The Language of Conversation

A. Conversationl English: Some Studies

  1. Abercrombie, David - Conversation and spoken prose

  2. Crystal and Davy - Investigating English style

  3. Wilkinson, Andrew - Spoken English

It may be helpful to review some of the most important points from the above writers, before studying any texts.


"Conversation and spoken prose" in Studies in Phonetics and Linguistics, London. OUP. 1965.

Abercrombie distinguishes three broad categories of spoken English: reading aloud, monologue and conversation. Since we are not concerned with the first two we will follow what he says about the third.

"Under 'conversation' I would include all those linguistic occasions when there is the opportunity for give and take. When it is understood that, at least in theory, there is more than one active participant, however long one of the participants may go on."

Conversation is the most natural, most frequent and most widely occurring of all kinds of spoken English.

He points out that there are many types of conversation - "it may be between strangers, or between acquaintances, or between intimates; it may involve exchanging information, or merely exchanging goodwill or perhaps ill-will."

He also makes the important point that the way to investigate spoken conversation is not through looking at examples of written conversation as found in plays and novels. There are in fact two sorts of conversation: the spoken kind and the kind found in literature. The latter is really language written as if it were spoken conversation. The difference is important and needs to be emphasised because there are major differences linguistically between the two kinds. Abercrombie's term for the language of fictional conversation is 'spoken prose'. He has this to say about the relationship of the two kinds of conversation (p. 4):

"Most people believe that spoken prose ... is at least not far removed, when well done, from the conversation of real life ... But the truth is that nobody speaks at all like the characters in any novel, play or film. Life would be intolerable if they did; and novels, plays and films would be intolerable if the characters spoke as people do in life. Spoken prose is far more different from conversation than is usually realised."

The most important method we have of studying conversation is to record them (by means of tape recorders) and then to transcribe them into written English. Abercrombie says that such investigations reveal that "the whole structure of conversation is different from that of prose, spoken or written. It comes as a surprise to find how different it is." (p. 5).

He also points out that transcriptions of spoken English are deficient in certain respects. The most important omission in ordinary writing is the absence of any means of recording intonation. Again Professor Abercrombie is very good at indicating the significance of this omission:

"... if you are reading aloud a piece of written prose, you infer from the text what intonation you ought to use, even if, as is almost always the case, you have a choice. The intonation, in other words, adds little information. But if you try to read aloud a piece of written conversation, you can't tell what the intonations should be - or rather what they actually were. Here the intonations contribute more independently to the meaning. To write conversation adequately, we must be able to include the intonation in the text, whereas there is no need at all to do so in writing prose." (p. 6).

Among the differences in sound that he lists between spoken prose and conversation are:

  1. the intonation patterns of conversation are more varied than those of spoken prose;

  2. the tempo of conversation is less even than that of spoken prose;

  3. pauses in spoken prose occur at the boundaries of important units (sentences, clauses) but in conversation their role is different and their occurrence is less predictable;

  4. silence is important in conversation, though often it goes unnoticed. The gap may be filled by gestures and facial expressions; as Abercrombie observes, "the conversation continues during them, though not in words." (p. 8);

  5. 'Normal non-fluency' is characteristic of conversation. This includes stammers and wrong pronunciation. These are rare in spoken prose.

Other differences mentioned are the apparent incompleteness of conversation. Because of the context not everything need be put into words. Parts of sentences may be lacking and in fact "the sentence as traditionally defined is really a unit of prose, not of conversation." (p. 8). Repetition is important, so also are such expressions as sort of, kind of, you see, shall I say, you know, I mean, I mean to say, hat I call, well. These may be called intimacy signals (Professor Quirk) or silence fillers (Abercrombie, p. 9), or temporisers. They enable us to play for time while we think of what to say next. Other temporisers are mm and er - these are referred to as spoken pauses.

2 CRYSTAL, David and DAVY, Derek

Investigating English Style, London. Longmans. 1969.

We can next look at what Crystal and Davy have to say. Their Chapter IV describes the language of conversation using a textual, largely inductive approach. They indicate the value of this variety of language for their purpose - as a 'neutral' kind of English which provides a yardstick with which to measure other varieties of language. That is, we can describe other varieties of language by listing the features by which they differ from conversational English. They also point to its value in that there is a lot to be said about conversation at all levels of language - pronunciation, grammar, vocabulary, semantics.

There are three main characteristics of conversation that they point to:

  1. the apparent inexplicitness of the language (see Abercrombie) due to the fact that participants rely for much of their information on the extra-linguistic context in which the conversation is taking place. Here they include features such as anaphoric reference by words such as this etc; also the incomplete-ness of many utterances;

  2. randomness of subject matter, and a general lack of planning. "Conversation, as opposed to such concepts as discussion and debate, regularly lacks an overall theme." (p. 103). It is always possible to change from one topic to another in a conversation;

  3. its 'normal non-fluency' (see Abercrombie). They include here features such as hesitation markers, slips of the tongue, and overlapping or simultaneous speech. "The really significant fact about informal conversation is the toleration of these features when they occur, and indeed the expectation that they will occur ... What must be avoided at all costs is prejudging this issue by inculcating a perjorative attitude towards hesitation features in conversation - to refer to conversation as if it were 'disjointed', or to talk about these features as if they were 'errors', without further qualification ... is in part to judge conversation against some other (usually written) standard ... Considered in its own situation (that is, with gestures, facial expressions, and so on all included), conversation does not seem 'disjointed' at all." (pp. 104-5).

Crystal and Davy provide a lot of detailed phonological and grammatical information about conversation. Many of the most important points, (pauses, intonation, tempo) have been covered generally in the review of Abercrombie. This coverage is sufficiently detailed for secondary schools.

Among their observations about grammar are the large number of loosely co-ordinated clauses. Should these be considered as sequences of sentences or as single compound sentences? This kind of problem of course arises from the fact that the sentence as Abercrombie noted is a unit that applies more easily to writing than it does to speech. Crystal and Davy treat them as sequences of linked sentences. Sentences tend to be short and minor clauses (those without finite verbs) are frequent. Many sentences are incomplete due to anacoluthon (restarting a sentence in order to express more adequately what a speaker has in mind) eg, There are some of them have not come. They also make an interesting point about how a conversation is kept going by tag sentences. (Well, it feels healthier, doesn't it? But that's life, isn't it?) and also by introducing 'stand-by' topics such as references to the weather.

Groups (both nominal and verbal) tend to be simple. There are few adjectives in the nominal groups, but some intensifiers are frequent, very, a bit; wh-words are usually omitted in relative clauses, and prepositions normally come at the end of relative clauses.

Verbal contracted forms are frequent (I'll, he's), also 'filler' verbs such as got. (I got all the way home before finding out.)

Points of disputed usage tend to pass over without drawing comment. Expressions such as "everybody made their contribution" occur. Such expressions are apparently readily acceptable in spoken English and do not provoke comment in the situation in which they are used even from educated people who also use them. "Similarly, repetitious structures, looseness of syntax, 'weak' words like got, and nice ... are a standard and indeed valuable part of informal conversation." (p. 113).

Some vocabulary items are more or less restricted to conversation. These include yeah, cos, phone, a lot, a bit, all right, phrasal verbs (such as take to) as well as items already listed from Abercrombie (I mean, a lot, a bit, sort of). Other vocabulary features are colloquial idioms and lexical hyperbole.


Spoken English, Birmingham, University of Birmingham. 1965.

We will now look briefly at some of the things that are said about the language of conversation by Andrew Wilkinson. He is not directly concerned with this so much as with the broader question of how oracy should be developed and tested in the school. But there are nevertheless some interesting points about the language of conversation that are raised.

He points out that speech instructions involving several participants may be divided into two sorts: those in which there is continuous and immediate response of listeners and those in which this is absent. The first type he describes as a reciprocal speech situation; the second as a formal speech situation. This seems a useful and valid distinction but might be broadened by saying that reciprocal speech situations (or informal speech situations) are those in which there is more than one active participant in the speech situation; formal speech situations are those in which there is only one active speech participant.

Formal speeches are best exemplified in prepared speech situation, such as radio talks or public lectures. In such situations the speaker may still be capable of some slight modification of what he is saying to fit in with his observations of his audience - the feedback. But at the extreme case of the fully prepared lecture or radio talk, when there is little or no feedback, then the situation is fully formal.

Conversation on the other hand is a reciprocal speech situation. Any speaker has direct feedback in the form of looks, nods, grunts, facial expressions to help him communicate effectively. "Acting on these the speaker may continue, or may repeat or rephrase his point if he judges it to have been missed. The contribution of the others may bring a new direction to the conversation or introduce a new topic. In either case they will serve as a spur to further utterance on the part of the first speaker. This interchange is the essence of conversation; it serves to advance the communication between the speakers, and also by supplying a constant feedback, to help them judge whether they are still on the same wavelength. There are some people who do not need a relevant response to encourage them to proceed; and others like the characters in Pinter's dialogue who each pursues the train of his own thoughts in the gaps of silence provided by the others. For some people however, it is unnerving not to receive immediate responses, even non-verbal ones, although the speaker appears to be attending carefully; it might be taken to indicate hyper-criticism or disapproval of what is being said, and might cause the speaker to dry up or to attempt to elicit a response by such a phrase as 'don't you think so?' But in any case it will not provide a stimulus to his next utterance, and the conversation will languish."

"Ordinary conversation is often aimless and desultory - one does not necessarily converse with a neighbour met on the bus in order to arrive by debate at a solution of some problem, nor to convince him of the superiority of one's own method of cultivating roses to his - though one may if one wants to do so. One converses often enough on the weather or some other trivial subject in order to show goodwill, to make human contact, or to put it negatively - so as not to appear unfriendly, for only people who know one another well can be sure their silence will not be misinterpreted. If there is a group of people together instead of two only, the less likely it is that silence will prevail, though conversation may be more or less desultory." (pp. 73-74).

Back to Language of Conversation Contents