standard use expressions. An examination of usage will throw light on two critical areas: the accuracy of the dictionary and the frequency with which each core definition is used. Caution must be exercised, as the strength of a definition may not correlate with its frequency. Applied Linguistics can show the specialised nature of how terms are used. Differences between common and specialised usage may be observed. The value of the presented definition may be judged according to how much, if at all, it clarifies any difference. Finally, Morrow's definition from the communicative methodology will be repeated as it forms the basis for my own.

[ p. 125 ]

A dictionary definition

The New Oxford Dictionary of English (1998, p 371) defines 'communicative' as, "willing, eager, or able to talk or impart information." From the same source, the verb 'communicate' is defined as "to share or exchange information, news, or ideas." or to "impart or pass on information, news, or ideas: convey or transmit in a non-verbal way: succeed in conveying one's ideas or in evoking understanding in others." This word is derived from the Middle English term communicativus and Latin communicat.
There are two critical differences between the adjectival and verbal forms of this term: one concerning number and one regarding the criterion of success. The adjective 'communicative' refers to a single person, giving us information about the individual. The verb, 'communicate' includes a criterion for success, and as one-way communication is not possible, 'communicate' requires at least two players. The singular nature of 'communicative' is reinforced as we recognise that it is possible to speak of a 'communicative' person simply on the strength of their reputation as a communicator even if they are being inactive at the time the label is being applied. Moreover, we may usefully ask questions about the nature of sharing and commonness, too. This task underpins the work done in Communication Theory.


The following sentences, taken from The British National Corpus Sampler, show how the word 'communicative' is used commonly. The method used to select these four examples was to include everything that seemed to me to be a non-specialised use of the term, except for the last excerpt listed here. The search term 'communicative' produced 470 results. From these, four typical examples are:
"It frequently happens that feminist women have relationships with men who are sensitive, communicative, share emotional and domestic chores and are committed to feminist ideals."

"That he wasn't very communicative, kept himself to himself."

"Egocentrism manifests itself in communicative terms as an inability to take account of another's point of view; instead, the world is seen rigidly from the standpoint of the self."

"Our pupils will, therefore, learn basic grammar as well as developing their communicative skills."

These examples do indeed clarify the dictionary position at least as far as the common use goes. But the last potentially specialised example highlights the ironic difficulty in clarifying which meaning is being referred to. Does the school wish to train pupils to become willing, or eager to communicate? However, as an attributive adjective for 'skills', we are left on our own as to which skills are in question here, as well as wondering about the theoretical basis for grammar being dislocated from communication.

A linguistic definition

The Longman Dictionary of Language Teaching and Applied Linguistics (2002, p90) defines the communicative approach, which it also terms 'communicative language teaching' as: "an approach to foreign or second language teaching which emphasises that the goal of language learning is communicative competence and which seeks to make meaningful communication and language use a focus of all classroom activities". This entry does not define 'communicate' or 'communicative', although it does point to another entry, 'communicative competence', which helps in outlining some of the features a full definition of 'communicate' might include. The term 'competence' is complicated and points to a value system and a hierarchy. Key questions here are 'Who controls the giving of the term 'competence?', which is a political point and 'At which point can competence be said to have been reached?', which relies on an unstated system of evaluation. 'Competence' indicates a reached state, but 'communication' is an on-going activity. This distinction explains why Longman's competence is described as a state of knowledge and the communicative approach as one of actions and skills to be practiced. Also, 'communicative' is defined as what it is not: it is not any other method.

[ p. 126 ]

In Longman's definition, five principles are given. Let us look at these in detail. #1 statement that "learners use a language through using it to communicate" seems fine prima facie as the meaning of the verb 'to communicate' may be inferred from their entry on the noun, 'communication'. Rephrased, through the dictionary definition, #1 gives something like this: the only language which learners use is that which results in successful information transfer, as 'to communicate' implies success. Of course, this is not always possible so, it would be better to qualify 'to communicate' by replacing it with 'to try to communicate'. #2 uses terms whose meanings are disputed, "authentic and meaningful communication should be the goal of classroom activities": What constitutes 'authentic'? Is this really a feasible goal? Who decides on the boundaries of 'meaningful', especially if getting higher grades through the memorisation of grammar rules is meaningful for many? #3 and #5 make claims which would fit all other language teaching methods, ("fluency and accuracy are both important goals" in #3 and "learning . . . involves trial and error" in #5, and #4 makes a claim which is shared by every other language learning activity, "communication involves the integration of different language skills": We can't, for example, translate without using various language skills; such as memory recall, reading ability, psychomotor movements. Anyway, what is a 'language skill' precisely?
Finally, near the bottom of the entry we find, "other communicative approaches"! The entry title is 'communicative approaches' yet it specifies just one type, CLT, and leaves us needing much more information in order to get a picture of the whole meaning.

Use within the Communicative Language Teaching literature

Larsen-Freeman (2000, p129) writes of communicative language teaching telling us that, "Activities that are truly communicative according to Morrow have three features in common: information gap, choice and feedback". This definition is usefully restrictive in a number of key ways: The 'information gap' refers not to what the teacher knows but students don't, but to a set-up in the activities where information is divided among students who then have to piece it back together. 'Choice' means that learners should have the freedom to use whichever language they want to communicate. And 'feedback' is where the success or not of students' language can be judged.
As we saw earlier, this definition is in no way shared by all writers who use the term. Morrow, through Larsen Freeman, and Stern differ in key ways over their treatment of the communicative language teaching's methods. Yet they are precise in the own definitional boundaries. Both Larsen-Freeman's and Stern's texts are explanatory ones dealing with precise meaning and content of language pedagogy. We should expect them to be precise. That Larsen Freeman's text comes eight years after Stern's and describes a very different system from Stern's without recognising definitional differences within the same methodology is at once troubling and will be a further catalyst in the deepening confusion over this critical term. In most texts which are not specifically intended to explain differences in methods, we do not find operational boundaries being defined at all.

A working definition
"It is unrealistic to expect language activities to be purely 'real-world' discussions, especially at the low levels of language ability."

A further issue, not addressed by Stern, is the grey space between 'truly' communicative activities and those that prepare for communication but are not necessarily communicative in their own right. It is unrealistic to expect language activities to be purely 'real-world' discussions, especially at the low levels of language ability. Furthermore, although outlawed by CLT, I can see no logical reason to banish grammar, vocabulary, or any other more traditional area from what constitutes 'information'. However simply saying that any activity which prepares the ground for future communication is communicative would be tantamount to saying that anything that uses language is communicative. Arguably, this is so, but it is not productive.

[ p. 127 ]

So, I offer this definition of what constitutes a communicative activity. This definition is, of course, personal, but it has been reached after some consideration of existing definitions. I would urge you not to adopt it, but to find your own.
A communicative activity is one where some or all of the available information necessary to complete the activity is accessible to a student only by that student asking another for the information.

This definition uses the concept of the information gap, as does Morrow's CLT, but it includes any teaching point I feel necessary. It seems to place the whole sphere of language learning firmly within the speaking/ listening matrix, but I do not use this kind of activity exclusively. Various kinds of reading and purely grammar explanations followed by questions also feature in my teaching. More fundamentally, the definition places a responsibility on one student for the success of another. That is to say, if the answer given is faulty, the activity will fail. This failure is also a part of the definition. Anything which precedes this kind of communicative activity having the intention of leading to it is defined as a preparation for communication activity.


I would like to end by noting a new phenomenon. During the presentation at the 2005 Pan-SIG Conference, a number of attendees expressed the opinion that, to them, if an activity does not involve any degree of personalisation, the activity is not communicative. Subsequently, I asked a few colleagues via email and face-to-face how they would define communicative, and almost all concurred that without personalisation they would not consider an activity communicative. The literature is divided over whether or not 'communicative' carries the imperative of personalisation. In my view, however valuable the role of personalisation is in language education, it is not a necessary element of communication.


British National Corpus Sampler. (n.d.). Retrieved on April 16th 2005 from

Brown, J. D. (1995). The Elements of Language Curriculum. Boston: Heinle & Heinle.

Ellis, R. (2001 October). Untitled. Paper presented at the British Council. Tokyo.

Larsen-Freeman, D. (2000). Techniques and Principle in Language Teaching. Oxford: OUP.

Lewis, M. (1993). The Lexical Approach. Hove: Language Teaching Publications.

Nunan, D. (1988b). Language Teaching Methodology. Harlow: Longman.

Persall, J. (1998). The New Oxford Dictionary of English. Oxford: OUP.

Prabhu, N. (1987). Second Language Pedagogy. Oxford: OUP.

Prahbu, N.S. (1990, Summer). There is no best method - why? TESOL Quarterly, 24 (2). 161-176.

Richards, J. C. (2001). Curriculum Development in Language Teaching. Cambridge: CUP.

Richards, J. C. & R. Schmidt, (3rd edition, 2003). Longman Dictionary of Language Teaching and Applied Linguistics. Harlow: Longman.

Stern, H.H. (1992). Issues and Options in Language Teaching. Oxford: OUP.

Ur, P. (1996). A Course in Language Teaching. Cambridge: CUP.

2005 Pan SIG-Proceedings: Topic Index Author Index Page Index Title Index Main Index
Complete Pan SIG-Proceedings: Topic Index Author Index Page Index Title Index Main Index

Last [ p. 128 ] Next